University of Southern California Dissertations and Theses

Design and application of a C-shaped miniaturized coil for transcranial magnetic stimulation in rodents

(USC Thesis Other)

Design and application of a C-shaped miniaturized coil for transcranial magnetic stimulation in rodents

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content DESIGN AND APPLICATION OF A C-SHAPED MINIATURIZED COIL FOR
TRANSCRANIAL MAGNETIC STIMULATION IN RODENTS
by
Wenxuan Jiang
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(BIOMEDICAL ENGINNERING)
December 2023
Copyright 2023 Wenxuan Jiang

ii
Acknowledgements
First and foremost, I would like to thank my advisor, Dr. Dong Song, for his support and guidance
throughout my graduate studies. Dong’s knowledge, patience, integrity, and encouragement have been
invaluable in moving this project forward. I am grateful and proud to be one of Dong’s students.
My co-advisor, Dr. Charles Liu, deserves equal acknowledgment for his insights, leadership, enthusiasm,
and wisdom. I'm especially grateful for the trust and confidence he placed in me, shaping the trajectory of
my academic journey.
I would like to thank the members of my thesis committee: Dr. Ellis Meng, Dr. Hossein Hashemi, and
Dr. Qifa Zhou, for their time, expertise, advice, and support. A special acknowledgment is for Dr. Ellis
Meng, whose consistent support and collaboration have significantly enriched my research journey.
Additionally, I would like to thank Dr. Peter Wang. His encouragement remains a beacon, always reminding
me of my potential and the importance of persistence in research.
I would like to thank the entire Song lab. I consider myself fortunate to have collaborated with such a
distinguished group. A particular note of appreciation goes to Dr. Huijing Xu and Dr. Sahar Elyahoodayan
for their mentorship and skill sharing.
Many have contributed significantly to my research journey. In particular, Robert Isenhart, Dr. Darrin
Lee, Dr. Vasileios Christopoulos, Dr. Xuechun Wang, and Dr. Kee Scholten have been steadfast
collaborators throughout my doctoral projects. Additionally, during my undergraduate years at the National
University of Singapore, Dr. Angelo All and Dr. Daniel Teh provided foundational guidance in research.
Lastly, I would like to thank my family for their support. I especially want to express my deepest
gratitude to my parents and grandparents for their love and unwavering support. I would also like to thank
my husband, Dr. Xinyan Liu, for his unconditional love. Additionally, my gratitude extends to my cat, who
has devoted her lifetime to loving me, and who occasionally bites me — out of goodwill, I believe.

iii
Table of Contents
Acknowledgements ....................................................................................................................... ii
List of Tables................................................................................................................................ vi
List of Figures.............................................................................................................................. vii
Abbreviations ............................................................................................................................... ix
Abstract.......................................................................................................................................... x
Chapter 1: Introduction and Background.................................................................................. 1
1.1 Motivation............................................................................................................................. 1
1.2 TMS coils.............................................................................................................................. 3
1.2.1 Types of TMS coils........................................................................................................ 3
1.2.2 Challenges of making small TMS coils......................................................................... 5
1.3 TMS parameters.................................................................................................................... 8
1.3.1 Pattern ............................................................................................................................ 8
1.3.2 Intensity........................................................................................................................ 10
1.3.3 Orientation ................................................................................................................... 11
1.4 Mechanisms of TMS........................................................................................................... 13
1.4.1 Physics behind TMS .................................................................................................... 13
1.4.2 Synaptic plasticity........................................................................................................ 14
1.4.3 Neurotransmitters......................................................................................................... 16
1.4.4 Other mechanisms........................................................................................................ 17
1.5 TMS and neuromodulation ................................................................................................. 18
1.5.1 Electrophysiological effects of TMS ........................................................................... 18
1.5.2 Treatments of neurological and neuropsychiatric disorders........................................ 20
Chapter 2: Design and Evaluation of a Miniaturized Coil ..................................................... 23
2.1 Introduction......................................................................................................................... 23
2.2 Method ................................................................................................................................ 27
2.2.1 TMS coil design........................................................................................................... 27

iv
2.2.2 TMS circuit design....................................................................................................... 27
2.2.3 Measurements of magnetic and electric fields............................................................. 30
2.2.4 Finite element modeling of magnetic and electric fields............................................. 32
2.2.5 Animal Surgery............................................................................................................ 33
2.2.6 Single-unit recordings.................................................................................................. 33
2.2.7 Somatosensory evoked potentials................................................................................ 35
2.2.8 Motor evoked potentials .............................................................................................. 36
2.2.9 Repetitive transcranial magnetic stimulation............................................................... 36
2.2.10 Temperature measurement......................................................................................... 36
2.2.11 Histology.................................................................................................................... 37
2.2.12 Data processing and statistical analyses .................................................................... 37
2.3 Results................................................................................................................................. 39
2.3.1 Characterization of TMS coil....................................................................................... 39
2.3.2 Comparison of simulated and measured field distributions......................................... 41
2.3.3 Magnetic and electric field distribution in the rat brain............................................... 45
2.3.4 rTMS increases firing rates of primary somatosensory and motor cortical neurons ... 46
2.3.5 rTMS suppresses somatosensory evoked potentials.................................................... 49
2.3.6 rTMS facilitates motor evoked potentials.................................................................... 51
2.4 Discussion........................................................................................................................... 53
Chapter 3: Subthreshold rTMS Suppresses Ketamine-Induced Poly Population
Spikes in Rat Sensorimotor Cortex ........................................................................................... 60
3.1 Introduction......................................................................................................................... 60
3.2 Method ................................................................................................................................ 62
3.2.1 Pharmacological manipulations and electrophysiological recording .......................... 62
3.2.2 Statistical analysis........................................................................................................ 65
3.3 Results................................................................................................................................. 66
3.3.1 Ketamine induces poly population spikes in addition to slow-wave activities ........... 66
3.3.2 rTMS suppresses ketamine-induced poly population spikes and changes the
power of LFP bands.............................................................................................................. 68
3.3.3 rTMS increases spontaneous neuronal spike firing rates............................................. 75
3.4 Discussion........................................................................................................................... 80

v
Chapter 4: Subthreshold rTMS Alters Learning and Memory Performance of Rats
in the Barnes Maze Task ............................................................................................................ 87
4.1 Introduction......................................................................................................................... 87
4.2 Methods............................................................................................................................... 89
4.2.1 Animals........................................................................................................................ 89
4.2.2 Barnes maze protocols................................................................................................. 89
4.2.3 rTMS treatment............................................................................................................ 93
4.2.4 Data processing and statistical analyses ...................................................................... 95
4.3 Results................................................................................................................................. 96
4.3.1 rTMS disrupts spatial learning in the acquisition phase .............................................. 96
4.3.2 rTMS enhances cognitive flexibility in the reversal learning phase............................ 99
4.3.3 rTMS strengthens memories of previous target locations in probe trials .................. 102
4.3.4 Acute effects of rTMS are not evident on spatial learning and memory ................... 104
4.4 Discussion......................................................................................................................... 106
Chapter 5: Conclusion and Prospects..................................................................................... 109
5.1 Future development of the miniaturized TMS coil........................................................... 109
5.2 Future applications of the miniaturized TMS coil ............................................................ 111
5.2.1 Effects of rTMS on functional connectivity of neuron populations in primary
somatosensory and motor cortex ........................................................................................ 111
5.2.2 Modeling of the dynamic responses of neural activities during
pseudo-random TMS .......................................................................................................... 112
5.2.3 Effects of rTMS on working memory in a rodent DNMS task.................................. 114
References.................................................................................................................................. 117

vi
List of Tables
Table 1.1. Summary of typical TMS protocols for the treatment of various neurological
and neuropsychiatric disorders ..................................................................................................... 22
Table 3.1. Targeted cortex, corresponding stereotaxic coordinates of the first shank, and
unit yields of animal used in this study......................................................................................... 65
Table 4.1. Definitions of behavioral metrics used in this study.................................................. 96

vii
List of Figures
Figure 1.1. Different types of TMS coils and corresponding E-fields on the brain surface .......... 4
Figure 1.2. Graphic illustration of different TMS protocols and their after-effects ...................... 9
Figure 1.3. Schematic of TMS coil orientation above the scalp .................................................. 12
Figure 1.4. Basic principles of TMS ............................................................................................ 14
Figure 2.1. E-fields simulated with different TMS coils in free space ........................................ 26
Figure 2.2. Simplified schematic of the driving circuit of the TMS coil..................................... 29
Figure 2.3. B- and E-field measurement ...................................................................................... 31
Figure 2.4. TMS combined with electrophysiological recording and stimulation....................... 35
Figure 2.5. Electrical properties of the TMS coil ........................................................................ 41
Figure 2.6. B- and E-field distributions from measurement and simulation................................ 44
Figure 2.7. Simulated B- and E-field distributions in the rat brain.............................................. 46
Figure 2.8. rTMS facilitated firings of S1 and M1 neurons......................................................... 48
Figure 2.9. rTMS suppressed SSEPs ........................................................................................... 50
Figure 2.10. rTMS facilitated MEPs............................................................................................ 52
Figure 3.1. Simultaneous rTMS and intracranial electrophysiological recording in rat
sensorimotor cortex....................................................................................................................... 63
Figure 3.2. Evolvement of sensorimotor cortical signals with ketamine injections .................... 68
Figure 3.3. Alternation between slow-wave activities (SWA) and PPS as the signature
pattern of ketamine ....................................................................................................................... 68
Figure 3.4. PPS were effectively and reversibly suppressed by rTMS........................................ 69
Figure 3.5. Changes of LFP power distribution in different frequency bands in the S1
and M1 after each course of rTMS ............................................................................................... 71
Figure 3.6. Comparison of power spectra between pre-rTMS LFP during PPS (blue),
pre-rTMS LFP without PPS (red), and post-rTMS LFP without PPS (yellow) ........................... 72
Figure 3.7. Changes of PPS features after the first (black) and second (grey) courses of
rTMS in S1 (left) and M1 (right) .................................................................................................. 74
Figure 3.8. Spontaneous unitary activities recorded pre- and post-rTMS ................................... 76
Figure 3.9. Changes of neuronal spike firing rates (bin size: 2 s) before and after each
course of rTMS ............................................................................................................................. 77
Figure 3.10. Scatter plots of pre-rTMS versus post-rTMS firing rates (bin size: 1 min) of
neurons at different time points (1, 2, 3, 4, 5, 6, 10, 15 min) after each course of rTMS............. 78

viii
Figure 3.11. Comparison of neuronal spike firing rates between pre-rTMS signal during
PPS, pre-rTMS signal without PPS, and post-rTMS signal without PPS..................................... 79
Figure 4.1. Schematic representation of the TMS treatment combined with standard
Barnes maze protocols.................................................................................................................. 91
Figure 4.2. Schematic representation of the alternating TMS/sham treatment combined
with modified Barnes maze protocols........................................................................................... 93
Figure 4.3. Simulation of magnetic and electric fields within the rat brain................................ 94
Figure 4.4. rTMS hindered spatial learning during the acquisition phase of the Barnes
maze task....................................................................................................................................... 98
Figure 4.5. rTMS enhanced cognitive flexibility during the reversal learning phase of
the Barnes maze task................................................................................................................... 101
Figure 4.6. rTMS strengthened memories of previous target locations in the acquisition
(A) and reversal learning probe trials (B) of the Barnes maze task............................................ 103
Figure 4.7. Acute effects of rTMS were not evident in the Barnes maze task performance .... 105
Figure 5.1. Graphic illustration of C-shaped coils with different gaps...................................... 110
Figure 5.2. General structure of MIMO and MISO models....................................................... 111
Figure 5.3. Graphic illustration of the closed-loop neuromodulation system............................ 113
Figure 5.4. EEG recordings from different brain regions of rats via a customized
parylene-based EEG probe ......................................................................................................... 114
Figure 5.5. Diagram of DNMS task including sample phase, delay phase, and two
consequences in the nonmatch phase.......................................................................................... 115

ix
Abbreviations
ACC Anterior cingulate cortex
B-field Magnetic field
BM Barnes Maze
cTBS Continuous theta burst stimulation
DLPFC Dorsolateral prefrontal cortex
E-field Electric field
EEG Electroencephalography
EMG Electromyography
FDA Food and Drug Administration
GABA Gamma-aminobutyric acid
imTBS Intermediate theta burst stimulation
iTBS Intermittent theta burst stimulation
LFP Local field potential
LTD Long-term depression
LTP Long-term potentiation
M1 Primary motor cortex
MEA Multi-electrode array
MEP Motor evoked potential
mPFC Medial prefrontal cortex
MRI Magnetic resonance imaging
MWM Morris water maze
NMDA N-methyl-D-aspartate
OCD Obsessive-compulsive disorder
OFC Orbitofrontal cortex
PFC Prefrontal cortex
PPS Poly population spikes
rTMS Repetitive transcranial magnetic stimulation
S1 Primary somatosensory cortex
SSEP Somatosensory evoked potential
SUA Single-unit activity
SWA Slow-wave activity
TBS Theta burst stimulation
TMS Transcranial magnetic stimulation

x
Abstract
Transcranial magnetic stimulation (TMS) is a non-invasive technique widely used for
neuromodulation. Animal models are essential for investigating the underlying mechanisms of
TMS. However, the lack of miniaturized coils hinders the TMS studies in small animals, since
most commercial coils are designed for humans and thus incapable of focal stimulation in small
animals. Furthermore, it is difficult to perform electrophysiological recordings at the TMS focal
point using conventional circular or figure-eight coils.
To address this, we designed, fabricated, and tested a novel miniaturized TMS coil that
consisted of a C-shaped iron powder core and insulated copper wires. The resulting magnetic and
electric fields were characterized with experimental measurements and finite element modeling.
The efficacy of this coil in neuromodulation was validated with electrophysiological recordings of
single-unit activities (SUAs), somatosensory evoked potentials (SSEPs), and motor evoked
potentials (MEPs) in rats. With subthreshold repetitive TMS (rTMS) focally delivered over the
sensorimotor cortex, distinct modulatory effects on SUAs, SSEPs, and MEPs with the same
stimulation protocol were observed in anesthetized rats. It indicates that multiple neurobiological
mechanisms in the sensorimotor pathways are differentially modulated by rTMS.
In addition, we developed and applied a rodent model that enabled simultaneous rTMS
treatment and pharmacological manipulations. The effect of rTMS and ketamine on brain
oscillations was explored. A novel form of synchronized activities, poly population spikes (PPS),
was discovered as the biomarker of ketamine in local field potentials. Brief subthreshold rTMS
effectively and reversibly suppressed PPS while increasing neuronal firing. These results suggest
that ketamine and rTMS have convergent but opposing effects on cortical oscillations and circuits.

xi
Furthermore, we investigated both the chronic and acute effects of rTMS on learning and
memory in rats performing the Barnes maze task. The complex effects of subthreshold rTMS on
various forms of cognition were observed, from the initial acquisition of spatial information to
flexibility in adapting to changed environments and the strength of spatial memory retention.
Specifically, our findings reveal that chronic subthreshold rTMS initially disrupted spatial learning
in rats during the acquisition phase of the Barnes maze task. It enhanced cognitive flexibility
during the reversal learning phase and strengthened spatial memories of previous target locations
in the acquisition and reversal learning probe trials. However, the acute effects of rTMS were not
evident on spatial learning and memory.
This novel miniaturized coil and the associated experimental paradigms facilitate the
integration of TMS, standard electrophysiology, and pharmacological manipulations in rat brains.
It provides a powerful tool for investigating the underlying neurobiological mechanisms of TMS
from the single neuron to behavioral level in small animal models, as well as for optimizing
therapeutic strategies for the treatment of neurological and neuropsychiatric disorders.

1
Chapter 1: Introduction and Background
1.1 Motivation
Neuromodulation alters the activity in the nervous system through the targeted delivery of
controlled physical energy or pharmaceutical agents. It is the process of inhibition, excitation,
modification, or regulation of neural circuits and brain functions to improve the quality of life in
humans. In the last few decades, we have witnessed a dramatic increase in neuromodulation
technology such as deep brain stimulation (Perlmutter & Mink, 2006), transcranial direct current
stimulation (Nitsche et al., 2008), optogenetics (Mickle et al., 2019), transcranial focused
ultrasound (K. Yu et al., 2021), targeted drug delivery system (Christo & Bottros, 2014), etc. One
of the neuromodulation techniques, transcranial magnetic stimulation (TMS), has gained
popularity due to its high-spatiotemporal resolution (compared to transcranial direct current
stimulation), its non-invasive nature (compared to deep brain stimulation and targeted drug
delivery system), and its well-established protocols (compared to optogenetics and transcranial
focused ultrasound).
TMS has been used extensively for the treatment of depression (J. Chen et al., 2013; Levkovitz
et al., 2015; McDonald et al., 2011; O’Reardon et al., 2007), obsessive-compulsive disorder
(Carmi et al., 2018, 2019), migraine (Lipton et al., 2010), stroke (Mansur et al., 2005), epilepsy
(W. Sun et al., 2012), schizophrenia (Brunelin et al., 2006), autism (Enticott et al., 2012), and other
neurological and neuropsychiatric disorders. However, the underlying mechanisms of its effects
remain largely unknown. Animal models are essential for investigating such underlying
mechanisms since they allow more comprehensive electrophysiological recordings and invasive
experiments that are difficult or impractical on humans. However, most commercial TMS coils

2
are designed for humans. The large geometric size of human TMS coils will cause non-focal highintensity stimulation, which are not applicable to small animal studies (Alekseichuk et al., 2019;
Tang et al., 2017; Vahabzadeh-Hagh et al., 2012). Therefore, there is a strong need for
miniaturized TMS coils in neuroscience, neural engineering, and clinical research communities.
Several studies have reported the designs of rodent-specific TMS coils (Bagherzadeh et al.,
2022; Cobos Sánchez et al., 2020; Khokhar et al., 2021; Tang et al., 2016). In general, these coils
either cannot be conveniently used in conjunction with standard electrophysiology or have distinct
magnetic and electric field (B- and E-field) distributions compared to those circular or figure-eight
coils, which are most commonly used in humans (Deng et al., 2013; Ueno & Sekino, 2021).
Consequently, the neural responses elicited with these coils cannot be directly compared with those
in human TMS studies.
To develop a TMS coil that can (1) generate focal B- and E-fields similar to those of human
figure-eight coils, and (2) is compatible with standard electrophysiological recording and
stimulation, we designed, fabricated, characterized, and evaluated a novel C-shaped miniaturized
coil for use in small animals (Jiang et al., 2021, 2022). The resulting B- and E-fields were
simulated with finite element modeling and further characterized with experimental measurements.
The efficacy of this TMS coil in neuromodulation was validated with electrophysiological
recordings of single-unit activities (SUAs), local field potentials (LFPs), somatosensory evoked
potentials (SSEPs), and motor evoked potentials (MEPs) before and after repetitive TMS (rTMS)
in rats using single electrodes and multi-electrode arrays (MEAs).
This novel miniaturized coil and the associated experimental paradigms facilitate the
integration of TMS, standard electrophysiology, and pharmacological manipulations in rat brains.
It provides a powerful tool for investigating the underlying neurobiological mechanisms of TMS

3
from the single neuron to behavioral level in small animal models, as well as for optimizing
therapeutic strategies for the treatment of neurological and neuropsychiatric disorders.
The body of work presented here begins with a background review of the need, to TMS system
design, prototyping, testing on benchtop and then in rodents, and finally concluding with the plan
for next-generation miniaturized coil design and its future applications, including developing
closed-loop neuromodulation systems and exploring the effects of TMS on brain functional
connectivity and memory performance in behaving animals.
1.2 TMS coils
1.2.1 Types of TMS coils
A vast of TMS coils have been designed or proposed for use in humans. Different geometries
or placements of coils result in different electromagnetic field distributions. Knowledges on coil
design rationale and electromagnetic field distribution are important for understanding the
mechanisms behind TMS, as well as for developing novel coils with improved performance. The
first modern TMS system introduced in 1985 used a flat circular coil with an outer diameter of 100
mm (Barker et al., 1985). A brief, high-current pulse (peak value of 4 kA after 110 s) passing
through the coil above the motor cortex elicited limb movements in patients. However, the circular
coil induces a ring-shaped E-field with maximum strength under the coil perimeter (Fig. 1.1A). It
is difficult to deliver focal TMS over specific brain regions (L. G. Cohen et al., 1990; Deng et al.,
2013). In order to improve the stimulation focality, a figure-eight coil was developed subsequently
(Ueno et al., 1988). A pair of oppositely connected coils were placed adjacently to induce two
ring-shaped E-fields in opposite directions (Fig. 1.1B). The superposition of those two fields

4
reinforced the stimulation intensity under the center of the coil, realizing a focal TMS above the
motor cortex with a 5-mm resolution (Ueno et al., 1990).
Figure 1.1. Different types of TMS coils and corresponding E-fields on the brain surface: (A) circular coil,
(B) figure-eight coil, (C) butterfly or double-cone coil (D) eccentric figure-eight coil, (E) cloverleaf coil,
(F) slinky coil, (G) figure-eight coil with iron core, and (H) figure-eight coil with conductive shield plate.
Modified from Deng et al. (2013).
Various modifications of the figure-eight coil have been made to further improve the
performance. A butterfly or double-cone coil (Fig. 1.1C) was constructed based on the
configuration of the figure-eight coil by creating an angle between each wing. It fits the curvature
of human head and increases the stimulation depth (Deng et al., 2013; Ueno & Sekino, 2021).
Additionally, an eccentric figure-eight coil (Fig. 1.1D) was made by changing the in-plane winding
geometry. The center-dense windings induce a more focal E-field with a lower driving current
than the concentric windings of conventional figure-eight coils (Deng et al., 2013; Sekino et al.,

5
2015). Furthermore, the combination of four sets of circular coils (or two sets of figure-eight coils)
forms a cloverleaf coil (Fig. 1.1E). This configuration shows flexibility in E-field distribution.
When opposing currents are delivered simultaneously in adjacent coils, the peaks of E-field exist
at the center of each pair of coils (Roth et al., 1994; Ruohonen et al., 1998). Besides, one group
has demonstrated that a rotating E-field can be created by applying biphasic currents in the
cloverleaf coil with a phase shift in time domain to relieve orientation sensitivity (Rotem et al.,
2014). Another variation of the figure-eight coil is the slinky coil (Fig. 1.1F). It has multiple
circular loops forming a helical coil on a half torus. The coil windings concentrated over the target
area further increase the stimulation focality (Chunye Ren et al., 1995; Krasteva et al., 2002;
Zimmermann & Simpson, 1996).
Other efforts have been made to improve coil performance including the use of ferromagnetic
cores and conductive shield plates. A modified figure-eight coil can be fitted into a C-shaped iron
core (Fig. 1.1G). The use of an iron core increases the stimulation intensity and power efficiency,
while also reducing heat generation (Epstein & Davey, 2002). However, the iron core must be
laminated to minimize eddy currents, and the saturation of the core at high coil currents could be
a potential issue. A conductive shield plate can be added underneath the figure-eight coil (Fig.
1.1H). The introduction of the shield plate reroutes and attenuates the B-field. It largely increases
the stimulation focality at the expense of intensity reduction and energy loss (D.-H. Kim et al.,
2006; Lu & Ueno, 2009).
1.2.2 Challenges of making small TMS coils
Although many attempts have been made to scale down the TMS coils, the outcome is not fully
satisfactory. Challenges exist in making small TMS coils for the use in small animals such as

6
rodents. E-field focality and depth of penetration are two key features in coil design. E-field
focality is conventionally defined as the surface area with an intensity greater than a certain
percentage of the maximum value (e.g., 50%: half-value area). Besides, Deng et al. (2013)
proposed a more robust method to quantity the focality by the tangential spread, which can be
calculated as the ratio between half-value volume and depth of penetration (defined as half-value
depth). However, there is a trade-off between the E-field focality and depth of penetration. In
general, the E-field induced by smaller coil is more localized but superficial (Deng et al., 2013).
Human TMS coils have larger dimensions compared with the size of rodent brains. Considering
that the volume of human brains is about 1000 times larger than that of rodent brains (Welniak–
Kaminska et al., 2019; X. Yu et al., 2014), the coil’s diameter should be reduced by ~90%
proportionally. Cohen and Cuffin (1991) suggested that the E-field intensity will reduce to 4%
when the diameter of the figure-eight coil is reduced to 20%. Therefore, an extremely high current
is required to drive the coil with a similar intensity. However, high current pulse may cause
mechanical stress and heating problems, which limits the diameter of the figure-eight coil to a
minimum of 25 mm (Yunokuchi & Cohen, 1991). Otherwise, a further reduction in coil size comes
at the expense of reduced stimulation intensity.
Several studies reported rodent-specific coils designed with high-density windings around
ferromagnetic cores (Makowiecki et al., 2014; Rodger et al., 2012; Tang et al., 2016). The use of
ferromagnetic cores with high permeability concentrates the magnetic flux in the core material,
making a more focal E-field. However, the side effects of ferromagnetic cores must be taken into
account. The changing B-field produced by the coil will induce eddy currents in the core and thus
cause heat and energy losses. The eddy current loss can be alleviated by using laminated,
powdered, or nonconductive cores. Besides, core materials like iron or ferrite have high

7
permeability but low saturation flux density. The small saturation threshold limits the maximum
B-field strength, and it may cause undesirable loading of the coil driving circuit (Bagherzadeh et
al., 2022). Thus, materials with high saturation flux density should be considered.
Although high-density windings and ferromagnetic cores increase the magnetic flux density,
they also increase the inductance of the coil, which will cause a longer rise time of coil current
(Wilson et al., 2018). According to Faraday’s law of induction, the E-field intensity largely
depends on the rate of change of magnetic flux density which is determined by the rate of change
of coil current. A slow response in current will decrease the E-field intensity. Hence, a balance
between magnetic flux density and coil inductance should be considered to optimize the E-field
intensity.
In addition, skin and proximity effects in the coil windings could be another potential problem
(Ravazzani et al., 2002). The skin effect is the tendency for the coil current to flow mostly near
the surface of the windings; the proximity effect is the tendency for nonhomogeneous distribution
of current density due to the changing currents in nearby conductors. Both effects will reduce the
effective area of the conductor and increase the overall resistance. Undesirable heating and lower
efficiency are expected especially at high current frequencies. These effects can be mitigated by
using litz wires. Litz wires consist of multiple insulated wire strands twisted together, so the
current will be distributed evenly within the conductor.
Coil heating is inevitable during long-term stimulation such as rTMS. A liquid- or air-cooling
system is often combined with the human TMS device (Navarro de Lara et al., 2021; Pridmore &
Ang, 2008) to reduce excessive heating. However, few cooling system exists for small coils. It is
unknown how effective the cooling system will be after the size is reduced proportionally.
Additionally, most animal TMS studies were performed in anesthetized or restrained animals

8
(Vahabzadeh-Hagh et al., 2012). Not all observations obtained in those studies can be directly
translated to humans. Thus, efforts should be made to deliver TMS in freely moving animals. A
headpost implanted into the animal’s skull can be used to stabilize the coil and guide TMS over
specific brain regions (Cermak et al., 2020).
1.3 TMS parameters
1.3.1 Pattern
Numerous TMS protocols have been explored to evaluate the after-effects (Fig. 1.2). In general,
the stimulation pattern can be classified into single-pulse TMS and rTMS. Single-pulse TMS was
introduced in the 1980s to assess motor function. Limb movements accompanied by MEP
recordings were elicited by single-pulse TMS over the motor cortex (Barker et al., 1985). To date,
single-pulse TMS in conjunction with electrophysiological recordings and brain imaging
techniques are used to assess and modify brain circuitries and functions (Bergmann et al., 2016).
For example, the combination of electroencephalography (EEG) and TMS enables the
investigation and mapping of brain regions through recordings of TMS-evoked potentials and
cortical oscillations (Taylor et al., 2008). In addition, SUA has been investigated to evaluate the
neural effects at the single-cell level. With artifact suppression techniques, excitation or inhibition
of different types of neurons were observed immediately following single-pulse TMS in rats (B.
Li et al., 2017), cats (Moliadze et al., 2003), and non-human primates (Mueller et al., 2014; Romero
et al., 2019; Tischler et al., 2011). The effects can last from a few milliseconds to hundreds of
milliseconds. Paired-pulse TMS is another type of single-pulse TMS: a subthreshold conditioning
stimulus is delivered followed by a suprathreshold test stimulus with variable inter-stimulus

9
intervals. The modulatory effect of paired-pulse TMS depends on the intensity as well as the
interval (Moliadze et al., 2005; Tokimura et al., 1996; Valls-Solé et al., 1992).
Figure 1.2. Graphic illustration of different TMS protocols and their after-effects.
The application of repeated TMS pulses induces prolonged changes in brain activities that
outlast the period of stimulation. The frequency of rTMS is crucial for determining its after-effects.
In general, high frequency (>5 Hz) rTMS has a facilitatory effect on cortical excitability, whereas
low frequency (0.2-1 Hz) rTMS usually has an inhibitory effect (Hallett, 2007; Hoogendam et al.,
2010; Matheson et al., 2016). The effects of rTMS will typically last for about half the duration
of the stimulation train (R. Chen et al., 1997; Guse et al., 2010). Besides simple rTMS protocols,
another form of rTMS, theta burst stimulation (TBS), is widely used to induce brain plasticity.

10
The typical paradigm of TBS contains three TMS pulses at 50 Hz which are repeated at 5 Hz (di
Lazzaro et al., 2005; Y.-Z. Huang et al., 2005). This pattern was designed based on the theta
rhythm (4-8 Hz) generated by the hippocampus (Capocchi et al., 1992). There are three different
types of TBS including continuous TBS (cTBS), intermittent TBS (cTBS), and intermediate TBS
(imTBS). Huang et al. (2005) suggested that cTBS (40 s of uninterrupted TBS) leads to an
inhibitory effect on MEPs, whereas iTBS (2 s of TBS repeated every 10 s for a total of 190 s) leads
to an excitatory effect. However, imTBS (5 s of TBS repeated every 15 s for a total of 110 s) has
no obvious effects on MEPs. It is might due to a mixture of opposing effects on long-term
potentiation (LTP) and long-term depression (LTD).
1.3.2 Intensity
The intensity of TMS plays an important role in determining the resulting effects on brain
activities. Motor threshold is typically defined as the minimum TMS intensity to elicit a positive
MEP in 50% of 10-20 consecutive trials (Rossini et al., 1994). It reflects the excitability of
descending motor pathways. Now it is common practice to set the intensity of TMS as the
percentage of motor threshold. Suprathreshold and subthreshold TMS will lead to different
patterns of neural responses. Suprathreshold intensity is used in single-pulse or paired-pulse TMS
protocols to elicit MEPs. In addition, both suprathreshold and subthreshold intensity are used in
rTMS protocols. However, the effects of rTMS with same frequency can be reversed due to the
stimulation intensity. For example, low frequency suprathreshold rTMS (1 Hz, 115% of motor
threshold) decreases the MEP amplitude (R. Chen et al., 1997), whereas subthreshold rTMS at 1
Hz often fails to induce measurable changes on the motor cortex excitability (Klomjai et al., 2015).
Furthermore, previous studies have shown significant improvement following suprathreshold

11
rTMS (110-120% of the motor threshold) delivered over the dorsolateral prefrontal cortex (DLPFC)
in patients with major depression (Blumberger et al., 2012; Galletly et al., 2012). Although
subthreshold rTMS has inferior therapeutic effects for treating depression compared with
suprathreshold rTMS (Boutros et al., 2002; Padberg, 2002), subthreshold TBS (80% of motor
threshold) has been shown to have similar antidepressant effects as conventional suprathreshold
rTMS protocols (Chung et al., 2015; Voigt et al., 2021). It requires less stimulation time and
produces longer lasting effects (Y.-Z. Huang et al., 2005). Moreover, subthreshold rTMS mitigates
the safety concerns about long duration of stimulation. Even at low intensities (less than ~100
mT), subthreshold rTMS has been shown to have positive effect in patients with depression
(Martiny et al., 2010; Rohan et al., 2014). Additionally, recent work suggests that low intensity
rTMS alters the structure and function of neural circuits in vivo (Makowiecki et al., 2014; Rodger
et al., 2012), and lowers action potential threshold while increasing evoked spike firing rate in
vitro (Tang et al., 2016).
1.3.3 Orientation
It is well known that TMS coils with different orientations and positions will produce distinct
electromagnetic field distributions in the brain. Previous studies have shown the MEP amplitude
and motor threshold as a function of coil orientation over the human motor cortex (Richter et al.,
2013; Rösler et al., 1989). It is common practice to set the coil orientation yielding the maximum
MEP amplitude as the optimal orientation. When using a figure-eight coil, the coil placed in the
posterior-lateral orientation over the hand motor area (Fig. 1.3A) is assumed as the standard
protocol for hand muscle stimulation (Brasil-Neto et al., 1992; Mills et al., 1992), whereas a coil
placed in the medial-lateral orientation over the foot motor area (Fig. 1.3B) is standard for foot

12
muscle stimulation (Terao et al., 2009). Different coil orientations may activate different
microcircuits in the brain. Li et al. (2017) suggested that the current induced in the medial-lateral
orientation (Fig. 1.3C) directly activates pyramidal neurons in layer 5 of the M1, whereas the
current in the posterior-anterior orientation (Fig. 1.3D) evokes trans-synaptic spiking activities
involving multiple layers of the M1 in rodents.
Figure 1.3. Schematic of TMS coil orientation above the scalp. (A) Standard coil orientation for hand
muscle stimulation in humans. (B) Standard coil orientation for foot muscle stimulation in humans. (C)
Medial-lateral coil orientation in rodents. (D) Posterior-anterior coil orientation in rodents. The arrow
indicates the current vector induced in the brain.
For efficient and focal stimulation, it is critical to optimize the coil orientation and position.
However, it might be difficult to directly measure the responses for activation of some cortical
areas such as the DLPFC (the target for treating depression). Various strategies have been
proposed including the use of cranial landmarks (George et al., 1995), 10-20 EEG system
(Okamoto et al., 2004), and MRI-guided neuronavigation (Ahdab et al., 2010; Herwig et al., 2001).
In addition, computational modeling with realistic head models can be used to further optimize the

13
coil positioning, minimizing the error caused by the secondary E-field generated inside the head
due to charges build-up at tissue boundaries (Janssen et al., 2015; Salinas et al., 2009).
1.4 Mechanisms of TMS
1.4.1 Physics behind TMS
In TMS, a coil is placed over the scalp. A brief, high-current pulse passing through the coil
produces a time-varying B-field perpendicular to the plane of the coil. The B-field in turn induces
an E-field in the brain with a circular current (an eddy current), which flows in the direction
opposite to the coil current (Fig. 1.4A). According to Faraday’s law of induction, the E-field (E)
induced by the coil can be expressed in terms of potentials:
= −

−
where A is the magnetic vector potential whose curl is equal to the B-field (B): ∇ × = , and
is the scalar electric potential. The first term (−

) in the above equation describes the nonconservative E-field contributed by the changing current in the coil, whereas the second term (−∇)
describes the conservative E-field contributed by the surface charge accumulated at the boundaries.

14
Figure 1.4. Basic principles of TMS. (A) Illustration of the current flowing in the TMS coil and the induced
current in the brain. Modified from Ridding and Rothwell (2007). (B) Illustration of the magnetic vector
potential produced by the current in the circular coil.
When using a circular coil (Fig. 1.4B), the magnetic vector potential arising from the coil
current (I) can be calculated as:
=
0
4
∮

where 0
is the permeability of free space; is number of turns; the line integral is around the coil
loop with distance . As a result, the induced E-field is proportional to the rate of change of the
coil current.
1.4.2 Synaptic plasticity
The cellular and molecular mechanisms of the long-lasting effects of rTMS is insufficiently
understood. The most widely accepted hypothesis is that rTMS induces changes in synaptic
efficacy between cortical neurons, including LTP- and LTD-like plasticity (Hoogendam et al.,

15
2010). This phenomenon was first observed in the hippocampus triggered by electrical stimulation
at a certain frequency (Bliss & Gardner-Medwin, 1973; Levy & Steward, 1983). The cellular basis
of these changes in synaptic strength is related to the glutamatergic N-methyl-D-aspartate (NMDA)
receptor. At the resting membrane potential, Mg2+ blocks the ion channel in the NMDA receptor.
When electrical stimulation or TMS is applied, the postsynaptic membrane will be depolarized,
expelling the Mg2+ from the NMDA receptor. As a result, the NMDA receptor is activated,
allowing the opening of the ion channel and the Ca2+ influx. It appears that LTP can be induced
by a large and rapid rise in the Ca2+ concentration triggered by high frequency stimulation, whereas
LTD can be induced by a small and slow rise triggered by low frequency stimulation (Luscher &
Malenka, 2012). Combination of rTMS and pharmacological interventions can be used to evaluate
the NMDA dependency of the after-effects of rTMS. Previous studies indicated that the
administration of NMDA-receptor antagonists blocked the facilitatory and inhibitory effects of
different rTMS protocols in humans (Y.-Z. Huang et al., 2007; Stefan et al., 2002; Teo et al., 2007).
Additionally, studies using rodent hippocampal slice cultures suggested that the Ca2+
-dependent
strengthening of excitatory synapses induced by 10 Hz rTMS depended on a cooperative activation
of voltage-gated Na+
channels, L-type voltage-gated Ca2+ channels, and NMDA receptors (Lenz
et al., 2015; Vlachos et al., 2012). Furthermore, rTMS can alter the structure and function of not
only excitatory synapses but also inhibitory synapses. A reduction in Ca2+
-dependent gammaaminobutyric acid (GABA)-ergic synaptic strength was observed in CA1 pyramidal neurons in
vitro after 10 Hz rTMS (Lenz et al., 2016). These studies support the hypothesis that rTMS is
highly likely to induce LTP or LTD-like plasticity between cortical neurons.

16
1.4.3 Neurotransmitters
The activation of neurotransmitters could be another possible working mechanism of rTMS.
Neurotransmitters are signaling molecules allowing for communication between neurons
throughout the body. Major neurotransmitters involved in rTMS include dopamine, serotonin,
glutamate, and GABA. Dopamine plays an important role in helping people feel pleasure as part
of the brain's reward system and regulates body movements. Several studies revealed that 10 Hz
rTMS over the left DLPFC or the left M1 led to the release of dopamine in the striatum, producing
beneficial effects for the treatment of Parkinson’s disease (Strafella, 2003; Strafella et al., 2001).
Moreover, it is well known that dysregulation of the dopamine system correlates with depression
(Belujon & Grace, 2017). The increase in dopamine release might be related to the therapeutic
action of rTMS in depressive patients (Pogarell et al., 2007). Serotonin, an inhibitory
neurotransmitter mainly for modulating emotion and mood, has also been extensively investigated
in the treatment of depression (Moncrieff et al., 2022). Similarly, acute 10 Hz rTMS over the left
DLPFC modulates serotonin release in limbic areas, which may contribute to the antidepressant
mechanism of rTMS (Sibon et al., 2007). Glutamate and GABA are the major excitatory and
inhibitory neurotransmitters in the central nervous system. The balance between glutamatergic
excitation and GABAergic inhibition is crucial for maintaining proper functioning of neuronal
circuits (Hampe et al., 2018; Lazarevic et al., 2013). Glutamatergic or GABAergic deficits may
play an additional role in the pathophysiology of depression (Croarkin et al., 2011; Sanacora et al.,
2012). Previous studies showed that 10 Hz rTMS over the left DLPFC not only increased the
glutamine/glutamate ratio but also the GABA levels in patients with major depression (Croarkin
et al., 2016; M. J. Dubin et al., 2016). Interestingly, glutamate is formed from glutamine and
subsequently synthesized into GABA by the action of glutamate decarboxylase (Bak et al., 2006).

17
It implies that the synthesis and metabolism of neurotransmitters might be involved in the
mechanism of rTMS.
1.4.4 Other mechanisms
rTMS outcomes affect regulation of gene and protein expression in the brain. Brain-derived
neurotrophic factor (BDNF) is a key protein supporting neuronal survival and growth. It also plays
an important role in neuronal plasticity related to learning and memory (Bathina & Das, 2015).
Previous studies revealed that long-term rTMS applied at 20 Hz activated the expression of BDNF
in the rodent hippocampal slice cultures (Müller, 2000) and increased the plasma BDNF levels in
patients with depression (Yukimasa et al., 2006). Moreover, high frequency iTBS was shown to
enhance the expression of zif268 in rodent brains, which is a key protein for memory consolidation
and facilitation of hippocampal synaptic plasticity (Aydin-Abidin et al., 2008; Penke et al., 2014).
Furthermore, an increase in expression of plasticity genes NTRK2 and MAPK9 induced by iTBS
was observed in human neurons in vitro (Thomson et al., 2020). This evidence supports the
plasticity mechanism for the long-lasting effects induced by rTMS.
rTMS also has an impact on the mechanisms of neurogenesis and apoptosis. One study
suggested that 1 Hz rTMS over the superior temporal gyrus increased gray matter volume in the
auditory cortex in humans (May et al., 2007). Additionally, high frequency rTMS has been shown
to enhance neurogenesis in the dentate gyrus (Ueyama et al., 2011) and promote neural stem cell
proliferation in rodents (Guo et al., 2014). Furthermore, studies revealed that 20 Hz rTMS
increased glucose metabolism and inhibited apoptosis in rodent models of transient ischemic attack
(Gao et al., 2010). Similarly, an anti-apoptotic effect via regulating Bcl-2 and Bax protein

18
expression in the hippocampus was observed following low frequency rTMS in rodent models of
vascular dementia (H.-Y. Yang et al., 2015).
The mechanism of the relatively short-term effects (of the order of seconds or a few minutes)
of rTMS may differ from the long-term effects. One plausible hypothesis is that rTMS induces
acute changes in ironic balance or membrane capacitance in neurons, which modulates the cortical
excitability in a short-term manner (Kuwabara et al., 2002). Besides, the reafferent feedback to
the stimulation target may also contribute to the effect of suprathreshold rTMS over the
sensorimotor cortex (Ridding & Rothwell, 2007).
1.5 TMS and neuromodulation
1.5.1 Electrophysiological effects of TMS
Electrophysiological measurements including SUA, LFP, EEG, and evoked potentials can be
used to capture the changes in neuronal activities induced by TMS. SUA, LFP and EEG are
extracellularly recorded signals from a local network of neurons. SUA reveals the microscopiclevel spiking activity of neuronal assemblies; LFP (also known as intracranial EEG) represents the
total synaptic activity located within a small volume surrounding the recording electrode in the
brain, while EEG measures similar cortical activity from the scalp at a macroscopic level. Distinct
from the spontaneous potentials detected by LFP or EEG, an evoked potential constitutes a form
of neural or muscular activity which is elicited by an external stimulus such as electrical
stimulation or TMS.
The main challenges in combining TMS and electrophysiology are the positioning of the TMS
coil in conjunction with the electrode implant, and the interference caused by the TMS pulse during

19
recordings. To assess the immediate effects of TMS, artifact suppression techniques are usually
required ensuring full recovery of true brain signals. At the single-cell level, excitation or
inhibition of different types of neurons can be observed within milliseconds to hundreds of
milliseconds after the single-pulse TMS (B. Li et al., 2017; Moliadze et al., 2003; Mueller et al.,
2014; Romero et al., 2019; Tischler et al., 2011). Additionally, rTMS has been shown to
effectively modulate cortical oscillations in different frequency bands of LFP or EEG signals (Ding
et al., 2014; Paus et al., 2001; J. Yang et al., 2021; Zmeykina et al., 2020). Moreover, the
combination of TMS and EEG enables the investigation of changes in connectivity strength
between different brain regions, mapping the dynamics of functional connectivity induced by TMS
(Bortoletto et al., 2015; Corlier et al., 2019; Miniussi & Thut, 2010; Rogasch & Fitzgerald, 2013).
Evoked potentials allow studying the integrity, connectivity, and excitability of the nervous system.
The waveform is typically obtained by isolating the EEG, LFP, or electromyography (EMG)
responses that are phase-locked to the suprathreshold single-pulse TMS. TMS-evoked potentials
recorded from the brain have been used to assess cortical physiology in health and disease (Farzan
et al., 2016; Kaskie & Ferrarelli, 2018; Paolo, 2015). In particular, the MEP elicited by TMS over
the motor cortex is one of the hallmark measures for the integrity and excitability of the descending
motor pathways. Besides, the MEP is the most commonly used measurement in evaluating the
effects of various TMS protocols (Y.-Z. Huang et al., 2005; Klomjai et al., 2015; Muller et al.,
2014; Tang et al., 2016; Tsuji & Rothwell, 2002).
In addition, TMS has been shown to modulate motor learning and memory formation in humans
and animals (Kirschen et al., 2006; Mix et al., 2010; Reis et al., 2008; Tang et al., 2018). These
behavior changes might be associated with specific neurophysiological characteristics. Although
the feasibility of simultaneous TMS and electrophysiology has been demonstrated, the immediate

20
effects of TMS on spontaneous, evoked or task-related brain activities from the microscopic to
macroscopic levels remain to be investigated.
1.5.2 Treatments of neurological and neuropsychiatric disorders
TMS was approved by the Food and Drug Administration (FDA) for the treatment of major
depression in 2008, pain associated with certain migraine headaches in 2013, and OCD in 2018
(Table 1.1).
Major depression is a common mental health disorder marked by persistently depressed mood
or loss of interest. Specific brain regions such as the DLPFC and deeper limbic areas were found
to be functioning abnormally in patients with major depression (George et al., 1994; Ketter et al.,
1996). There are two FDA-cleared rTMS protocols for treating major depression in adults. One
used a figure-eight coil to deliver suprathreshold rTMS at 10 Hz over the left DLPFC (O’Reardon
et al., 2007). The other one used an H1 coil to deliver suprathreshold rTMS at 18 Hz over the
same brain region with shorter pulse trains and intertrain intervals (Levkovitz et al., 2015). In
addition, another group used a figure-eight coil to deliver suprathreshold rTMS at 1 Hz over the
right DLPFC (McDonald et al., 2011). Both high frequency and low frequency rTMS protocols
have been shown to have antidepressant effects in patients with major depression (J. Chen et al.,
2013).
Migraines are a recurring type of headache and are sometimes preceded by an aura. eNeura
Therapeutics’ Cerena TMS device is the first FDA-approved single-pulse stimulator for relieving
pain associated with migraines with aura. Studies suggested that single-pulse TMS was
significantly more effective for treating migraines with aura after the first attack, with a higher

21
proportion of patients who were pain-free at 2 hours after TMS compared to those treated with a
sham stimulation (Lipton et al., 2010).
OCD is characterized by recurring, unwanted thoughts, ideas, or sensations that lead to
repetitive behaviors. Some deep brain structures such as the medial prefrontal cortex (mPFC) and
the anterior cingulate cortex (ACC) were found to be hyperactive in patients with OCD (Fitzgerald
et al., 2005; Herrmann et al., 2004). The FDA-cleared rTMS protocols use either an H7 coil or a
double-cone coil to deliver 20 Hz deep rTMS over the bilateral mPFC and ACC; it has been shown
to significantly improve OCD symptoms in patients (Carmi et al., 2018, 2019).
In addition, TMS has been reported to have a beneficial effect in stroke (Mansur et al., 2005),
epilepsy (W. Sun et al., 2012), schizophrenia (Brunelin et al., 2006), autism (Enticott et al., 2012),
and other neurological and neuropsychiatric disorders (Table 1.1). Unfortunately, many of these
studies are pilot studies with a small sample size. Further investigation is required to improve the
understanding of the mechanisms behind these reported successes.
Applications Coil Location Frequency Intensity Duration Measures Comments
Major
depression
Figureeight
coil
Left
DLPFC
10 Hz (4s
pulse trains
with 26s
ITI)
120%
motor
threshold
3000
pulses per
session
(~38 min)
Hamilton
Rating
Scale for
Depression
Antidepressant
effect was
observed
(FDA-cleared
protocol).
Major
depression
H1 coil Left
DLPFC
18 Hz (2s
pulse trains
with 20s
ITI)
120%
motor
threshold
1980
pulses per
session
(~30 min)
Hamilton
Rating
Scale for
Depression
Antidepressant
effect was
observed
(FDA-cleared
protocol).
Major
depression
Figureeight
coil
Right
DLPFC
1 Hz 120%
motor
threshold
1800
pulses per
session
(30 min)
Hamilton
Rating
Scale for
Depression
Antidepressant
effect was
observed.
Migraine with
aura
Circular
coil
Occiput Single
pulse
0.9T
peak
with
2 pulses
about 30s
apart
Pain-free
response
rate
It was effective
for the acute
treatment of

22
180μs
rise time
migraine with
aura after the
first attack.
OCD H7 coil
or
doublecone
coil
mPFC
and ACC
20 Hz (2s
pulse trains
with 20s
ITI)
100%
motor
threshold
2000
pulses
(18 min)
Yale-Brown
Obsessive
Compulsive
Scale
Positive effects
on OCD
symptoms were
observed
(FDA-cleared
protocol).
Stroke Figureeight
coil
Contrales
-ional
motor
cortex
1 Hz 100%
motor
threshold
600 pulses
(10 min)
SRT, CRT,
PPT,
finger
tapping
speed
It significantly
improved
performance on
reaction time
and PPT.
Epilepsy Figureeight
coil
Epileptogenic
focus
0.5 Hz 90%
motor
threshold
3 sessions
of 500
pulses per
day
Seizures
frequency,
epileptogenic
localization,
SCL-90
Significant
antiepileptic
effects on
patients with
refractory
partial seizures
were observed.
Schizophrenia Figureeight
coil
Left
temporop
-arietal
cortex
1 Hz 90%
motor
threshold
2 sessions
of 1000
pulses per
day
AHRS,
SAPS
There was
significant
improvement in
AHRS
compared to
sham
stimulation.
Autism Figureeight
coil
Left M1
or SMA
1 Hz 100%
motor
threshold
900 pulses
(15 min)
Movementrelated
cortical
potentials
rTMS over
SMA and left
M1 improved
the early and
late component
of MRCPs,
respectively.
Table 1.1. Summary of typical TMS protocols for the treatment of various neurological and
neuropsychiatric disorders. ITI, intertrain interval; CRT, choice reaction time; PPT, Purdue pegboard test;
AHRS, Auditory Hallucinations Rating Scale; SAPS, Scale for the Assessment of Positive Symptoms;
SMA, supplementary motor area; MRCPs, movement-related cortical potentials; SCL-90, Rating Scale of
Symptom Checklist-90.

23
Chapter 2: Design and Evaluation of a Miniaturized Coil
2.1 Introduction
TMS is a non-invasive neuromodulation technique extensively used in clinical applications and
basic research. During TMS, a coil is placed over the scalp. A brief high-intensity current pulse
passes through the coil producing a time-varying B-field. The B-field further induces an E-field
inside the brain, which modulates brain activities (Barker, 1991; Hallett, 2007). Despite the
widespread use of TMS for treating neurological and neuropsychiatric conditions such as
depression (J. Chen et al., 2013; Levkovitz et al., 2015; McDonald et al., 2011; O’Reardon et al.,
2007), OCD (Carmi et al., 2018, 2019), migraine (Lipton et al., 2010), stroke (Mansur et al., 2005),
epilepsy (W. Sun et al., 2012), schizophrenia (Brunelin et al., 2006), and autism (Enticott et al.,
2012), the underlying mechanisms of its effects remain largely unclear. Animal models are
essential for investigating such underlying mechanisms since they allow more comprehensive
electrophysiological recordings and invasive experiments that are difficult or impractical on
humans. However, most commercial TMS coils are designed for humans. The large geometric
size of human TMS coils will cause non-focal high-intensity stimulation, which are not applicable
to small animal studies (Alekseichuk et al., 2019; Tang et al., 2017; Vahabzadeh-Hagh et al., 2012).
Therefore, there is a strong need for miniaturized TMS coils in neuroscience, neural engineering,
and clinical research communities.
Several miniaturized TMS coils have been developed for rodent studies, among which circular
coil is one of the most popular designs. For example, a circular coil (8 mm outer diameter) with
300 turns of copper wire around an iron core was built to achieve a maximum B-field of 12 mT
(Makowiecki et al., 2014; Rodger et al., 2012). By increasing the windings to 780 turns, it

24
produced a maximum B-field of 120 mT and an E-field of 12.7 V/m on the brain surface at ~1 mm
below the coil (Tang et al., 2016). A similar circular coil (9.2 mm outer diameter) with 70 turns
of copper wire around a soft ferrite core produced a maximum B-field of 180 mT and an E-field
of 2.5 V/m at 2 mm below the coil (Wilson et al., 2018). In another circular coil (5 mm outer
diameter) with 50 turns of copper wire around a tapered iron powder core (Khokhar et al., 2021),
a maximum B-field of 685 mT and an E-field of 15 V/m at the base of the coil were generated.
However, the maximum E-field of circular coils has a circular shape under the coil perimeter thus
is not focal (Fig. 2.1A) (L. G. Cohen et al., 1990; Deng et al., 2013). In several recent studies,
figure-eight coils were built for focal TMS in rodents. For example, a figure-eight coil with 9
turns of copper wire in each wing (19 mm outer diameter) could produce more focal B-field (300
mT as maximum) and E-field (0.55 V/m as maximum) (Parthoens et al., 2014). A more powerful
figure-eight coil (25 mm outer diameter) was designed to use a commercial stimulator to generate
B-field and E-field intensities similar to those of human TMS coils that could elicit unilateral
MEPs in rats, which was not achieved in most rodent studies (Boonzaier et al., 2020). With the
aid of convex optimization for TMS in realistic rat head models (Koponen et al., 2017; Koponen,
Stenroos, et al., 2020), the figure-eight coil could be further optimized with windings extending to
a wide region (116 mm by 84 mm) to achieve minimum energy without sacrificing the stimulation
focality (Nieminen et al., 2022). To date, figure-eight coils in conjunction with specially modified
flat electrodes or probes have been used to assess brain circuitry and function (B. Li et al., 2017;
Murphy et al., 2016). However, the figure-eight coil cannot be easily combined with standard
electrophysiology since it is difficult to implant a straight penetrating electrode under the center of
the coil (where maximum stimulation intensity exists) while keeping the coil close to the brain
surface (Fig. 2.1B). Other alternative designs (Bagherzadeh et al., 2022; Cobos Sánchez et al.,

25
2020; Meng, Jing, et al., 2018; Rastogi et al., 2016) improved the stimulation focality using
analytical models but at the same time generated E-field distributions different from those of
circular or figure-eight coils, which are most commonly used in humans. Consequently, the neural
responses elicited with these coils cannot be directly compared with those in human TMS studies.
To develop a TMS coil that can (1) generate focal B- and E-fields similar to those of human
figure-eight coils, and (2) is compatible with standard electrophysiological recording and
stimulation, we designed, fabricated, characterized, and evaluated a novel C-shaped miniaturized
coil for use in small animals (Jiang et al., 2021, 2022) (Fig. 2.1C). This coil consisted of insulated
copper wires around a C-shaped iron powder core, which allowed convenient implantation of
electrophysiological recording and stimulation electrodes between the two bases of the coil. The
resulting B- and E-fields were simulated with finite element modeling and further characterized
with experimental measurements. The efficacy of this TMS coil in neuromodulation was validated
with electrophysiological recordings of SUAs, SSEPs, and MEPs before and after rTMS in rats
using single electrodes and MEAs. This novel miniaturized coil and the associated experimental
paradigm enabled the combination of TMS, electrophysiological recording, and electrical
stimulation in rat brains. Using this paradigm, we for the first time observed distinct modulatory
effects on SUAs, SSEPs, and MEPs with the same rTMS protocol in anesthetized rats. It provided
a useful tool to investigate the neural responses as well as the underlying neurobiological
mechanisms of TMS in small animal models.

26
Figure 2.1. E-fields simulated with different TMS coils in free space. (A) A circular coil generated a nonfocal maximum E-field under its perimeter. Standard multi-electrode array (MEA) could be implanted to
the location with high E-field without obstruction. (B) A figure-eight coil generated a focal maximum Efield under the center of the coil. However, standard MEA could not be vertically implanted under the
center of the coil unless the coil was far apart or tilted at a very large angle, which might result in weak or
different E-field distribution. (C) A C-shaped coil generated a focal maximum E-field under the center of
the coil. Standard MEA could be implanted to the focal point with the coil or the MEA slightly tilted.

27
2.2 Method
2.2.1 TMS coil design
A miniaturized TMS coil was built by winding 30 turns of insulated 23-gauge copper wire on
a C-shaped magnetic core. The core was made from an iron powder toroid (T60-26, Micrometals,
Anaheim CA, USA). The toroid has an outer diameter of 15.2 mm, an inner diameter of 8.5 mm,
and a height of 6.0 mm. A small gap was cut on the toroid to form the C-shaped geometry. The
angle and shortest distance between the two bases of the C-shaped coil were 150 degrees and 5
mm, respectively. The winding at each base of the coil was rectangular with a dimension of 7 mm
x 4 mm. The core material was composed of ferromagnetic particles coated with organic
compounds to ensure electrical insulation. This insulation makes the core tolerant of thermal aging
and provides high saturation flux density with low loss (Khokhar et al., 2021; Lefebvre et al., 1997;
K. Sun et al., 2020). The permeability of the core material is 75 times that of the air, which greatly
increases the magnetic flux density of the coil. The TMS coil was dip-coated with conformal
silicone coating (422C, MG Chemicals, Canada) to provide extra protection and insulation. To
characterize the electrical properties of the coil, the coil resistance and inductance were measured
via a precision LCR meter (Keysight E4980AL, Santa Rosa, CA, USA) at 1V input voltage from
20 Hz to 100 kHz (201 points in log scale).
2.2.2 TMS circuit design
A custom-made stimulator was designed to generate TMS pulses through the coil (Fig. 2.2). A
DC voltage source (Kungber SPS605 Variable DC Power Supply, Shenzhen, China) was used as
the power supply to drive the circuit. Its voltage could be adjusted to control the amplitude of the

28
TMS pulse. The maximal output of the DC voltage source (60 V) was used for all experiments.
Two aluminum electrolytic capacitors (capacitance: 680 F; voltage: 200 VDC) and one polymer
film capacitor (capacitance: 40 F; voltage rating: 500 VDC) were connected in parallel for
generating the pulses. Since electrolytic capacitors had larger capacitance and resistance compared
to film capacitor, the combination of those capacitors improved performance by providing high
capacitance at low frequencies while providing low resistance at high frequencies. Two N-channel
MOSFETs (IRF540, Vishay, Malvern PA, USA) were used to control the charging and discharging
of the capacitors. When the voltage applied at the gate terminal was below the threshold voltage,
the MOSFET was off, and the capacitors was charged up to the supplied DC voltage. When the
voltage applied at the gate terminal was above the threshold voltage, the flow of current through
the MOSFET led to discharging of capacitors. As a result, a current pulse was generated in the
TMS coil. A Schottky diode (SR560, EIC semiconductor, Bangkok, Thailand) parallelly
connected to the coil was used to protect the MOSFETs from damage. The gate voltage of the
MOSFET was determined by the output of an operational amplifier (OPA548, Texas Instruments,
Dallas TX, USA). The noninverting input terminal of the amplifier was connected to an Agilent
33500B arbitrary waveform generator (Keysight, Santa Rosa, CA, USA), which could be
programmed to generate monophasic waveforms as the stimulator’s input. The inverting input
terminal of the amplifier was connected in feedback to a sense resistor (0.05 Ω) and the source
terminal of the MOSFET. The high forward gain and differential input nature of the amplifier
drove the gate of the MOSFET. When it was above the threshold, the voltage of the input signal
appeared across the sense resistor. The current passing through the sense resistor would appear in
the drain of the MOSFET and the TMS coil. Two identical MOSFET circuits were used and

29
connected in parallel to double the coil current. Therefore, the coil current was measured as the
sum of currents across two sense resistors in the circuit.
To evaluate the performance of the stimulator, Gaussian pulses with different peak amplitudes
(0-10 V) but the same standard deviation (30 s) were generated as input signals to the circuit.
The coil current was measured at each intensity in an incremental order with steps of 0.2 V. In
addition, the coil current was measured when the Gaussian pulses were delivered at different
frequencies (1, 5, 15, and 20 Hz) with the same peak amplitude of 7.5 V and standard deviation of
30 s. To further test the flexibility and real-time capability of the device, arbitrary stimulation
patterns of Gaussian pulses with random peak amplitude (3-8 V) and frequency (1-20 Hz) were
generated via real-time SCPI commands sent to the arbitrary waveform generator. The coil current
was recorded simultaneously.
Figure 2.2. Simplified schematic of the driving circuit of the TMS coil. The stimulation pattern was
generated with a waveform generator as the input to the circuit. The output voltages of two operational
amplifiers controlled the on and off conditions of two MOSFETs in parallel. When the MOSFETs were

30
off, the capacitors were charged up to the voltage of the DC power supply. When the MOSFETs were on,
the capacitors were discharged, and a current pulse was generated in the TMS coil.
2.2.3 Measurements of magnetic and electric fields
To characterize the C-shaped TMS coil (Fig. 2.3A) driven by the stimulator, experimental
measurements of B- and E-fields were performed with a stereotaxic system in a Faraday cage. The
center point between two bases of the coil was defined as the origin of the 3D coordinates. The
x-, y-, and z-component of B-field (,,
) produced by the TMS coil were measured using a
two-loop search coil (radius r=1.15 mm, number of turns N=2) (Fig. 2.3B). The search coil was
placed above the base of the TMS coil and oriented to the direction of each measured component.
Voltage V was induced in the search coil during stimulation and amplified 100 times with a Model
1700 Differential AC Amplifier (A-M systems, Sequim WA, USA). The B-field was sampled
with a 1 mm spatial resolution. The induced voltages along each direction (, ,
) in the
transverse (x-y) plane were captured with an Axon Digidata 1322A acquisition system (Molecular
Devices, Sunnyvale CA, USA). Assuming the B-field was homogeneous over the search coil,
Faraday’s law of induction was used to calculate each component of B-field (Meng, Daugherty, et
al., 2018):
,, = −
1

2
∫ ,,
∙
The corresponding x- and y-component of E-field (, ) were measured using a dipole probe
in saline (Fig. 2.3C). The dipole probe consisted of two twisted insulated wires with a 3 mm
separation at the exposed ends (Mueller et al., 2014; Tofts & Branston, 1991). The TMS coil was
placed underneath a glass beaker that was filled with sodium chloride solution (0.9%). The dipole
probe was placed in the solution and oriented to the direction of each component. A biphasic

31
waveform was measured via the dipole probe during stimulation. It was amplified 1000 times
with a DAM50 differential amplifier (World Precision instruments, Sarasota FL, USA). The Efield was sampled with a 2 mm spatial resolution. The induced voltages along each direction in
the transverse plane were measured with the Axon data acquisition system. Assuming the induced
E-field was approximately constant between the exposed ends of the twisted wires (Tofts &
Branston, 1991), the x- and y-components of E-field were calculated as:
, = −
Δ
∆,
where Δ was the first peak of the biphasic waveform; ∆ and ∆ were known distances of the
exposed ends along x- and y-direction, respectively.
Figure 2.3. B- and E-field measurement. (A) Photograph of the miniaturized C-shaped TMS coil. (B)
Schematic of the set-up for B-field measurement. Three types of search coils were placed above the coil to
measure the B-field in x-, y- and z-directions. (C) Schematic of the set-up for induced E-field measurement.
Two dipole probes were placed above the coil and oriented to measure the E-field in x- and y-directions
within saline.

32
2.2.4 Finite element modeling of magnetic and electric fields
Finite element modeling was conducted using Multiphysics 6.0 (COMSOL, Burlington MA,
USA) to simulate the B- and E-fields induced by the TMS coil. The magnetic core was constructed
as a half circle (-105 to 105 degrees) with a rectangular transection (6 mm x 3 mm). Relative
permeability of the core material was set to 75; the electrical conductivity was set to 0 due to the
dielectric coating of iron powder; the winding was simulated as a single copper conductor with a
diameter of 0.2 mm. The 30-turn helical coil was evenly distributed on the core. In the frequency
domain analysis, the coil current, , was modeled as a sinusoidal current at frequency, ,
with the same peak amplitude, , and the same maximum rate of change at time, = 0
:

|
=0
= 2
As a result, the TMS coil was injected with a 142 A sinusoidal current at 3217 Hz to replicate
the maximum time derivative of 2.87 A/s in measured Gaussian current pulse.
For comparison with the experimental measurements, the B-field was simulated in the air; the
E-field was simulated in saline. A cylinder with a height of 60 mm and a diameter of 60 mm was
created to approximate the saline solution in the beaker. Its isotropic electrical conductivity was
set at 1.45 S/m (Sauerheber & Heinz, 2015). The simulated TMS coil was positioned underneath
the cylinder as in the experimental setup. B- and E-fields were calculated using COMSOL
Magnetic and Electric Fields (mef) interface. Complete meshes of B- and E-fields consisted of
2043805 and 492000 elements, respectively.
To estimate the B- and E-fields induced in the brain, a 3D rat brain model was built based on
an existing model (M. Pohl et al., 2013), which was constructed from 240 cross-sectional magnetic
resonance imaging (MRI) slices. Electrical conductivity of the brain was set to 0.106 S/m

33
corresponding to gray matter (Mueller et al., 2014; Slopsema et al., 2018; Tang et al., 2021). The
TMS coil was positioned 1.5 mm above the right hemisphere and rotated 15 degrees as in the
experimental setup. The complete mesh consisted of 188150 elements.
2.2.5 Animal Surgery
Female Sprague-Dawley rats (n=32, 11-12 weeks, 220-250 g) were used in this study; eight
animals were used for single-unit recordings from the primary somatosensory cortex (S1); eight
animals were used for single-unit recordings from the primary motor cortex (M1); eight animals
were used for SSEP recordings while another eight animals were used for MEP recordings.
Animals were anesthetized with ketamine-xylazine (K, 75 mg/kg; X, 10 mg/kg; intraperitoneal
administration) following a rapid inhalational induction with 3-4% isoflurane. Anesthesia was
maintained with additional boluses of 36 mg/kg ketamine. The animal was mounted on a
stereotaxic frame with Parafilm-wrapped ear bars and nose cone in a Faraday cage. Neural
recordings were conducted when the anesthesia state was stable. Physiological status of animals
(pedal reflex, body temperature, breathing rate, etc.) was monitored throughout the experiments.
A craniotomy was performed to expose the right sensorimotor cortex. One burr hole was incised
to attach the ground wire over the occipital bone.
2.2.6 Single-unit recordings
To conduct single-unit recordings, a MEA (Neuronexus A8x8-Edge-5mm-50-150-177, Ann
Arbor MI, USA) was implanted in the S1 (AP: 0.00 mm; ML: 3.80 mm) or the M1 (AP: 2.40 mm;

34
ML: 3.00 mm) regions (Fig. 2.4A). SUAs were recorded and sorted via OmniPlex neural recording
system (Plexon, Dallas TX, USA).
By connecting the recording system with implanted electrodes, wire loops between electrodes
were formed (B. Li et al., 2017; Mueller et al., 2014). During TMS, the changing B-field would
induce current in the loop. The injected current via electrodes might evoke neural responses at
high amplitude. To estimate the amount of TMS-induced current in the recording electrode, the
induced electromotive force around the wire loop formed with the recording assembly and coupled
to the maximal B-field during TMS without any offset was calculated (Mueller et al., 2014):
|| = |

| = |

−
1
2

| =

−
1
2 × 473 × 500 2
30
= 4.8
where is the induced electromotive force; is the magnetic flux through the loop; is the
maximal B-field during TMS; S is the area of the loop formed by the recording electrode and
reference electrode; is the standard deviation of the Gaussian-shaped pulse. The induced current
in the loop could be converted using the input impedance of the headstage (40 MΩ) yielding 120
nA. It is unlikely to induce significant neural responses via microsimulation at this amplitude
(Boinagrov et al., 2012; Stoney et al., 1968), and the induced current could be even smaller due to
the offset of the wire loop. Similarly, the induced current in the ground electrode could be
converted using the impedance of the rat body with a larger loop area. However, the injected
current density via ground electrode could be negligible due to a large contact area.

35
Figure 2.4. TMS combined with electrophysiological recording and stimulation. (A) Experimental set-up
for single-unit recordings. Single-unit activities were recorded with the MEA implanted in the S1 and the
M1 regions. The coil was placed above the sensorimotor cortex over the craniotomy window. The location
of MEA placement was verified with an implantation lesion (S1 layer 4-5 and M1 layer 5). (B)
Experimental set-up for SSEP recording. SSEPs elicited by electrical stimulation in the forelimb were
recorded with the microelectrode implanted in the S1 region. Microelectrode location was verified with an
electrolytic lesion (S1 layer 2-3). (C) Experimental set-up for MEP recording. MEPs elicited by electrical
stimulation in the M1 were recorded with needle electrodes inserted in the forelimb. Stimulating
microelectrode location was verified with an electrolytic lesion (M1 layer 5).
2.2.7 Somatosensory evoked potentials
To record SSEPs, a single Platinum-Iridium microelectrode (PI2PT30.5A5, Microprobes,
Gaithersburg MD, USA) was implanted in the S1 (AP: -0.25 mm; ML: 3.80 mm) (Fig. 2.4B). The
microelectrode and ground wire were connected to the DAM50 differential amplifier (gain: 1000).
Two subdermal needle electrodes (Natus, Middleton WI, USA) were inserted into the contralateral
forelimb to stimulate the median nerve (Bazley et al., 2012). Another needle electrode was inserted
into the tail as the ground for a STG1004 stimulator (Multi Channel Systems, Reutlingen BW,
Germany). A biphasic square current pulse (pulse amplitude: 0.8 mA; pulse width: 0.1 ms) was
applied every 5 seconds to evoke SSEPs. All signals were recorded with the Axon data acquisition
system.

36
2.2.8 Motor evoked potentials
To record MEPs, a single microelectrode was implanted in the M1 (AP: 1.25 mm; ML: 3.00
mm) (Fig. 2.4C). The microelectrode and ground wire were connected to the STG1004 stimulator.
Two biphasic current pulses (pulse amplitude: 0.8 mA; pulse width: 0.5 ms; inter-pulse interval:
2 ms) were applied every 5 seconds to evoke MEPs. To record muscle activities, two subdermal
needle electrodes were inserted into the biceps brachii muscle and the finger pad of the
contralateral forelimb as the recording and reference electrode, respectively (Bazley et al., 2012;
B. Li et al., 2017). A third needle electrode was inserted to the tail as the ground. MEPs were
recorded with the DAM50 differential amplifier (gain: 1000) and Axon data acquisition system.
2.2.9 Repetitive transcranial magnetic stimulation
rTMS in all animal experiments were delivered as Gaussian input pulses with a peak amplitude
of 7.5 V and standard deviation of 30 s at 10 Hz. During single-unit or evoked potential
recordings, the TMS coil was placed ~1.5mm above the cortex. It was rotated 15 degrees to make
the tip of the electrode align with the coil center. Following 5-min baseline recordings,
subthreshold rTMS (3 min, 10 Hz) or control condition (rTMS turned off) was delivered to the
sensorimotor cortex (n=5, rTMS groups; n=3, control groups). To evaluate the time-course of
rTMS effects, neural signals were monitored for 15 minutes after stimulation.
2.2.10 Temperature measurement
To evaluate the heating effects produced by the coil, the temperature of the coil and the brain
surface before and after rTMS (3 min, 10 Hz) was measured three times using a thermometer

37
(Famidoc, Guangdong, China) in three animals. The temperature of the coil was measured to be
23.8±0.1 °C before rTMS (10 Hz) and increased to 46.7± 0.3 °C after 3 minutes of rTMS (n=3).
Meanwhile, the temperature of rat brain surface was measured to be 35.8±0.2 °C before rTMS
and slightly increased to 36.2±0.2°C after 3 minutes of rTMS (n=3).
2.2.11 Histology
Histology was performed to verify the implantation site in the brain. For evoked potential
recordings, an electrolytic lesion was created by delivering a constant current (amplitude: 0.3 mA;
duration: 10 s) to the implanted microelectrode. For single-unit recordings, no electrolytic lesions
were created since the MEA could not be used to deliver a strong enough current. Perfusion was
performed using 10% formalin solution (Sigma-Aldrich, Burlington MA, USA) via the vascular
system. After the brain was extracted and dehydrated with 18% sucrose solution, brain slices (50
μm) were cut on a coronal plane with a Cryostat (Leica, Buffalo Grove IL, USA). Nissl staining
was applied on the brain slices to visualize the neuron populations and implantation lesions.
2.2.12 Data processing and statistical analyses
All neural recordings were imported and processed in MATLAB (MathWorks, Natick MA,
USA). The baseline of the neural recordings was defined as the 5-minute recording when the
animal was under a stable anesthesia state prior to rTMS. For single-unit recordings, a 250 Hz
highpass filter was first applied to the data. Spike sorting was performed in the Plexon offline
sorter (Plexon, Dallas TX, USA) to extract the timestamps of neuronal firing. The mean firing
rate of each neuron was calculated and averaged with a one-minute bin size. For evoked potential

38
recordings, timestamps of current stimulus artifacts were captured by setting a threshold using
MATLAB function ‘findpeaks’. The DAM50 amplifier was set with a bandpass filter of 0.1 Hz10 kHz, and no other filter was applied to the evoked potentials. After the timestamps of each
stimulus artifact were obtained, a 100 ms window following the artifact was extracted to isolate
the evoked potentials. The individual peak-to-peak amplitude was calculated as the difference
between the highest and the lowest value in each evoked potential. The individual latency was
calculated as the duration between the stimulus artifact and the first negative peak of each evoked
potential. The peak-to-peak amplitude and latency of each evoked potential were further averaged
every minute (12 consecutive sweeps).
All values were normalized with their corresponding 5-min baseline recordings and reported as
the mean ± standard error of the mean (SEM). Two-tailed t-tests with a significance level of 0.01
were conducted to compare the normalized neural signals before and after rTMS or control
condition.

39
2.3 Results
2.3.1 Characterization of TMS coil
Gaussian pulses (peak amplitude: 7.5 V; standard deviation : 30 s) were generated as input
signals to the circuit (Fig. 2.5A). The corresponding coil current was measured across the sense
resistors (Fig. 2.5B). The current had the same shape as the input pulse and a peak amplitude
(max ) at 142 A. The current as a function of time could be approximated by the following
equation:
() = e
−
1
2
(

)
2
The induced E-field waveform was proportional to the first derivative of the coil current. The
resulting biphasic waveform was measured via a dipole probe in saline (Fig. 2.5C). The induced
E-field waveform as a function of time could be approximated by the following equation:
() =

−
1
2
t ∙ e
−
1
2
(

)
2
where was measured to be 5.2 V/m when the dipole probe was placed ~3.5 mm from the
center of the coil. The input-output curve demonstrates the relationship between the input signal
and the output (coil) current (Fig. 2.5D). The response of the circuit was linear when the input
signal was within the range of 3-8 V. The lower limit (3 V) was determined by the gate-source
threshold voltage of the MOSFET (2-4 V). However, saturation occurred when the input signal
was greater than 8 V. In addition, the output current was affected by the rTMS frequency (Fig.
2.5E). The coil current decreased as the frequency increased. At 20 Hz, the coil current dropped
to 95% of the value at 1 Hz. Furthermore, the resistance and inductance of the coil were measured
via a precision LCR meter at 1 V input voltage from 20 Hz to 100 kHz. The resistance remained

40
at 0.1 Ω when the frequency was lower than 10 kHz (Fig. 2.5F). This limit was much higher than
the frequency of the input pulse (3217 Hz); therefore, the skin and proximity effects, which
reduced the effective area of the copper wire and increased the overall resistance (Ravazzani et al.,
2002), were not obvious during stimulation. The inductance of the coil was measured to be
constant at 11.7 H when the input frequency is between 100 Hz to 10 kHz, which was similar to
the simulated inductance of 13.4 H in the finite element modeling. However, the inductance
might vary with the amplitude of the coil current due to the use of the magnetic core which could
introduce nonlinearity at high amplitude (RamRakhyani & Lazzi, 2014). The TMS coil was able
to deliver arbitrary stimulation patterns (Fig. 2.5G) via the custom-made stimulator. The pulse
amplitude and frequency were precisely controlled through real-time SCPI commands sent to the
arbitrary waveform generator.

41
Figure 2.5. Electrical properties of the TMS coil. (A) A Gaussian voltage pulse generated by a waveform
generator was used as the input signal. (B) Coil current. (C) E-field waveform measured in saline. (D)
Output (coil) current as a function of input signals. (E) Output current relative to current measured at 1 Hz
as a function of rTMS frequency. (F) Coil resistance measured via a precision LCR meter. (G)
Representative arbitrary stimulation pulse patterns generated by the coil. Top: stimulation pattern with
multi-level amplitudes and frequencies. Bottom: stimulation pattern with random amplitudes and intervals.
2.3.2 Comparison of simulated and measured field distributions
B-field induced by the coil in the air was measured on benchtop and simulated with finite
element modeling method. The measured and simulated B-fields were highly consistent (Fig.

42
2.6A). At the center of the TMS coil, the peak strength of x-component of B-field was measured
and simulated to be 227 mT and 247 mT, respectively. Both measurement and simulation showed
two smaller peaks with opposite polarity at the bases of the coil. The measurement showed a
slightly larger peak at the right base (south pole) compared with the left base (north pole). This
deviance in magnitude might be due to the slight asymmetry of the handcrafted magnetic core.
The y-component of B-field had four peaks with different polarity, which were apparent in both
measurement and simulation. Among the four peaks, each pair of the diagonal peaks shared the
same polarity. The combination of x- and y-component was shown in the quiver plots. The
measurement and simulation showed near identical B-field distributions. The minimum points
existed at the base of the coil. The main B-field vectors pointed from the north pole to the south
pole. The z-component was the dominant component of B-field in the transverse plane. The
simulation showed that the north pole and the south poles had peak strengths of 446 mT and -455
mT, respectively. The measurement showed that the north pole and south pole had peak strengths
of 406 mT and -473 mT, respectively.
E-field induced by the coil in saline was further measured and simulated. The dipole probe was
placed ~4.5 mm from the base of the coil considering the glass thickness (~3.5 mm). The measured
and simulated E-fields were also highly consistent (Fig. 2.6B). The x-component of E-field had
four peaks with different polarity in both measurement and simulation. Similar to the y-component
of B-field, each pair of the diagonal peaks shared the same polarity, and each pair of the adjacent
peaks had opposite polarity. At the center of the TMS coil, the peak strength of y-component of
E-field was measured and simulated to be 3.7 V/m and 3.9 V/m, respectively. Both measurement
and simulation showed two smaller peaks with opposite polarity at the base of the coil. The
combination of x- and y-component was shown in the quiver plots. The field vectors followed the

43
direction of induced current produced by the changing B-field. The two ring-shaped fields under
the two bases of the coil reinforced each other along y-direction and led to a stronger and more
focal maximum E-field strength at the center. When the dipole probe was placed ~3.5 mm from
the center of the coil, a maximum strength of 5.2 V/m was achieved. To compare the measurement
and simulation along z-direction, the dipole probe was moved vertically at the center of the coil.
Results showed high degree of consistency in the decay rate for both x- and y-components.

44
Figure 2.6. B- and E-field distributions from measurement and simulation. (A) Measured (top) and
simulated (bottom) B-field profiles in the transverse plane. The x-, y-, and z-components of B-field and
their combination of x- and y-components of B-field were in the air when the TMS coil was placed ~0.5
mm away. (B) Measured (top) and simulated (bottom) E-field profiles in the transverse plane. The x- and
y-components of B-field and their combination were in saline when the TMS coil was placed ~4.5 mm
away. In addition, normalized measured and simulated x-component (top) and y-component (bottom) of
E-field as a function of depth along the z-direction showed consistent decay rates.

45
2.3.3 Magnetic and electric field distribution in the rat brain
B- and E-fields in the rat brain were further simulated with the finite element modeling method
(Fig. 2.7). In the simulation, the C-shaped coil was placed above the sensorimotor cortex to match
the placement in real animal experiments (Fig. 2.7A). The peaks of B-field were underneath the
two bases of the coil (Fig. 2.7B). The maximum strength was 460 mT on the surface of cortex.
The sagittal view revealed that B-field decayed rapidly as the distance increases. It reduced to 100
mT at ~3.8 mm from the surface. Simulated E-field distribution in the brain indicated that the
focal point was under the center of the TMS coil (Fig. 2.7C). The maximum strength of E-field
was estimated to be 7.2 V/m on the surface of cortex. The coil was able to generate an
approximately 6-by-8 mm rectangular area with E-field strength above 3.6 V/m when it was tilted
15 degrees. It could focally deliver TMS to the M1, secondary motor cortex (M2), and S1. Similar
to B-field, the induced E-field decayed rapidly as the distance increased. It reduced to 1 V/m at
~4.4 mm from the surface.

46
Figure 2.7. Simulated B- and E-field distributions in the rat brain. (A) Superior (left) and sagittal (right)
views of coil placement relative to a 3D rat brain. (B) B-field distribution on the surface (left) and sagittal
section (right) of the brain. (C) E-field distribution on the surface (left) and sagittal section (right) of the
brain.
2.3.4 rTMS increases firing rates of primary somatosensory and motor cortical neurons
To investigate the effect of subthreshold rTMS (3 min, 10 Hz) on cortical neurons,
spontaneously occurring electrical activities from the S1 and M1 were recorded with the 64-
channel MEA. Representative wideband and high-pass filtered (250 Hz) neural signals in one
animal demonstrate the changes in single-unit recordings before and after rTMS (Fig. 2.8A). To
further quantify the changes, spike sorting was performed to extract the SUA, and the firing rate

47
of each neuron was calculated. Results showed that the firing rates of S1 (Fig. 2.8B) and M1
neurons (Fig. 2.8C) in 10 animals (n=5 per rTMS group) both increased after rTMS compared with
the firing rates of baseline recordings. The normalized mean firing rates (S1 neurons: n=163 per
rTMS group, n=88 per control group; M1 neurons: n=149 per rTMS group, n=101 per control
group) were further averaged for every minute. The mean firing rates of S1 and M1 neurons
significantly increased to 154±5% (t=6.93, p=0.002) and 160±9% (t=7.20, p=0.002) of the
baseline level immediately after rTMS, respectively. They gradually decreased to 121±10%
(t=6.37, p=0.003) and 113±6% (p0.01) over 15 minutes. Two-tailed t-test reveals no significant
changes after control condition (p0.01).

48
Figure 2.8. rTMS facilitated firings of S1 and M1 neurons. (A) Representative wideband and 250 Hz
highpass filtered signals before and after rTMS, single-unit waveforms, and PCA clusters from one animal.
(B) The firing pattern of S1 neurons before and after rTMS (n=163) or control condition (n=88). (C) The
firing pattern of M1 neurons before and after rTMS (n=149) or control condition (n=101). In (B) and (C),
Top: firing rate histogram for rTMS group (bin size: 2 s); Bottom: mean firing rate for both rTMS and
control groups (bin size: 1 min). Comparisons of mean firing rates before and after rTMS or control
condition were performed with two-tailed t-tests (*p < 0.01). Values were reported as the mean ± SEM.

49
2.3.5 rTMS suppresses somatosensory evoked potentials
To evaluate the effect of rTMS with the same stimulation parameters (3 min, 10 Hz) on the
ascending sensory pathways, SSEPs elicited by electrical stimulation of the forelimb muscle were
recorded from the contralateral S1 before and after rTMS. Representative traces of one animal
demonstrate the suppression of SSEP following rTMS compared with the mean trace of baseline
recordings (Fig. 2.9A). The normalized SSEP amplitude and latency of each group (n=5 per rTMS
group; n=3 per control group) were further averaged for every minute. The SSEP amplitude was
significantly suppressed to 74±4% (t=-11.9888, p=0.0003) and 82±5% (t=-8.4367, p=0.0011) of
the baseline level during the first and second minute after rTMS, respectively (Fig. 2.9B). After
~5 minutes, the SSEP amplitude returned to the baseline level. Two-tailed t-test reveals no
significant changes after control condition (p0.01). The SSEP latency was obtained by
calculating the duration between the stimulus artifact and the first negative peak of each evoked
potential. Figure 2.9C demonstrates the average of normalized SSEP latency at each timepoint. It
indicates that there is no significant change before and after rTMS and control condition (p0.01).

50
Figure 2.9. rTMS suppressed SSEPs. (A) Representative raw SSEP traces (gray) at 1-15 minutes after
rTMS from one animal. Red: averaged trace; Blue: baseline trace. (B) Mean SSEP amplitude before and
after rTMS (n=5) and control condition (n=3). (C) Mean SSEP latency before and after rTMS (n=5) and
control condition (n=3). In (B) and (C), comparisons of the amplitude or latency before and after rTMS or
control condition were performed with two-tailed t-tests (*p < 0.01). Values were reported as the mean ±
SEM.

51
2.3.6 rTMS facilitates motor evoked potentials
To further compare the effects of rTMS (3 min, 10 Hz) on different pathways, MEPs
representing the integrity and excitability of descending motor pathways were elicited by
intracortical electrical stimulation of the M1 and recorded from the contralateral forelimb muscle
before and after rTMS. Different from SSEP, representative traces demonstrate the facilitation of
MEP following rTMS compared with the mean trace of baseline recordings (Fig. 2.10A). The
normalized MEP amplitude and latency of each group (n=5 per rTMS group; n=3 per control group)
were further averaged for every minute. The MEP amplitude significantly increased to 136±9%
(t=6.5420, p=0.0028), 137±14% (t=6.7441, p=0.0025), and 126±11% (t=4.6522, p=0.0096) of
the baseline level during the first, third, and forth minute after rTMS, respectively (Fig. 2.10B).
After ~5 minutes, the MEP amplitude returned to the baseline level. Two-tailed t-test reveals no
significant changes after control condition (p0.01). The MEP latency was obtained by calculating
the duration between the stimulus artifact and the first negative peak of each evoked potential.
Figure 2.10C demonstrates the average of normalized MEP latency at each timepoint. Similar to
SSEPs, there is no significant change in MEP latency before and after rTMS and control condition
(p0.01).

52

Figure 2.10. rTMS facilitated MEPs. (A) Representative raw MEP traces (gray) at 1-15 minutes after
rTMS in a representative animal. Red: averaged trace; Blue: baseline trace. (B) Mean MEP amplitude
before and after rTMS (n=5) and control condition (n=3). (C) Mean MEP latency before and after rTMS
(n=5) and control condition (n=3). In (B) and (C), comparisons of the amplitude or latency before and after
rTMS or control condition were performed with two-tailed t-tests (*p < 0.01). Values were reported as the
mean ± SEM.

53
2.4 Discussion
In this study, we developed and tested a novel miniaturized coil and its driving circuit for rodent
TMS studies. As we knew, different TMS coil geometries resulted in different B- and E-field
distributions, and further led to distinguishing effects of TMS on neural activities and behaviors.
Our coil was designed with the main aims to (1) scale down the geometric size to generate focal
B- and E-fields similar to those of human TMS coils, and (2) enable simultaneous
electrophysiological recording and stimulation at its focal point. The resulting design consisted of
a C-shaped core with two adjacent but separate coil windings at the two bases, which carried
currents in opposite directions. Like the conventional figure-eight coils, the opposite currents
induced a focal E-field under the center of the coil. Different from the figure-eight coils, there was
a gap between the two bases of the coil, which allowed convenient placements of electrodes.
Our coil had multiple advantages over existing ones. First, the use of magnetic core reduced
the flux leakage and resulted in a more focal B-field (Spaldin, 2010). In addition, the iron powder
core was composed of ferromagnetic particles which were coated with organic compounds to
ensure electrical insulation. It eliminated the eddy current in the core and largely reduced heat loss
(Lefebvre et al., 1997). Although its permeability was lower than that of iron or ferrite core, it had
higher saturation flux density (Khokhar et al., 2021; K. Sun et al., 2020), which produced a
significantly higher B-field intensity (473 mT measured at ~0.5 mm) compared to most of the
existing small coils (Makowiecki et al., 2014; Parthoens et al., 2014; Rodger et al., 2012; Tang et
al., 2016; Wilson et al., 2018). Second, our TMS coil contained less turns (30 turns) of copper
wires. Higher winding density would lead to higher coil inductance, which in turn reduced the
current in the coil (Wilson et al., 2018). Moreover, it would cause undesired heating which might
result in irreversible damage on the coil as well as the biological tissue (Rossi et al., 2009). Third,

54
as we stated above, our coil could be used in conjunction with standard electrophysiological
recording and stimulation due to the air gap between its bases. The 5 mm gap allowed any rotation
or placement of the MEA (1.4 mm width at the tip and 3.3 mm width at the end) used in this study.
Additionally, a 150-degree angle between two bases of the coil fitted the curvature of the rat head
and made the coil close to the brain surface. The coil was able to generate a focal E-field at the
center of the base and at the same time allowed implantation of electrodes at the same location.
Although the resulting E-field intensity (5.2 V/m measured at ~3.5 mm in saline) was much smaller
compared to those of human TMS coils (of the order of 100 V/m (Koponen, Stenroos, et al., 2020;
Smith & Peterchev, 2018)), it was higher than many of the existing miniaturized circular
(Makowiecki et al., 2014; Rodger et al., 2012; Wilson et al., 2018) and figure-eight (Parthoens et
al., 2014) coils customized for rodents. Fourth, our TMS coil was able to deliver arbitrary
stimulation pulse patterns with different intervals and amplitudes. These features are essential for
optimizing stimulation parameters and further building a closed-loop TMS system to achieve
precise neuromodulation.
We saw this combination of TMS, electrophysiological recording, and electrical stimulation in
small animal brain as the most important feature of this miniaturized C-shape coil. As we knew,
SUAs provided the most precise information of neuronal activities at the single-neuron level. The
effects of rTMS on SUAs were highly informative about its effects on neural circuits, brain
functions, and behaviors. In this study, we demonstrated that brief subthreshold rTMS could
significantly facilitate the firings of S1 and M1 neurons. Previous studies showed that rTMS at 10
Hz decreased the GABAergic synaptic strength (Lenz et al., 2016) and increased glutamatergic
synaptic strength (Lenz et al., 2015; Vlachos et al., 2012) in vitro. These two forms of

55
modifications of synaptic strengths might account for the increasing firing rates observed after
rTMS in this study.
MEP elicited by electrical stimulation or single-pulse TMS over the M1 was one of the hallmark
measures for the quality and integrity of the descending motor pathways (Bestmann & Krakauer,
2015; Rothwell, 1997). It was used to assess the effects of rTMS on the corticospinal tract
excitability in both human and animal studies (Y.-Z. Huang et al., 2005; Klomjai et al., 2015;
Maeda et al., 2000; Muller et al., 2014; Tang et al., 2016; Tsuji & Rothwell, 2002). In this work,
we used MEPs elicited by electrical stimulation of layer 5 neurons in the M1 to evaluate the effects
of subthreshold rTMS generated by our coil. Results showed that MEP amplitude significantly
increased after rTMS. This increase of MEP amplitude was consistent with previous studies
(Maeda et al., 2000; Tang et al., 2016). The change of MEP amplitude provided indirect evidence
on the neuronal mechanism of rTMS. It was likely that these changes were due to a mixture of
changes on the intrinsic excitability of neurons and synaptic connectivity in the activated pathways
(Matheson et al., 2016; Mozzachiodi & Byrne, 2010). For example, the effects of rTMS on SUAs
might contribute to its effects on MEPs; synaptic strengths might be modified between these
neurons. In addition, MEP latency was another indicator of the corticospinal tract excitability,
which reflected the neural conduction rate from the descending impulse to the target muscle (G.
Huang & Mouraux, 2015). Our result showed that the change in MEP latency after rTMS was less
prominent.
In general, high frequency (>5 Hz) rTMS had a facilitatory effect on cortical excitability lasting
for seconds to minutes, whereas low frequency (≤1 Hz) rTMS typically had an inhibitory effect
(Hoogendam et al., 2010; Matheson et al., 2016). Interestingly, we observed a facilitatory effect
of rTMS on SUAs and MEPs but an inhibitory effect on SSEPs with the same rTMS parameters

56
(3 min, 10 Hz). The SSEP was elicited by electrical stimulation over the median nerve and
recorded form the corresponding somatosensory cortex. It reflected the function and integrity of
ascending sensory pathways (Fehlings et al., 1988; Strahm et al., 2003). The SSEP was often used
to evaluate the effects of rTMS in patients with migraine (Coppola et al., 2012; de Tommaso et al.,
2014; Kalita et al., 2017) or stroke (Sohn et al., 2014; Xie et al., 2015). Our results showed that
SSEP amplitude significantly decreased after rTMS. Meanwhile, SSEP latency, which represented
the neural conduction rate from the ascending impulse to the target cortex, had less prominent
changes after rTMS. Previous studies showed improvement in habituation of SSEP in migraineurs
following 10 Hz rTMS treatment (Coppola et al., 2012; de Tommaso et al., 2014; Kalita et al.,
2017). It suggested that the increase of plasma beta-endorphin level (Kalita et al., 2017; Misra et
al., 2013) and the excitation of GABAergic neurons (Coppola et al., 2012; Kalita et al., 2017)
might account for the changes in sensitization and habituation. However, the inhibitory effect of
rTMS on SSEP amplitude in healthy anesthetized subjects has not been studied yet. The
neurobiological mechanisms of the effect of rTMS on SSEP require further investigations.
One main challenge of designing miniaturized coils was caused by the trade-off between coil
diameter and E-field strength (D. Cohen & Cuffin, 1991). To make the coil size suitable for rodent,
extremely high current was required to achieve the same stimulation strength as in human TMS
coil (Bagherzadeh et al., 2022; Khokhar et al., 2021), which might cause excessive heating,
electromagnetic stress, and other technical difficulties. Given this physical constraint, further
improvements could still be made to the current design. For example, the size of the gap largely
influenced the magnetic flux in a gapped toroid (Christiansen et al., 2017). Optimization of coil
performance could be made via a finite element model. We could further improve the focality of
the E-field by optimizing the distance and angle of the gap of the C-shaped coil. In addition, our

57
current DC voltage source had a maximal output of 60 V and might not provide sufficient power
for higher rTMS frequency to prevent the drop of the pulse amplitude. We could modify the
driving circuit and use a more powerful DC power supply to increase the coil current. Furthermore,
it was noted that the temperature of the coil increased ~23 degrees after rTMS (3 min, 10 Hz).
Since the resistance of the coil remained low (0.1 Ω) at the frequency of the input pulse, the rise
of temperature was likely caused by the core losses. The electrical properties of the coil could
change due to the increased temperature of the windings, which might partly explain the drop of
coil current at higher rTMS frequency. A temperature control or cooling system might be required
when the current is further increased, since heating of the coil would potentially lead to a local
brain temperature change that might influence neural excitability (Burek et al., 2019; Shibasaki et
al., 2007, 2015). Moreover, the coil generated light vibrations and sounds during stimulation. The
sound could potentially affect the auditory sensory pathways (Koponen, Goetz, et al., 2020;
Nikouline et al., 1999). However, the ears of the animal were protected and plugged with Parafilmwrapped ear bars throughout the experiments under anesthesia, hence the sound should have
negligible effects on the auditory sensory system. It was worth noting that the control condition
(rTMS turned off) in this study was different from the sham condition for conventional figureeight coils. In sham condition, the current of the two circular windings flowed in the same
direction to cancel the E-field right under the center of the coil. Hence, sham TMS could reproduce
the vibration, sound, and superficial E-field on the head but not cause direct neural effects in the
cortex as the active TMS (Duecker & Sack, 2015; Memarian Sorkhabi & Denison, 2022;
Ruohonen et al., 2000; Zhi-De Deng & Peterchev, 2011). To realize an appropriate sham condition
for our coil, modifications could be made by dividing the 30 windings into two groups of 15
windings. Each group of windings could be driven by the circuit separating from the parallel

58
system with opposite current directions. Besides, Gaussian pulses were used as the input signal to
the stimulator. Different from the conventional monophasic pulse, the Gaussian pulse had a more
symmetric waveform similar to the conventional half-sine pulse (Sommer et al., 2006). The effect
of TMS with different TMS parameters, including pulse shape, frequency, duration, and intensity,
remained to be further studied.
It is widely accepted that the neurobiological effects of TMS are mediated through the E-field
induced in the brain (Aberra et al., 2020; Hallett, 2007; B. Wang et al., 2018). The induced Efield is not only determined by the TMS coil, but also influenced by the geometry and conductivity
of the head (Koponen, Stenroos, et al., 2020; Stenroos & Koponen, 2019). In this study, due to a
large craniotomy performed on the rat head, a brain-only model was used to simulate the E-field
distribution for simplicity. The resulting E-field was less accurate than a full head model, since
tissue boundaries with a contrast in conductivity would cause the accumulation of surface charges
whose secondary E-field significantly affected the primary E-field (Ruohonen, 2003). Further
improvements of the E-field simulation could be made using a realistic head model with all
experimental conditions. To be noted, the main purpose of the E- and B-field measurements in
this study was to compare with the simulated results under identical conditions to validate the finite
element model. The measured E-field in saline differed from the E-field induced in free space or
the brain due to the secondary E-field created by the accumulation of surface charges at the
boundaries of the beaker. Besides, the glass of the beaker prevented measurement of E-field closer
to the coil. To have a more accurate measurement of the induced E-field, a saline-free method
(Nieminen et al., 2015) would be considered for future experiments.
In summary, we designed, fabricated, and validated a novel miniaturized C-shaped TMS coil
and tested it in rat experiments. We reported the methodology of developing and characterizing

59
the miniaturized TMS coil and its stimulator design, experimental measurement, and simulation
of field distributions. The efficacy of this coil in neuromodulation was validated with
electrophysiological recordings of SUAs, SSEPs, and MEPs in rats. This miniaturized coil and
the associated experimental paradigm enabled the combination of TMS, electrophysiological
recording, and electrical stimulation in rat brains. It provided a powerful tool to investigate the
neural responses and underlying mechanisms of TMS in small animal models. Using this
paradigm, we for the first time observed distinct modulatory effects on SUAs, SSEPs, and MEPs
with the same rTMS protocol in anesthetized rats. These results suggested that multiple
neurobiological mechanisms in the sensorimotor pathways were differentially modulated by rTMS.
In the future, we will extend these studies to awake and behaving animals to further investigate
the neurobiological mechanisms of TMS and make the connections between electrophysiological
activities to behaviors and cognitive functions.

60
Chapter 3: Subthreshold rTMS Suppresses Ketamine-Induced Poly
Population Spikes in Rat Sensorimotor Cortex
3.1 Introduction
Cortical oscillations within or across multiple brain regions play fundamental roles in sensory
(Buzsaki & Draguhn, 2004; Engel & Singer, 2001; Fries et al., 2001), motor (Davis et al., 2012;
Farmer, 1998; van Wijk et al., 2012), and memory functions (Lisman, 2010; Nyhus & Curran,
2010). Prominent rhythmic oscillations have been classically divided into delta (1-4 Hz), theta (4-
8 HZ), alpha (8-12 Hz), beta (12-35 Hz), and gamma (35-100 Hz) bands. Since endogenous
cortical oscillations are often disrupted in neurological and neuropsychiatric conditions, they are
often used as biomarkers in diagnosis and treatment of these diseases (Popovych & Tass, 2014;
Rabiller et al., 2015).
Cortical oscillations can be altered by neuromodulation and pharmacological manipulations.
For example, in rTMS, a non-invasive neuromodulation procedure, an external TMS coil is used
to induce a changing electromagnetic field to activate or inhibit the nervous system (Barker, 1991;
Hallett, 2007). Although the underlying mechanisms remains largely unknown, it has been
reported that rTMS can effectively modulate cortical oscillations in different frequency bands with
different stimulation parameters (Ding et al., 2014; Paus et al., 2001; J. Yang et al., 2021;
Zmeykina et al., 2020). Ketamine, a medication originally used for anesthesia and analgesia
(Mazzeffi et al., 2015; McCarthy et al., 1965; Meyer & Fish, 2008; Vadivelu et al., 2016), has
drawn much attention recently for its use as an antidepressant and a psychedelic agent (Andrade,
2017; Ballard & Zarate, 2020; Krystal et al., 1994, 2019; Steeds et al., 2015). It has been shown
that ketamine decreases powers of neural oscillations in low frequency (delta, theta, and alpha)

61
bands while increasing power in high frequency (gamma) band at both anesthetic and
subanesthetic doses(Chauvette et al., 2011; Hong et al., 2010).
Due to their overlapping effects, rTMS and subanesthetic ketamine have been combined to treat
neuropsychiatric disorders in humans and shown synergetic effects (Best & Griffin, 2015; Davila
et al., 2021; Pradhan & Rossi, 2020). However, the interaction of rTMS and anesthetic ketamine
on cortical activities including cortical oscillations and neuronal spike firing has not been explored
yet. To understand the neurobiological basis of the effects of rTMS and ketamine as well as their
interactions on cortical oscillations, in this study, we developed and applied a rodent model that
enabled simultaneous rTMS treatment, pharmacological manipulations, and invasive
electrophysiological recordings, which is difficult or impossible in human studies.
Specifically, a miniaturized TMS coil was designed, fabricated, and characterized for rodent
brain stimulation. This coil can deliver focal subthreshold TMS to the S1 and M1 in rats. MEAs
were also implanted in the S1 and M1 to record both LFPs and SUAs. Using this rodent model,
we discovered a novel form of synchronized activities, i.e., poly population spikes (PPS), as the
biomarker of ketamine in LFPs. Such activities could be highly reliably induced by ketamine.
More intriguingly, we found that brief (3-min duration) subthreshold rTMS can effectively and
reversibly suppress PPS while increasing the firing rates of spontaneous SUAs. These results
demonstrated that ketamine and rTMS have convergent but opposing effects on cortical
oscillations and circuits. This highly robust phenomenon may have important implications to
understanding the neurobiological mechanisms of rTMS and ketamine as well as developing new
therapeutic strategies involving both neuromodulation and pharmacological agents.

62
3.2 Method
3.2.1 Pharmacological manipulations and electrophysiological recording
All animal experiments were conducted following protocols approved by the Institutional
Animal Care and Use Committee of the University of Southern California. Sprague-Dawley rats
(n = 11, female, 220-250 g, 11-12 weeks) were used in this study. Animals underwent a rapid
inhaled induction with 4% isoflurane, followed by an intraperitoneal injection of ketamine (75
mg/kg) and xylazine (10 mg/kg) mixture. Additional doses of ketamine (36 mg/kg) were injected
intraperitoneally to maintain a constant anesthesia level, which was evaluated via the hindlimb
pedal withdrawal reflex and breathing rate. Body temperature was maintained with a feedbackcontrolled heating pad. Rats were mounted on a stereotaxic frame via ear bars and a nose cone in
a Faraday cage. A right-sided craniotomy was conducted to expose the S1 and M1. Dura mater
was removed. A 64-channal silicone probe (Neuronexus A8x8-Edge-5mm-50-150-177, Ann
Arbor MI, USA) was implanted in layer 5 of the M1 or layer 4-5 of the S1 using a
micromanipulator (Fig. 3.1). The probe was parallel to the sagittal suture (midline) of the skull.
The stereotaxic coordinates of the implantation sites relative to the bregma are shown in Table 3.1.
Two small holes were drilled on the occipital bone to place the ground wire and the reference wire.
The TMS coil was parallel to the midline and tilted at 15° to ensure that its center point was
right above the electrode position (Fig. 3.1). The coil was placed above the skull with a distance
of ~1 mm. Two courses of 3-minute, 10 Hz rTMS (~1800 pulses) were delivered to the S1 or M1
with a 15-minute interval. LFPs and spontaneous unitary (spiking) activities were monitored
before and after each course of rTMS in all animals. Five minutes of recordings prior to each
stimulation was used as baseline. To keep the consistency of electrophysiological recordings, the
experiment was restricted to a 45-minute window when the animal was under a stable anesthesia

63
depth. Wideband signals were acquired with an OmniPlex Neural Recording Data Acquisition
System (Plexon, Dallas TX, USA) at a 40 kHz sampling rate.
Figure 3.1. Simultaneous rTMS and intracranial electrophysiological recording in rat sensorimotor cortex.
(A): TMS coil was positioned above the skull and parallel to the midline with an incidence angle of 15°. A
64-channel silicon probe was implanted to the primary somatosensory (S1) or motor cortex (M1). (B):
Locations of silicon probes were verified in brain slices with cresyl violet (Nissl) staining. Probe tracks
were identified at layer 4-5 of the S1 (left) or layer 5 of the M1(right).
Spike sorting was performed with a Plexon offline sorter. Only units with a clear refractory
period, an above 50 μV peak-to-peak amplitude, and a consistent waveform shape were included in
the analysis. Frequency and time-frequency analysis was performed with a custom MATLAB
script. The data was first lowpass filtered and downsampled to 1000 Hz. To visualize the frequency
characteristics across the continuous recording, the Welch’s method was used to compute the power
spectral density (power spectrum). Time-frequency data was visualized via the short-time fast
Fourier transform (FFT) using a Hann window of 1 second with 75% overlap. The resulting output
was a spectrogram with a temporal resolution of 0.25 s and a frequency resolution of 1 Hz.

64
A single silicon probe was utilized throughout the study. The electrode impedance was remeasured in vitro via a NanoZ impedance tester (White Matter, Seattle WA, USA) after the
experiment. All functional electrodes remained at an impedance of about 0.5 MΩ at 1 kHz. The
seventh shank (Site49-57) of the probe was broken in the recordings of Rat04 to Rat11. Signals
recorded from those unfunctional channels were not included in analysis. Recordings in Rat03
were 5-minute shorter than recordings in other animals.
After the experiment, animals were perfused transcardially with 10% formalin. The rat brain
was extracted from the skull, dehydrated in 18% sucrose, stored at 4 °C, and then sliced into 50-
um-thick coronal brain slices with a Cryostat (Leica, Buffalo Grove IL, USA). To verify recording
locations, brain slices were stained with cresyl violet and photographed under a microscope (Leica,
Buffalo Grove IL, USA).

65
Animal ID Cortex
Stereotaxic coordinates
Unit yield
AP ML Depth
Rat01 S1 0.00 3.80 -2.00 46
Rat02 S1 0.00 3.80 -2.59 29
Rat03 S1 0.00 3.80 -2.12 21
Rat04 S1 0.00 3.80 -2.05 34
Rat05 S1 0.00 3.80 -2.30 33
Rat06 M1 2.40 3.00 -2.00 28
Rat07 M1 2.40 3.00 -2.10 19
Rat08 M1 2.40 3.00 -2.14 44
Rat09 M1 2.40 3.00 -2.49 31
Rat10 M1 2.40 3.00 -2.11 27
Rat11 S1 0.00 3.80 -2.10 -
Table 3.1. Targeted cortex, corresponding stereotaxic coordinates of the first shank, and unit yields of
animal used in this study. S1: primary somatosensory cortex, M1: primary motor cortex, AP: anteriorposterior, ML: medial-lateral. Unitary activity was not recorded in Rat11.
3.2.2 Statistical analysis
All data are presented as the mean  standard error of the mean (SEM). A two-tailed t-test was
performed to compare the normalized power of LFP bands for each minute after rTMS to the 5-
minute pre-rTMS baseline (significance level: = 0.05). A two-tailed t-test was performed to
compare the normalized PPS features for each minute after rTMS to the 5-minute pre-rTMS
baseline (significance level: = 0.05). A paired t-test was conducted to compare the pre-rTMS
and post-rTMS neuronal firing rates (significance level: = 0.01).

66
3.3 Results
3.3.1 Ketamine induces poly population spikes in addition to slow-wave activities
Using the MEA, we monitored electrophysiological activities from the S1 and M1 in vivo in
rats (Fig. 3.1). Neural activities recorded after anesthetic dosing of ketamine are divided into three
phases (Fig. 3.2). The first phase started right after MEA implantation, which happened
approximately 40 minutes after the first ketamine/xylazine administration. This phase was close
to the recovery from anesthesia. It showed low-amplitude activities in the LFP and continuous
neuronal firing (Fig. 3.2Aa, Ba). When the animal had a positive pedal withdrawal reflex, the first
additional dose of ketamine was administered, which initiated the second phase. Slow-wave
activities (SWAs) were rapidly induced by ketamine, which was evident by the increasing
amplitude of LFP and the periodic patterns of spiking activity (Fig. 3.2Ab, Bb). This phenomenon
was reported in several previous studies (Chauvette et al., 2011; Fiáth et al., 2016; Horváth et al.,
2021). Two states alternated in SWAs: up-states with intensive neuronal spike firing and downstates with cessation of neuronal spike firing. As the animal was still at a light plane of anesthesia,
the second additional dose of ketamine was administered 10 minutes after the first additional dose
to start the third phase. After the second additional dose, the amplitude of LFP further increased,
and the neuronal spike firing became highly rhythmic. Most interestingly, a novel LFP pattern
was observed: within a few minutes of the second additional dose, a train of high-voltage (> 1 mV)
population spikes, i.e., PPS started to appear. The PPS pattern was irregular at first. Ten minutes
after the second additional dose, while the animal was under deep anesthesia without any reflexes,
the PPS pattern gradually became more regular until it reached a stable state, characterized as a
train of 10-20 population spikes with a 50-100 ms duration and a 3-5 Hz frequency lasting for 3-5
seconds (Fig. 3.2Ac, 3.2Bc, and Fig. 3.3). The PPS then alternated with the SWA (Fig. 3.3A).

67
Like neuronal spike firings during SWAs (up-states), the neuronal spike firing occurred during
depth-negative phases of PPS. This ketamine-induced PPS persisted for >45 minutes across the
cortex, and it was synchronous across all recording channels.

68
Figure 3.2. Evolvement of sensorimotor cortical signals with ketamine injections. (A): Intracranial
recording (S1, Rat11) are divided into three phases by two doses of ketamine. Representative data segments
from the three phases are expanded to different timescales (a, b, and c). a: Low-amplitude activity before
the first additional dose of ketamine. b: Slow-wave activity after the first additional dose of ketamine. c:
Poly population spikes (PPS) after the second additional dose of ketamine. These segments are further
expanded and high pass filtered (250 Hz) to show the correlation between different states of LFP and
neuronal spike firing (fourth row). (B): Spectrograms of the signals shown in (A).
Figure 3.3. Alternation between slow-wave activities (SWA) and PPS as the signature pattern of ketamine.
(A), top: wideband signal showing SWA and PPS; (A), bottom: spectrogram of the signal. The horizontal
strips during PPS reflect the harmonics of the fundamental frequency of population spikes within PPS. (B):
Distributions of number of population spikes (PS) during PPS, PPS duration, frequency of PS during PPS,
and PPS-PPS intervals.
3.3.2 rTMS suppresses ketamine-induced poly population spikes and changes the power of
LFP bands
After PPS were induced, we further applied two courses of rTMS (10 Hz, 3 min) to the cortices
and monitored electrophysiological activities continuously. Compared with the baseline recording,

69
rTMS effectively and reversibly suppressed the ketamine-induced PPS after both courses of
stimulation (Fig. 3.4A, B). In the first course of rTMS, no PPS were observed after the end of
rTMS until 21916 seconds later in the S1 (n=5) and 18114 seconds later in the M1 (n=5). In
the second course of rTMS, the PPS reappeared 15310 seconds in the S1 (n=5) and 85+13
seconds in the M1 (n=5) after the end of rTMS.
Figure 3.4. PPS were effectively and reversibly suppressed by rTMS. (A): Wideband signal and its
spectrogram before and after the first course of rTMS. (B): Wideband signal and its spectrogram before
and after the second course of rTMS. Red triangles denote the first PPS occurred after rTMS.

70
Quantitative analysis was conducted on power spectra of LFPs with one-minute resolution (Fig.
3.5A). LFP frequency was classified into delta (1-4 Hz), theta (4-8 HZ), alpha (8-12 Hz), beta
(12-35 Hz), and gamma (35-100 Hz) bands in this analysis. Mean power of each individual
frequency band was computed for each course of rTMS. Results showed that the first course of
rTMS significantly decreased the beta and gamma power of the LFP in the S1, while there was an
increase in all power bands after ~5 minutes, compared to the baseline (Fig. 3.5B). In the second
course of rTMS in the S1, there was a significant suppression in theta, alpha, beta, and gamma
band powers, while there was an enhancement in the delta band power (Fig. 3.5B). Different from
the first course of rTMS, the mean power of each frequency band returned to the baseline level ~5
minutes after the second course of rTMS. In the M1, the mean power of LFP had similar results
to those observed in the S1. rTMS significantly decreased the theta, alpha, beta, and gamma band
powers in both courses of stimulation, while the delta band power increased in the second course
of rTMS (Fig. 3.5C). The effect of rTMS persisted for ~5 minutes in both courses. After that,
there was an increase in alpha, beta, and gamma band powers after the first course of stimulation.
However, the mean power of each band returned to the baseline level after the second course of
stimulation. Those results indicated that rTMS modulated the neural oscillations in different
frequencies in both S1 and M1.

71
Figure 3.5. Changes of LFP power distribution in different frequency bands in the S1 and M1 after each
course of rTMS. (A): Power spectra at every minute before and after two courses of rTMS (Rat01). The
mean power of delta (1-4 Hz), theta (4-8 HZ), alpha (8-12 Hz), beta (12-35 Hz), and gamma (35-100 Hz)
bands are calculated and compared with the baseline level (p < 0.05, two-tailed t-test). (B): The mean
power of each frequency band in the S1 before and after each course of stimulation (n=5). (C): The mean
power of each frequency band in the M1 before and after each course of stimulation (n=5). Error bars:
SEM.

72
To further assess whether the suppression of PPS by rTMS alone could explain the changes in
powers of frequency bands caused by rTMS, three types of signals, i.e., pre-rTMS LFP during PPS,
pre-rTMS LFP without PPS, and post-rTMS LFP without PPS, were separated from the data, and
their power spectra were compared (Fig. 3.6). Pre-rTMS LFP during PPS and pre-rTMS LFP
without PPS signals were obtained from 5 minute before rTMS. Post-rTMS LFP without PPS
signals were selected from the first minute after rTMS. Results showed that there was little
difference in power spectra of LFP without PPS between pre-rTMS and post-rTMS signals (Fig.
3.6, red and yellow lines). LFP during PPS showed a decreased power in delta band and increased
powers in alpha, beta, and gamma bands (Fig. 3.6, blue lines). The first peak in the PPS power
spectrum, which was caused by the fundamental frequency (3-4 Hz) of population spikes during
the PPS, and the following peaks caused by the harmonics of the fundamental frequency, formed
a signature “sawtooth” pattern of PPS in the power spectrum. These results clearly indicated that
the suppression of PPS by rTMS caused the changes of LFP power.
Figure 3.6. Comparison of power spectra between pre-rTMS LFP during PPS (blue), pre-rTMS LFP
without PPS (red), and post-rTMS LFP without PPS (yellow). (A): The first (left) and second (right) course
of rTMS in the S1. (B): The first (left) and second (right) course of rTMS in the M1. Note the signature
“sawtooth” pattern in PPS.

73
Since PPS had a non-sinusoidal broadband waveform with canonical oscillation that could not
be sufficiently indicated by different frequency bands in its power spectra(Cole & Voytek, 2017),
additional waveform-specific metrics such as slope of individual population spike (PS) during PPS,
amplitude of PS during PPS, frequency of PS during PPS, number of PS during PPS, PPS duration,
and PPS-PPS intervals were used to further quantify PPS features (Fig. 3.7). The slope of PS was
measured between the peak of PS and 10 ms before the peak; the frequency of PS in PPS was
defined as the mean frequency of PS during a PPS. These metrics were further averaged for every
minute in each course of rTMS in both cortices. Results showed that both slope (Fig. 3.7A) and
amplitude (Fig. 3.7B) of PS significantly increased several minutes following the recovery of PPS
after rTMS in the first course of rTMS (p < 0.05). Interestingly, this increase happened ~5-10
minutes after rTMS in the S1 but happened ~10-15 minutes after rTMS in the M1. Such increase
was not observed in the second course of rTMS. This pattern (increase after the first course of
rTMS but not the second course of rTMS) was also observed in the frequency bands of LFP (Fig.
3.5). No significant changes were observed in the second course of rTMS (p > 0.05) in the S1
except for an increase of PS slope in the eighth minute after stimulation (p = 0.0088, t = 4.7719).
However, the slope and amplitude of PS significantly decreased in the fifth and sixth minutes after
the second course of rTMS in the M1 (p < 0.05), which might be caused by the gradual recovery
of PPS. However, there was no obvious changes during PPS recovery after the first course of
rTMS in frequency of PS during PPS (Fig. 3.7C), number of PS during PPS (Fig. 3.7D), PPS
duration (Fig. 3.7E), or PPS-PPS interval (Fig. 3.7F), which indicated that the suppressive effects
of rTMS on PPS and its recovery were mainly on slope and amplitude of individual PS.

74
Figure 3.7. Changes of PPS features after the first (black) and second (grey) courses of rTMS in S1 (left)
and M1 (right). (A) Slope of population spikes (PS) during PPS. (B) Amplitude of PS during PPS. (C)
Frequency of PS during PPS. (D) Number of PS during PPS. (E) PPS duration. (F) PPS-PPS intervals.
All values were normalized and compared with the baseline recordings (p < 0.05, two-tailed t-test). Error
bars: SEM.

75
3.3.3 rTMS increases spontaneous neuronal spike firing rates
Lastly, we compared the single neuron-level activities between pre- and post-rTMS. In total,
163 and 149 single neurons were recorded in the S1 and M1 (n = 5), respectively (Table 1). Figure
3.8 shows a representative example of one-minute-long high pass filtered (250 Hz) unit activities
immediately before and after two courses of rTMS (Site63 in Rat06). Both courses of rTMS
consistently increased firing rates of neuronal spikes in both S1 and M1 (Fig. 3.9). Different levels
of modulation to the SUAs were shown in each neuron. Linear regressions were performed on
mean firing rates of all neurons and compared between pre-rTMS and post-rTMS (Fig. 3.10).
Mean firing rates after stimulation were significantly higher than those before stimulation (Fig.
3.10, paired t-test, A: p<0.01, B: p<0.01, n = 163; C: p<0.01, D: p<0.01, n = 149). In both S1 and
MI, mean firing rates peaked within 1 or 2 minutes after rTMS then gradually decayed to a second
baseline, which was lower than the first baseline, after the first course of rTMS; while after the
second course of rTMS, mean firing rates decayed to the second baseline without further decaying
(Fig. 3.10E, F). While both S1 and M1 neurons showed multiplicative increases in mean firing
rates in both courses of rTMS, S1 neurons exhibited a more prominent increase in mean firing
rates compared to M1 neurons.
To evaluate the influence of PPS on the neuronal spike firing rates, firing rates were calculated
for pre-rTMS signal during PPS, pre-rTMS signal without PPS, and post-rTMS signal without PPS.
In both courses of stimulation, post-rTMS firing rates increased compared to pre-rTMS firing rates
despite the occurrence of PPS (Fig. 3.11). Pre-rTMS firing rates during PPS were higher than prerTMS firing rates without PPS in S1 neurons but not in M1 neurons.

76
Figure 3.8. Spontaneous unitary activities recorded pre- and post-rTMS. One-minute of high pass filtered
(250 Hz) signals from the M1 before and after the first (A) and second (B) course of rTMS. (C): Waveforms,
PCA clusters, and spike autocorrelograms (± 50 ms, bin size: 0.5 ms) of two units sorted from this signal.

77
Figure 3.9. Changes of neuronal spike firing rates (bin size: 2 s) before and after each course of rTMS. (A):
Firing rate histograms of all S1 neurons (n=163). (B): Mean firing rates of S1 neurons from each animal
(n=5). (C): Firing rate histograms of all M1 neurons (n=149). (D): Mean firing rates of M1 neurons from
each animal (n=5).

78
Figure 3.10. Scatter plots of pre-rTMS versus post-rTMS firing rates (bin size: 1 min) of neurons at
different time points (1, 2, 3, 4, 5, 6, 10, 15 min) after each course of rTMS. Firing rates of S1 neurons are
compared before and after the first (A) and second (B) course of rTMS (n=163). Firing rates of M1 neurons
are compared before and after the first (C) and second (D) course of rTMS (n =149). Linear regression
(solid line) is superimposed on the scatter plots. (E, F): The slope of the linear fits at each time points after
rTMS for both S1 and M1 neurons. Shaded areas: SEM.

79
Figure 3.11. Comparison of neuronal spike firing rates between pre-rTMS signal during PPS, pre-rTMS
signal without PPS, and post-rTMS signal without PPS. (A): Firing rate histograms of S1 neurons (n=46,
bin size = 2 s) during the three types of signals. (B): Firing rate histograms of M1 neurons (n=19, bin size
= 2 s) during the three types of signals. (C) and (D): Scatter plots of firing rates of neurons. Blue: prerTMS during PPS versus post-rTMS without PPS. Red: pre-rTMS without PPS versus post-rTMS without
PPS.

80
3.4 Discussion
Most commercially available TMS coils are designed for human subjects and thus have large
geometric sizes that cause high-intensity non-focal stimulation in small animals. In addition, those
commercially available coils such as classic figure-eight coils cannot be conveniently used in
conjunction with the standard MEA due to the limited space between the coil and the brain surface.
We have designed, fabricated, and characterized a miniaturized TMS coil for rodent studies. The
coil can generate a B-field strength (~400 mT) stronger than those of previously reported TMS
coils with similar dimensions (Makowiecki et al., 2014; Rodger et al., 2012; Tang et al., 2016). In
both human and animal experiments, motor threshold is determined as the minimum E-field
intensity produced by the TMS coil that can result in predefined MEPs in at least 5 of 10
consecutive trials (Herbsman et al., 2009). The induced E-field generated by our coil was
measured to be greater than 3 V/m in a 5-by-3 mm space (at 4 mm) in saline and much lower than
the motor threshold of 100 V/m measured in previous rodent TMS studies (Boonzaier et al., 2020;
Salvador & Miranda, 2009). However, even at this subthreshold intensity, 10 Hz rTMS, one of
the most commonly used TMS protocols, has been shown to increase the amplitude of MEPs in
both humans and rats (Maeda et al., 2000; Tang et al., 2016). To evaluate the reliability and
reproducibility of our findings, two courses of subthreshold rTMS at 10 Hz were focally delivered
via the miniaturized TMS coil to modulate brain activities of the sensorimotor cortex in rats.
The first finding of this study is that ketamine can reliably induce a highly synchronized
activities, i.e., PPS, in sensorimotor cortices. To the best of our knowledge, this is the first report
of this type of activities in LFP induced by ketamine. The other type of ketamine-induced activity,
SWA, is well documented in previous studies (Chauvette et al., 2011; Fiáth et al., 2016; Horváth
et al., 2021; Neske, 2016). SWA and PPS shows distinct waveforms and frequency-domain

81
characteristics: SWA is relatively sinusoidal with down-states and up-states. PPS is nonsinusoidal and has highly stereotypical waveforms that consist of multiple population spikes. We
found that under deep ketamine anesthesia, SWA and PPS alternated in a highly regular fashion.
This phenomenon presented a novel biomarker of ketamine’s effect on the sensorimotor cortex.
Furthermore, due to its high degree of repeatability (observed in all animals), this ketamineinduced PPS can also be used as an animal model for studying cortical oscillation and synchrony
in general.
The second finding of this study is that rTMS can effectively and reversibly suppress the
ketamine-induced PPS. rTMS changed the power distribution of frequency bands in LFP: it
suppressed alpha, beta and gamma bands, while enhanced the delta band, as a direct consequence
of its strong suppressive effects on PPS since PPS had high powers in alpha, beta and gamma
bands, and lower power in the delta band. After PPS recovery, a “rebound effect” on LFP power
was observed in the first rTMS but not in the second rTMS. This “rebound effect” was mainly
due changes in slope and amplitude of individual PS during PPS. These changes might be caused
by the effect of ketamine alone or the combined effects of rTMS and ketamine. The first possible
explanation is that such seeming rebound effect was caused by the gradual increase of ketamine
effect alone: ketamine did not reach its full effect before the first rTMS. Therefore, after the
suppressive effect of rTMS, ketamine effect kept increasing and eventually reached its steady-state
maximum level before the second rTMS. Consequently, such increase happened after the first
rTMS but not the second rTMS. If this was what happened, this “rebound effect” was not a real
rebound but essentially a summation of rTMS suppression on top of the increase of ketamine effect
in the baseline. Previous studies showed that ketamine injection led to dynamic changes in plasma
ketamine concentration and brain activities (D. Li & Mashour, 2019; Nicol & Morton, 2020; Sato

82
et al., 2004; Veilleux-Lemieux et al., 2013). The second possible explanation is that the rebound
effect was indeed caused by rTMS; after the short-term suppressive effect of the first rTMS, PPS
not only recovered but also rebounded to a higher level of baseline. Such rebound effect should
also have been saturated after the first rTMS and thus failed to happen again after the second rTMS.
Additional experiments on the pharmacokinetic and pharmacodynamic properties of ketamine and
a deeper understanding to the rTMS effect are required to address this question. Nonetheless, the
immediate suppressive effect of rTMS to PPS was robust in both courses of rTMS in both cortices.
In addition, we found that SWA and PPS responded very differently to rTMS: rTMS could
robustly abolish PPS, whereas rTMS had little effect on low frequency component of SWA. The
underlying mechanism of SWA is relatively well understood. It is a synchronized cortical
oscillation that involves neurons in multiple brain regions including cortical layers and thalamus
(Fiáth et al., 2016; Neske, 2016). Previous studies have suggested that the firing of cortical layer
5 neurons and thalamocortical neurons mainly contributes to the initiation of the up-states; the
NMDA receptors may be involved in the persistence of the up-states; the facilitation of inhibitory
interneurons, depression of excitatory synapses, or activation of calcium-dependent potassium
conductance may result in the termination of the up-states (Neske, 2016). By contrast, PPS
induced by high dose of ketamine is a new phenomenon that has never been reported before.
Therefore, its mechanisms remain unclear. The distinct effects of rTMS on PPS and SWA
suggested that PPS and SWA might be caused by different neuronal and neural network-level
mechanisms and these mechanisms could further be differentially modulated by rTMS. The
suppression of PPS indicated that rTMS and ketamine had a convergent but opposing effect on
synchronized activities in LFP. They may affect the cortical dynamics with overlapping cellular
and network-level mechanisms such as the glutamatergic and GABAergic signal transmissions.

83
Glutamate and GABA are the dominant excitatory and inhibitory neurotransmitters in the
central nervous system. Balance between glutamatergic excitation and GABAergic inhibition is
crucial for maintaining normal operations of neuronal circuits (Hampe et al., 2018; Lazarevic et
al., 2013). It was hypothesized that the administration of ketamine at the anesthetic level might
lead to a disruption of GABAergic and glutamatergic systems (Akeju et al., 2016; Sleigh et al.,
2014). Blockage of NMDA receptors on interneurons by ketamine reduces GABA signaling and
disinhibits pyramidal neurons(Seamans, 2008). Although the mechanism of the ketamine-induced
PPS remains unknown, given its similarity in waveforms with interictal spikes, it is likely that the
disinhibition of the glutamatergic excitation and the disruption of the GABAergic inhibition
contribute to the occurrence of PPS (Avoli et al., 2006; Staley & Dudek, 2006). It must be noted
that the effect of ketamine on glutamatergic signaling is dose-dependent. There is an excitatory
effect caused by a surge of glutamate following a subanesthetic dose of ketamine, while inhibitory
effects occur at an anesthetic dose (Moghaddam et al., 1997; Silberbauer et al., 2020). Interestingly,
GABAergic and glutamatergic systems are also potential targets of rTMS. Previous studies have
shown that 10 Hz rTMS increases the GABA level in depression patients(M. J. Dubin et al., 2016).
This increase of GABA may explain the suppression effect of rTMS on the PPS observed in this
study. Further experimental and modeling studies are required to fully elucidate the complex
interactions of ketamine and rTMS on the cortical oscillations and circuits.
The third finding of this study is that rTMS can significantly increase the firing rates of single
neurons in the sensorimotor cortices. The firing rates of neuronal spikes largely increased when
there was no PPS, e.g., immediately after rTMS, compared with the pre-rTMS firing rates either
during or without PPS. These results suggest that there might be multiple mechanisms contributing
to the changes of firing rates caused by rTMS. Indeed, it was also reported that 10 Hz rTMS

84
reduced the GABAergic synaptic strength (Lenz et al., 2016) and increased the glutamatergic
synaptic strength (Lenz et al., 2015; Vlachos et al., 2012) in mouse slice cultures. One plausible
hypothesis is that the modification of synaptic plasticity caused by rTMS accounts for the increase
of spontaneous spiking activities we report in this manuscript. It is worth noting that there is a
strong correlation between changes in firing rates and changes in LFP after the first and the second
rTMS. It is highly likely that the rebound effects observed in firing rate and LFP power share
some common mechanisms, either due to the effect of ketamine along or the combined effects of
rTMS and ketamine described above. However, it is unlikely that the changes in firing rate are
directly caused by the changes of PPS, since there is no significant difference in firing rates during
PPS or periods without PPS (Fig. 3.11). Despite the unknown mechanisms, there seem to be a
strong negative correlation between occurrence of PPS, a high-frequency oscillation presumably
caused by abnormally synchronized spiking activities, and firing rate of asynchronized spike
activities, which presumably underlay normal information processing and brain functions. Our
results show that ketamine tends to push cortical circuits to the abnormal synchronized state while
rTMS on the other hand is highly effective in restoring cortical circuits to the normal
asynchronized state. This opposing effect of ketamine and rTMS on cortical oscillation may have
important implications to understanding the underlying mechanisms of ketamine and rTMS.
Neocortical areas utilize similar types of neurons and circuit organizations to achieve different
functions (Harris & Shepherd, 2015). S1 is responsible for processing somatic sensations; M1
controls voluntary movements. In this study, rTMS showed similar but variable effects on the S1
and M1: rTMS suppressed PPS for a longer period of time and caused a larger increase of neuronal
spike firing rates in the S1 than in the M1. Previous studies have shown that LTP can be more
reliably induced by electrical stimulation in the S1 than in the M1 (Castro-Alamancos et al., 1995).

85
Although the rTMS effects showed in this study were much shorter than LTP (minutes vs. hours),
it is possible that the two phenomena (stronger responses in the S1 than M1) were caused by similar
mechanisms underlying synaptic plasticity.
In this study, the effects of subthreshold rTMS and ketamine on cortical activities were
investigated in rats deeply anesthetized with ketamine. The main discovery was that ketamine and
rTMS had converging effects on sensorimotor cortical oscillations, and such effects were indicated
by the robust induction of PPS by high-dose ketamine and effective abolishment of such PPS by
rTMS. It is worth noting that the anesthetic dose of ketamine administered in this study (75 mg/kg)
is much higher than the subanesthetic dose used in treating depression and inducing psychedelic
effects in humans (0.1-0.75 mg/kg infused intravenously over 40 min) even considering the dose
conversion between rat and human (divide by 6.2) (Andrade, 2017; Ballard & Zarate, 2020;
Krystal et al., 1994, 2019; Nair & Jacob, 2016). Since the effects of rTMS are inherently statedependent (Silvanto & Pascual-Leone, 2008), rTMS may produce different after-effects under
different doses of ketamine or different anesthesia regimens. Although the discovery of this study
could not be directly used to explain the combined therapy of rTMS and subanesthetic ketamine
in human studies, it provided a platform for studying the chemical and electromagnetic interactions
on cortical circuits at single neuron and neuronal population resolutions. In addition, the
alternation of SWA and PPS induced by ketamine observed in this study could be used as a novel
biomarker to monitor and characterize the anesthesia state. This phenomenon might provide new
insights into the anesthetic action of ketamine on brain activities.
In future studies, we will further investigate the effect of ketamine with different doses and the
effect of rTMS with different stimulation parameters in both anesthetized animals and behaving
animals performing cognitive tasks. Such studies will deepen our understanding to the underlying

86
mechanism of effects of ketamine and rTMS, as well as their interactions, and may have important
implications to the development of a combined chemical and electromagnetic therapeutic strategy
for treating neurological and neuropsychiatric disorders.

87
Chapter 4: Subthreshold rTMS Alters Learning and Memory
Performance of Rats in the Barnes Maze Task
4.1 Introduction
rTMS is a promising non-invasive neuromodulation technique widely used in clinical and
research settings. It has been approved by the FDA for treating neurological and neuropsychiatric
conditions such as major depression (Levkovitz et al., 2015; O’Reardon et al., 2007) and OCD
(Carmi et al., 2018, 2019), as well as for aiding in smoking cessation (Zangen et al., 2021).
Additionally, preliminary studies have shown potential efficacy of rTMS in improving various
cognitive functions in humans, including working memory (Brunoni & Vanderhasselt, 2014),
cognitive flexibility (Guse et al., 2010), and episodic memory (Yeh & Rose, 2019). In these studies,
the PFC is the primary rTMS target, attributed to its critical role in the short-term working memory
retention (Curtis & D’Esposito, 2003; Lara & Wallis, 2015), the flexible cognitive control (C. Kim
et al., 2011; Rougier et al., 2005), and the encoding and retrieval processes of episodic long-term
memory (Blumenfeld & Ranganath, 2007; Lepage et al., 2000).
While rTMS has primarily been studied in the context of human clinical applications, animal
models, especially rodents, have played a significant role in investigating the potential cognitive
effects and underlying mechanisms of this neuromodulation technique. Previous studies have
demonstrated improvements in cognitive functions across rodent models of Alzheimer's disease
(Lin et al., 2021; Tan et al., 2013), Parkinson's disease (H. Wang & Gao, 2023), vascular dementia
(H.-Y. Yang et al., 2015; Zhang et al., 2015), aging (Ma et al., 2014; Weiler et al., 2021), and sleep
deprivation (Estrada et al., 2015). Additionally, several investigations involving healthy animals
have been conducted to evaluate the baseline effects of rTMS on learning and memory. One study

88
revealed that high-frequency (10 Hz) subthreshold rTMS enhanced spatial episodic memory in
rats performing the Morris water maze (MWM) task (Wu et al., 2022). Another study
incorporating the reversal learning phase in the MWM found that both spatial and reversal learning
and memory were enhanced by 5 Hz suprathreshold rTMS treatment (Shang et al., 2016). In
contrast, low-frequency (0.5 Hz) suprathreshold rTMS was observed to impair spatial episodic
memory (W. Li et al., 2007). The effects of rTMS on various memory types, such as working (T.
Wang et al., 2023) and recognition memory (Ahmed & Wieraszko, 2006; Zhu et al., 2021), have
also been explored through paradigms like the T-maze and novel object recognition tests.
However, the effects of rTMS on the cognitive functions of healthy rodents remain to be
thoroughly investigated. In this study, to assess the chronic and acute effects of high-frequency
subthreshold rTMS, the Barnes maze (BM) task (Barnes, 1979) was employed, offering a less
physical stressful alternative to the MWM. Subthreshold 10 Hz rTMS or sham treatment was
administered to the rat’s PFC over ten days to evaluate the chronic cognitive effects. An alternating
rTMS and sham stimulation protocol was designed to evaluate the acute effects within a modified
BM setting. These experimental paradigms provide a robust framework for measuring various
forms of cognition, from the initial acquisition of spatial information to the flexibility in adapting
to changed environments and the strength of spatial memory retention. Given the growing clinical
significance of rTMS in treating disorders like autism (Barahona-Corrêa et al., 2018),
schizophrenia (Sciortino et al., 2021), epilepsy (Cooper et al., 2018), and stoke (Fisicaro et al.,
2019), understanding its broader cognitive implications becomes essential. Such insights will
guide the optimization of rTMS protocols, maximizing therapeutic benefits while minimizing
potential cognitive disruptions.

89
4.2 Methods
4.2.1 Animals
All animal experiments were conducted in accordance with protocols approved by the
Institutional Animal Care and Use Committee of the University of Southern California. Fifteen
male Sprague-Dawley rats (300-350 g, 10-11 weeks) were randomly divided into three groups: a
TMS group (n = 6), a sham group (n = 6), and an alternating TMS/sham group (n = 3). Animals
were housed under standard laboratory conditions with ad libitum access to food and water.
Following a one-week acclimation to the housing conditions, animals were introduced to a 4-day
restraint habituation prior to TMS or sham treatment. The process involved placing each rat in a
rodent snuggle (Lomir, Malone NY, USA) for 5 minutes to reduce any potential stress or
discomfort caused by restraint in the experiment.
4.2.2 Barnes maze protocols
The BM used in this study consisted of a white circular platform (122 cm in diameter) elevated
1 meter off the ground with 16 circular holes (9 cm in diameter) evenly spaced around the
circumference. A black cubic plastic box (15 cm in length) was placed under one of the holes as
the escape box, while the remaining holes were covered from below with black plastic plates.
Surrounding the platform were white curtains adorned with visual cues, including a blue circle, a
green triangle, and black and white stripes. A bright illumination source and a tracking camera
were mounted to the ceiling above the center of the platform. Additionally, a speaker, emitting a
buzzing sound, was positioned under the center of the platform.

90
A standard BM protocol (Gawel et al., 2019) was used for both the TMS and sham groups to
assess the chronic effects of rTMS (Fig. 4.1). The task consisted of 5 phases: habituation,
acquisition, an acquisition probe trial, reversal learning, and a reversal learning probe trial. TMS
or sham stimulation was delivered to the animal during the acquisition and reversal learning phases.
In the habituation phase, visual cues were removed, the light was turned on, and the buzzer was
deactivated. The animal was introduced to the BM platform for 60 seconds. Subsequently, it was
placed inside the escape box for an additional 120 seconds to acclimate to the environment. In the
acquisition phase, the animal was trained to locate the escape box over 6 consecutive days, with 2
trials per day. For each trial, the animal was placed at the center of the platform and covered by a
start box for 10 seconds, while the buzzer remained inactive. Subsequently, the animal was
allowed to freely explore the maze and find the escape box, with the buzzer activated. If the animal
successfully entered the escape box within 90 seconds, the box was covered, and the animal was
allowed to stay inside for an additional 30 seconds. However, if the animal failed to enter, it was
manually placed inside the escape box for the same 30-second duration. Once the animal was
inside the escape box, the buzzer was deactivated. After each trial, 70% ethanol was applied to
clean the surface and eliminate the odors, and a fan was employed for air circulation. Twenty-four
hours post-acquisition phase, an acquisition probe trial was initiated to evaluate spatial memory
retrieval. The escape box was removed from the platform, and its hole was covered like the others.
The animal was placed at the center of the platform, covered by a start box for 10 seconds, while
the buzzer remained inactive. Afterward, the animal was given a 90-second window to navigate
the maze freely, with the buzzer activated. On the same day, the reversal learning phase began to
test cognitive flexibility. Lasting 4 days with 2 trials per day, this phase followed the same protocol
as the acquisition phase, except the escape box was relocated to the opposite side of its initial

91
position. Twenty-four hours post-reversal learning phase, a final reversal learning probe trial was
conducted to assess the adaptability of the animal’s learned responses, following the same protocol
as the acquisition probe trial.
Figure 4.1. Schematic representation of the TMS treatment combined with standard Barnes maze protocols.
After a 4-day restraint habituation period, the animal began the Barnes maze task, which consisted of 5
phases: habituation, acquisition, an acquisition probe trial, reversal learning, and a reversal learning probe
trial. rTMS (5 min, 10 Hz) or sham stimulation was delivered to the restrained animal right before each trial
during the acquisition and reversal learning phases. The treatment was conducted in two daily sessions,
separated by a 10-minute interval, and was continued for 10 consecutive days.

92
A modified BM protocol was used for the alternating TMS/sham group to assess the acute
effects of rTMS (Fig. 4.2). The task consisted of two phases: baseline training and behavioral
assessments. During the baseline training, four BM sessions were conducted in a single day, with
the escape box locations alternating between sessions. The escape box was placed in one location
for the first and third sessions, while it was shifted to the opposite side for the second and fourth
sessions. The initial session comprised 12 trials, whereas the subsequent three sessions each had
8 trials. In each trial, animals were trained to locate the escape box, following the standard
acquisition phase protocol. Trials were spaced by a 3-minute interval, while there was
approximately an hour between sessions. Twenty-four hours after the baseline training, behavioral
assessments with alternating TMS and sham stimulation were initiated to evaluate the acute effects
of rTMS. Same as the baseline training, four BM sessions with alternating escape box locations
were carried out daily. Each session consisted of 8 trials and was repeated for 4 consecutive days.

93
Figure 4.2. Schematic representation of the alternating TMS/sham treatment combined with modified
Barnes maze protocols. On each day, four Barnes maze sessions were conducted, with the escape box (EB)
locations alternating between sessions. After a day of baseline training, the restrained animal was subjected
to either rTMS (5 min, 10 Hz) or sham stimulation in an alternating pattern prior to each session from Day
2 to Day 5. Each session consisted of 8 trials, separated by a 3-minute interval. An approximate 1-hour gap
was maintained between sessions.
4.2.3 rTMS treatment
A C-shaped miniaturized coil (Jiang et al., 2023) was used to deliver TMS or sham stimulation
to the animal. The coil was positioned about 2 mm above the head, specifically targeting the PFC
(Fig. 4.3A). To simulate the B- and E-fields generated by the TMS coil within the rat brain, we
utilized Multiphysics 6.0 (COMSOL, Burlington, MA, USA), as detailed in prior work (Jiang et
al., 2023). The peak B-field was estimated to be 0.29 T directly under the two bases of the coil
(Fig. 4.3B), and the peak E-field was estimated to be 5.2 V/m under the coil's center (Fig. 4.3C).

94
Simulations on the sagittal plane along the brain's midline indicated a rapid decay in both fields
with increased distance. Approximately 3 mm from the brain surface, the B-field reduced to 0.1 T,
while the induced E-field reduced to 1 V/m.
Figure 4.3. Simulation of magnetic and electric fields within the rat brain. (A) Coil placement relative to
a 3D model of the rat's head (left) and its brain (right). (B) Magnetic field simulation on the brain surface
(left) and in the sagittal section (right) along the brain’s midline. (C) Induced electric field simulation on
the brain surface (left) and in the sagittal section (right) along the brain’s midline.
For the TMS group, five-minute 10 Hz subthreshold rTMS was administered to the restrained
animal immediately before each BM trial during the acquisition and reversal learning phases. This

95
treatment consisted of two TMS sessions daily, separated by a 10-minute interval, and continued
for 10 consecutive days (Fig. 4.1). For the sham group, a disconnected sham coil was positioned
over the restrained animal in the same manner as in the TMS group. However, another identical
coil, connected to the stimulator, was used to replicate the auditory effects of TMS. For the
alternating TMS/sham group, either five-minute 10 Hz subthreshold rTMS or sham stimulation
was alternately administered to the restrained animal to counterbalance potential sequence effects
(Fig. 4.2). The treatment was administered right before each BM session throughout the behavioral
assessments.
4.2.4 Data processing and statistical analyses
Behavioral metrics, including latency, error, distance, velocity, hole deviation score, and search
strategy (Table. 4.1), were quantified to assess the animals' behavior. DeepLabCut (Nath et al.,
2019) was used to track the animal’s trajectories. For training the tracking model, the nose, back,
and tail of the rat were manually labeled in 20 frames across each of the 25 selected videos. The
shuffle was set to 3, with a maximum of 100,000 iterations. The resulting coordinates from these
trajectories were extracted for each video. Data processing was performed in MATLAB
(MathWorks, Natick MA, USA), and results were presented as the mean ± standard error of the
mean (SEM). To compare the TMS group with the sham group in the acquisition and reversal
learning phases, as well as TMS trials versus sham trials within the alternating TMS/sham group,
unpaired two-tailed t-tests were conducted at a significance level of 0.05. For comparisons
between the TMS and sham groups in the probe trials, unpaired one-tailed t-tests were conducted
at the same significance level.

96
Latency Time required for the animal to enter the escape box with its head
Error Count of instances where the animal deflects its nose and head into a nonescape hole before entering the escape box with its head
Distance Length of the path traveled by the animal to enter the escape box with its head
Velocity Average speed of the animal, calculated by dividing the distance by the latency
Hole deviation score Number of holes between the first hole the animal visited and the escape box
Search strategy (a) Direct: The animal either directly reaches the escape box or explores 1 or 2
adjacent holes before locating the target.
(b) Serial: Before reaching the escape box, the animal sequentially visits at least
two adjacent holes that are not directly beside the target, either in a clockwise
or counterclockwise direction.
(c) Mixed: The animal displays a mixture of search strategies, either crossing
through the center of the maze or demonstrating a random, unorganized
sequence of hole visits.
Table 4.1. Definitions of behavioral metrics used in this study.
4.3 Results
4.3.1 rTMS disrupts spatial learning in the acquisition phase
In the acquisition phase of the BM task, animals were trained to locate the escape box over six
consecutive days, with two trials per day (Fig. 4.4A). Immediately before each trial, they received
either rTMS or sham stimulation. The primary aim of this phase was to evaluate the effect of TMS
on the animals' spatial learning capabilities. To quantify their performance, several metrics were
employed, including latency to find the escape box, number of errors made, distance traveled,
average velocity, hole deviation score, and search strategy employed.
During this phase, both the TMS (n = 6) and sham groups (n = 6) demonstrated a decrease in
latency (Fig. 4.4B), error (Fig. 4.4C), and distance (Fig. 4.4D) over the trials. However, the TMS

97
group occasionally exhibited rebounding trends. On average, the TMS group took more time,
committed more errors, and traveled longer distances before locating the escape box compared to
the sham group. A significant difference between the two groups was observed in the fifth trial
(latency: p=0.036, t=2.417; error: p=0.011, t=3.115; distance: p=0.033, t=2.466). In addition, on
Day 5, the distance traveled by the TMS group significantly increased in both the ninth (p=0.040,
t=2.360) and tenth (p=0.031, t=2.500) trials, which corresponded to a higher hole deviation score
in the ninth trial (p=0.035, t=2.432; Fig. 4.4E). Despite these differences, the average velocity
increased over time for both groups (Fig. 4.4F), with no statistically significant difference between
them (p>0.05).
The animals' search strategies were categorized into three types: direct, serial, and mixed
searches (Fig. 4.4G). A direct search reflects an efficient pattern, suggesting that the animal has
acquired the escape box's location and employed spatial cues to navigate the maze. In contrast, a
serial search represents a systematic approach, indicating that while the animal hasn't solidified a
spatial memory of the escape box, it employs a procedural method for maze navigation. The mixed
search shows random search patterns, suggesting the animal neither recognizes the escape box's
location nor employs spatial or serial cues for maze navigation. During the acquisition phase, both
the TMS and sham groups predominantly adopted the serial search strategy in the initial five trials
(TMS group: 93.33±4.22%, sham group: 76.67±9.55%; Fig. 4.4H). From the sixth trial onward,
the sham group increasingly employed the direct search strategy, contrasting with the TMS group's
behavior (TMS group: 23.81±4.76%, sham group: 40.48±7.75%; p=0.048, t=-1.832; Fig. 4.4I).
Although half of the TMS group employed the direct search method in the seventh trial, they
reverted to the serial search as their primary approach thereafter. The sham group only exhibited
mixed search behaviors early in the phase, while the TMS group exhibited such behaviors more

98
sporadically. These observations suggest that rTMS disrupted spatial learning during the
acquisition phase of the BM task.
Figure 4.4. rTMS hindered spatial learning during the acquisition phase of the Barnes maze task. (A)
Representative trajectories of rats under rTMS (left) and sham treatments (right). Latency to find the escape
box (B), number of errors made (C), distance traveled (D), hole deviation score (E), and average velocity
(F) were statistically analyzed to compare performance between the TMS (red) and sham (blue) groups. (G)
Examples of representative trajectories of the direct (top), serial (bottom left), and mixed (bottom right)

99
search strategies. (H) Stacked bar charts presenting the ratios of search strategies(direct: blue, mixed: white,
serial: red) employed in each trial for the TMS (top) and sham (bottom) groups. (I) Bar plots indicating the
percentage of each search strategy (direct: left, mixed: middle, serial: right) in the last 7 trials for the TMS
(red) and sham (blue) groups. Comparisons of the TMS and sham groups were performed with t-tests (∗p
< 0.05, ∗∗p < 0.01, ns: no significance). Values were reported as the mean ± SEM.
4.3.2 rTMS enhances cognitive flexibility in the reversal learning phase
Following the acquisition and the acquisition probe trial, animals initiated the reversal learning
phase of the BM task. They were trained to locate the escape box, which had been moved to the
opposite side from its original position, over four consecutive days with two trials per day (Fig.
4.5A). Same as in the acquisition phase, animals received either rTMS or sham stimulation right
before each trial. The primary aim of this phase was to evaluate the effect of TMS on the animals'
cognitive flexibility.
During the first trial of this phase, the TMS group (n = 6) more efficiently located the newly
positioned escape box than the sham group (n = 6), taking less time (p=0.037, t=-2.003; Fig. 4.5B),
committing fewer errors (p=0.004, t=-3.776; Fig. 4.5C), and traveling shorter distances (p=0.003,
t=-3.852; Fig. 4.5D). However, in subsequent trials, the performances of the two groups did not
significantly differ (p>0.05). Two key observations can elucidate these findings. Firstly, the initial
hole chosen by the TMS group in the first trial was closer to the escape box compared to that of
the sham group (p=0.032, t=-2.493; Fig. 4.5E). This positioning likely increased the chances of
the TMS group locating the escape box with fewer errors. Secondly, compared to the TMS group,
the sham group spent more time (p=0.045, t=-1.884; Fig. 4.5F) and made more visits (p=0.008,
t=-3.313; Fig. 4.5G) around the escape box's previous location in the first trial. The patterns
observed suggest that the TMS group demonstrated a weaker adherence to the original target. This
can be linked to one of two potential reasons. One possibility is that rTMS resulted in a weaker

100
memory that could be easily reversed. However, the animals achieved the same level of
performance in the last two trials of the acquisition phase (Fig. 4.4B-F). Hence, the TMS group
did not exhibit a weaker memory compared to the sham group by the end of the acquisition.
Alternatively, an intriguing hypothesis is that rTMS enhanced the cognitive flexibility of learning
new locations at the onset of the reversal learning phase.
As trials progressed, both groups showed a rise in their average velocity (Fig. 4.5H). However,
in the final trial, the TMS group's velocity decreased in comparison to that of the sham group
(p=0.004, t=-3.657). This may imply that the TMS group might have engaged in a more
exploratory approach, possibly due to overlearning. This behavior aligns with their consistent use
of serial, rather than direct, searches to find the escape box (Fig. 4.5I). While both the TMS and
sham groups largely relied on serial searches (TMS group: 83.33±6.18%, sham group:
77.08±3.84%; Fig. 4.5J), mixed search patterns were occasionally evident in two rats from the
sham group. Nonetheless, there was no significant difference in search strategies between the two
groups (p>0.05).

101
Figure 4.5. rTMS enhanced cognitive flexibility during the reversal learning phase of the Barnes maze
task. (A) Representative trajectories of rats under rTMS (left) and sham treatments (right). Latency to find
the escape box (B), number of errors made (C), distance traveled (D), hole deviation score (E), time spent
around the previous escape box's location (F), number of visits to the previous escape box's location (G),
and average velocity (H) were statistically analyzed to compare performance between the TMS (red) and
sham (blue) groups. (I) Stacked bar charts presenting the ratios of search strategies (direct: blue, mixed:
white, serial: red) employed in each trial for the TMS (top) and sham (bottom) groups. (J) Bar plots
indicating the percentage of each search strategy (direct: left, mixed: middle, serial: right) in all trials for
the TMS (red) and sham (blue) groups. Comparisons of the TMS and sham groups were performed with ttests (∗p < 0.05, ∗∗p < 0.01, ns: no significance). Values were reported as the mean ± SEM.

102
4.3.3 rTMS strengthens memories of previous target locations in probe trials
An acquisition probe trial (Fig. 4.6A) was conducted to evaluate spatial memory retrieval one
day after the acquisition phase. The escape box was removed from the platform, and its hole was
covered like the others. No rTMS or sham treatments were administered to the animals prior to
this trial. Similarly, after the reversal learning phase, a reversal learning probe (Fig. 4.6B) was
implemented to assess the animals' adaptability to learned responses. To quantify the animals’
performance, various parameters were used, including time spent around the target zone where the
escape box was previously located (Fig. 4.1), frequency of visits to the target zone, total distance
traveled, total errors committed, and the overall search strategy.
Both the TMS (n = 6) and sham groups (n = 6) showed similar patterns in their search strategies
(Fig. 4.6Aa, 4.6Ba), using either mixed or serial searches (50%) in the acquisition probe trial (Fig.
4.6Ab), and converted to serial searches (TMS group: 100%; sham group: 83.3%) in the reversal
learning probe trial (Fig. 4.6Bb). Notably, the TMS group consistently spent more time around
the target zone compared to the sham group in both acquisition (TMS group: 20.67±2.32 s; sham
group: 12.83±2.47 s; p=0.022, t= 2.313; Fig. 4.6Ac) and reversal learning trials (TMS group:
20.67±2.78 s; sham group: 13.67±1.58 s; p= 0.027, t= 2.189; Fig. 4.6Bc). Additionally, the
frequency of visits to the target zone did not differ significantly between the groups (p>0.05; Fig.
4.6Ad, 4.6Bd), suggesting that the TMS group might have been more focused on the target zone
compared to the sham group. These results indicate that rTMS strengthened memories of previous
target locations in both probe trials, which further support the hypothesis that rTMS enhanced the
cognitive flexibility at the onset of the reversal learning phase (Fig. 4.5B-G). However, this
enhanced spatial memory did not lead to improved performance in navigating the maze, as there

103
were no significant differences in total distance traveled (Fig. 4.6Ae, 4.6Be) or errors made (Fig.
4.6Af, 4.6Bf) between the groups in both probe trials (p>0.05).
Figure 4.6. rTMS strengthened memories of previous target locations in the acquisition (A) and reversal
learning probe trials (B) of the Barnes maze task. (a) Trajectories of rats under rTMS (top) and sham
treatments (bottom) in the probe trials. (b) Stacked bar charts presenting the ratios of strategies (light grey:
mixed, dark grey: serial) employed in each trial for the TMS (left) and sham (right) groups. Time spent
around the target zone (c), number of visits to the target zone (d), total distance traveled (e), and total
number of errors made (f) were statistically analyzed to compare performance between the TMS (red) and
sham (blue) groups. Comparisons of the TMS and sham groups were performed with t-tests (∗p < 0.05, ns:
no significance). Values were reported as the mean ± SEM.

104
4.3.4 Acute effects of rTMS are not evident on spatial learning and memory
For the alternating TMS/sham group (n = 3), a modified BM protocol was implemented to
evaluate the acute effects of rTMS. Four BM sessions with alternating escape box locations were
carried out daily. Following baseline training on the first day, all animals reached the criteria
(latency≤30 s, error≤16) for both locations. In the subsequent 4-day behavioral assessments,
either rTMS or sham treatment was administered right before each of the BM sessions, which
consisted of 8 trials. This alternating approach (Fig. 4.2) was designed to counterbalance potential
sequence effects. To evaluate performance, we utilized multiple metrics for BM sessions under
both TMS (n = 24) and sham conditions (n = 24), including latency to find the escape box, error
count, traveled distance, average velocity, hole deviation score, and the employed search strategy.
BM sessions, under both TMS and sham conditions, demonstrated similar patterns in the
animals’ trajectories (Fig. 4.7A). A consistent increase in direct searches was observed as the trials
continued (Fig. 4.7B). Furthermore, the direct search was the dominant strategy in all BM trials
for both TMS (49.48±7.16%) or sham conditions (43.23±7.29%), with no significant difference
(p>0.05; Fig. 4.7C). Additional statistical analyses revealed no significant differences between
the TMS and sham sessions in terms of latency to the escape box (p>0.05; Fig. 4.7D), error count
(p>0.05; Fig. 4.7E), distance traveled (p>0.05; Fig. 4.7F), average velocity (p>0.05; Fig. 4.7G),
and hole deviation score (p>0.05; Fig. 4.7H). These findings indicate that the acute effects of rTMS
were not evident on spatial learning and memory during the BM task. One possible explanation
could be that a single rTMS session might be insufficient to induce observable changes in learning
and memory performance. Additionally, the thorough training the animals underwent to master
the task could also elevate the difficulty in distinguishing between the performances under TMS
and sham conditions.

105
Figure 4.7. Acute effects of rTMS were not evident in the Barnes maze task performance. (A)
Representative trajectories of a single rat under alternating rTMS (top) and sham stimulation (bottom). (B)
Stacked bar charts presenting the ratios of search strategies(direct: blue, mixed: white, serial: red) employed
during BM sessions for TMS (top) and sham (bottom) conditions. (C) Bar plots indicating the overall
percentage of each search strategy (direct: left, mixed: middle, serial: right) across all BM trials under rTMS
(red) and sham (blue) conditions. Latency to find the escape box (D), number of errors made (E), distance
traveled (F), average velocity (G), and hole deviation score (H) were analyzed. There were no significant
differences between the sessions under TMS (red) and sham (blue) conditions. Comparisons of behavioral
performance under TMS (red) and sham (blue) were conducted with t-tests (∗p < 0.05, ns: no significance).
Values were reported as the mean ± SEM.

106
4.4 Discussion
In this study, we for the first time reported the complex effects of high-frequency subthreshold
rTMS on various forms of cognition, from the initial acquisition of spatial information to the
flexibility in adapting to changed environments and the strength of spatial memory retention. Our
findings reveal that chronic subthreshold 10 Hz rTMS initially disrupted spatial learning in healthy
rats during the acquisition phase of the BM task. It enhanced cognitive flexibility during the
reversal learning phase and strengthened spatial memories of previous target locations in the
acquisition and reversal learning probe trials. However, the acute effects of rTMS were not evident
on spatial learning and memory.
While most studies have indicated that low-frequency rTMS (≤1 Hz) can hinder cognitive
functions (W. Li et al., 2007), high-frequency rTMS (≥5 Hz) has typically been associated with
enhancements in learning and memory performance (Shang et al., 2016; Wu et al., 2022). Contrary
to this trend, in our study, animals exhibited a decline in learning performance following 4 sessions
of subthreshold 10 Hz rTMS treatment. This underscores the cumulative nature of rTMS effects,
suggesting that multiple sessions of rTMS might be necessary for its disruptive impacts on spatial
learning to be evident. However, by the end of the 12th session, the animals' performance aligned
with that of the sham group. This outcome implies that sustained rTMS exposure might induce
neural adaptation (Benda, 2021), enabling the animals to reach performance levels comparable to
the sham group by the end of the acquisition phase.
Conversely, the enhancement of cognitive flexibility in the reversal learning phase offers a
different perspective on the neuromodulatory effects of rTMS. Flexible cognitive control involves
a complex interplay of various neural circuits and neurotransmitter pathways (Izquierdo et al.,

107
2017). It has been reported that high-frequency rTMS can induce LTP-like synaptic plasticity
(Lenz et al., 2015, 2016; Vlachos et al., 2012). The enhanced cognitive flexibility may indicate a
change of synaptic plasticity in the associated brain regions, notably the orbitofrontal cortex (OFC)
and the PFC (Uddin, 2021). Furthermore, neurotransmitter systems including serotonin, dopamine,
and glutamate (Izquierdo et al., 2017), known to be modulated by rTMS (Chervyakov et al., 2015;
M. Dubin, 2017), might also play an important role in these observed changes. Moreover, the
strengthening of memories related to previous target locations in probe trials aligns with findings
from prior studies (Shang et al., 2016; Wu et al., 2022). These studies have also shown that rTMS
promotes the upregulation of NMDA receptors and enhances the expression of BDNF, a key
molecule involved in neural plasticity related to learning and memory (Miranda et al., 2019), in
the PFC and hippocampus. These findings support the hypothesis that the enhancement of spatial
memory is attributed to changes in LTP-like plasticity induced by rTMS within memoryassociated brain regions.
For the alternating TMS/sham group, the acute effects of rTMS were not evident on spatial
learning and memory during the BM task. This absence of acute effects emphasizes the importance
of both the frequency and duration of rTMS treatments. Moreover, the intensive training regimen
that the animals underwent could have raised the challenge of differentiating between
performances under TMS and sham conditions. Similarly, during the reversal learning phase of
chronic TMS experiments, once animals identified the escape box location in the initial trial, there
was no notable difference in performance under TMS and sham conditions in subsequent trials.
This observation could be also linked to the comprehensive training received by the animals.
In conclusion, our findings underscore the complex effects of high-frequency subthreshold
rTMS on cognitive processes, revealing both potential challenges and opportunities for its

108
application in experimental and clinical settings. Further studies should investigate the cellular
and molecular mechanisms of rTMS, while also exploring different intensities, frequencies, and
durations. A comprehensive understanding of rTMS effects on cognition can pave the way for
optimizing therapeutic strategies, ensuring better outcomes in treating various neurological and
neuropsychiatric disorders.

109
Chapter 5: Conclusion and Prospects
5.1 Future development of the miniaturized TMS coil
The system presented in this work is a novel, robustly functional, and cost-efficient
neuromodulation technology for small animal TMS research. The design process and in vivo
evaluation demonstrate feasibility of use in rodent experiments. Future work concerns engineering
effort for delivering a more powerful and focal magnetic stimulation to specific brain regions in
rodents. To address this, further improvements can be made on the current design in three aspects:
1) Modifications of the driving circuit. The E-field intensity produced by our coil is still much
lower than human TMS coils. Higher coil current is required to increase the E-field intensity. This
can be done by increasing the DC power supply voltage, changing the value of sense resistors,
adding extra transistors and capacitors, etc.
2) Optimization of the TMS coil. The C-shaped TMS coil can be further optimized with the
aid of finite element modeling. Different geometries of the coils will result in different B- and Efield distributions. Comparisons can be made by varying the angle and shape of the gap in the Cshaped coil (Fig. 5.1). In addition, the windings can be replaced with litz wires to reduce the skin
effect and proximity effect losses at high frequencies.

110
Figure 5.1. Graphic illustration of C-shaped coils with different gaps. (A) Coils with different angles. (B)
Coils with an additional cut operation at the gap including minimal (left), moderate (middle), and excessive
(right) modifications.
3) Multi-channel TMS system with two or more coils. Multiple independently controlled TMS
coils can be used simultaneously to form a multi-channel TMS system. The system can deliver
focal stimulation over multiple brain regions with same or different stimulation protocols.
Additionally, it can create overlapping E-fields with specific coil placement to increase the
stimulation intensity and focality.

111
5.2 Future applications of the miniaturized TMS coil
5.2.1 Effects of rTMS on functional connectivity of neuron populations in primary
somatosensory and motor cortex
Large-scale recordings and input-output analysis of neuronal networks offer an opportunity for
identifying the functional connectivity between neurons. A multiple-input, multiple-output
(MIMO) nonlinear dynamical modeling approach (Song et al., 2013) will be used to quantity the
causal relations between pre-rTMS signals and post-rTMS signals (Fig. 5.2).
Figure 5.2. General structure of MIMO and MISO models. (A) A MIMO model consists of a series of
MISO models. (B) Each MISO model represents a biologically realistic model of input-output neuron
connections. Modified from Song et al. (2013).

112
A MEA will be implanted in the S1 or M1 in anesthetized rats. rTMS will be performed above
the sensorimotor cortex. SUAs will be recorded before and after rTMS to obtain the spike trains.
Each neuron in the S1 or M1 neuron population will serve as the output, and the rest neurons will
serve as the input to form a multiple-input, single-output (MISO) model. The MIMO model
consists of a series of MISO models. For both S1 and M1 neuron population, two MIMO models
employing pre-rTMS spike trains and post-rTMS spike trains will be constructed, respectively. In
addition, feature permutation will be applied to measure the significance of neurons by evaluating
the change of model loss before and after permutation.
The MIMO modeling approach captures nonlinear dynamical interactions and thus will provide
an accurate measure to the dependency between signals. The resulting MIMO models can be
interpreted as a quantitative mapping between pre-rTMS signals and post-rTMS signals. The
permutation importance of neurons and information rate will be assessed to evaluate the effects of
rTMS on functional connectivity within S1 and M1 neuron populations.
5.2.2 Modeling of the dynamic responses of neural activities during pseudo-random TMS
In a vast number of studies attempting to explore the effects of TMS, different stimulation
parameters have been investigated including pulse frequency, intensity, temporal pattern, etc. All
these parameters may produce different effects on neural activities. Hence, in order to achieve
precise neuromodulation, accurate MIMO modeling and informative datasets are required to
predict the effects of TMS on neural activities.

113
Figure 5.3. Graphic illustration of the closed-loop neuromodulation system.
As shown in Figure 5.3, a pseudo-random TMS pattern is designed as the input to the dynamic
MIMO model. It can stochastically switch between intensity and frequency therefore producing a
white spectrum in the input space. The TMS pattern will be delivered to the sensorimotor cortex
of anesthetized rats. LFPs and SUAs in the M1 will be recorded via a MEA during TMS. Signal
processing including artifact suppression, filtering, spike sorting, and LFP power calculation will
be performed on the raw signals. LFP power features and spike trains will be extracted to serve
as the output of the dynamic MIMO model. By predicting the real-time response of neural
activities during TMS, the MIMO model can be used to optimize the stimulation parameters. It
will facilitate the development of a closed-loop neuromodulation system to achieve desired
therapeutic effects in a variety of neurological and neuropsychiatric disorders.

114
5.2.3 Effects of rTMS on working memory in a rodent DNMS task
EEG is a method to record macroscopic cortical activities from the scalp. The combination of
rTMS and EEG can measure the perturbation induced by rTMS on cortical activities at both local
and network levels, which provides a unique means of studying the causal role of specific brain
regions in behaving animals. A customized parylene-based EEG probe with multiple electrodes
is designed to be used in conjunction with rTMS. The specific layout of 32 recording electrodes
covers the M1, M2, S1, posterior parietal cortex (PPC), and primary visual cortex (V1) of rats as
shown in Figure 5.4.
Figure 5.4. EEG recordingsfrom different brain regions of rats via a customized parylene-based EEG probe.
The EEG probe will be chronically implanted in rats. After the animal is recovered from
surgery, EEG signals will be monitored while the animal performs a delayed-nonmatch-to-sample
(DNMS) memory task (Berger et al., 2011) in a chamber (Fig. 5.5). In brief, the DNMS task will

115
consist of three phases: (1) sample phase: in which a single lever is presented randomly in either
the left or right position; when the animal presses the lever, the event is classified as a sample
response (SR), (2) delay phase: of variable duration (1–30 s) in which a nosepoke (NP) into a
photocell is required to advance to the (3) nonmatch phase: where both levers are presented and a
response on the lever opposite to the SR, i.e., a nonmatch response (NR), is required for delivery
of a drop of water in the trough. A response in the nonmatch phase on the lever in the same
position as the SR (match response) constitutes an ‘error’ with no water delivery and light turn-off
in the chamber for 5 s (a conditioned punishment). Following the reward delivery (or an error),
the levers are retracted for 10 s before the sample lever resets for the next trial. Individual
performance will be assessed as % correct NRs with respect to the total number of trials (100–150)
per daily (1–2 h) session.
Figure 5.5. Diagram of DNMS task including sample phase, delay phase, and two consequences in the
nonmatch phase.

116
Animals will be under food restriction throughout the experiment. Following a week of
restraint habituation, animals will receive chronic rTMS or sham stimulation during behavioral
training. rTMS or sham stimulation will be given immediately before each training session. At
different points of the experiments, brain tissue will be extracted for neuroplasticity marker
molecular analysis. In addition, neural activities recorded from multiple brain regions together
with memory performance will be analyzed to investigate when and how neural activities in one
brain region influence activities in other regions. Most importantly, the effects of rTMS on
working memory will be evaluated from the network to behavioral level.

117
References
Aberra, A. S., Wang, B., Grill, W. M., & Peterchev, A. v. (2020). Simulation of transcranial magnetic
stimulation in head model with morphologically-realistic cortical neurons. Brain Stimulation, 13(1),
175–189. https://doi.org/10.1016/j.brs.2019.10.002
Ahdab, R., Ayache, S. S., Brugières, P., Goujon, C., & Lefaucheur, J.-P. (2010). Comparison of
“standard” and “navigated” procedures of TMS coil positioning over motor, premotor and prefrontal
targets in patients with chronic pain and depression. Neurophysiologie Clinique/Clinical
Neurophysiology, 40(1), 27–36. https://doi.org/10.1016/j.neucli.2010.01.001
Ahmed, Z., & Wieraszko, A. (2006). Modulation of learning and hippocampal, neuronal plasticity by
repetitive transcranial magnetic stimulation (rTMS). Bioelectromagnetics, 27(4), 288–294.
https://doi.org/10.1002/bem.20211
Akeju, O., Song, A. H., Hamilos, A. E., Pavone, K. J., Flores, F. J., Brown, E. N., & Purdon, P. L. (2016).
Electroencephalogram signatures of ketamine anesthesia-induced unconsciousness. Clinical
Neurophysiology, 127(6), 2414–2422. https://doi.org/10.1016/j.clinph.2016.03.005
Alekseichuk, I., Mantell, K., Shirinpour, S., & Opitz, A. (2019). Comparative modeling of transcranial
magnetic and electric stimulation in mouse, monkey, and human. NeuroImage, 194, 136–148.
https://doi.org/10.1016/j.neuroimage.2019.03.044
Andrade, C. (2017). Ketamine for depression, 4: in what dose, at what rate, by what route, for how long,
and at what frequency? The Journal of Clinical Psychiatry, 78(7), 10106.
https://doi.org/10.4088/JCP.17f11738
Avoli, M., Biagini, G., & de Curtis, M. (2006). Do Interictal Spikes Sustain Seizures and
Epileptogenesis? Epilepsy Currents, 6(6), 203–207. https://doi.org/10.1111/j.1535-
7511.2006.00146.x
Aydin-Abidin, S., Trippe, J., Funke, K., Eysel, U. T., & Benali, A. (2008). High- and low-frequency
repetitive transcranial magnetic stimulation differentially activates c-Fos and zif268 protein
expression in the rat brain. Experimental Brain Research, 188(2), 249–261.
https://doi.org/10.1007/s00221-008-1356-2
Bagherzadeh, H., Meng, Q., Lu, H., Hong, E., Yang, Y., & Choa, F.-S. (2022). High-Performance
Magnetic-core Coils for Targeted Rodent Brain Stimulations. BME Frontiers, 2022, 1–11.
https://doi.org/10.34133/2022/9854846
Bak, L. K., Schousboe, A., & Waagepetersen, H. S. (2006). The glutamate/GABA-glutamine cycle:
aspects of transport, neurotransmitter homeostasis and ammonia transfer. Journal of
Neurochemistry, 98(3), 641–653. https://doi.org/10.1111/j.1471-4159.2006.03913.x
Ballard, E. D., & Zarate, C. A. (2020). The role of dissociation in ketamine’s antidepressant effects.
Nature Communications, 11(1), 1–7. https://doi.org/10.1038/s41467-020-20190-4
Barahona-Corrêa, J. B., Velosa, A., Chainho, A., Lopes, R., & Oliveira-Maia, A. J. (2018). Repetitive
Transcranial Magnetic Stimulation for Treatment of Autism Spectrum Disorder: A Systematic
Review and Meta-Analysis. Frontiers in Integrative Neuroscience, 12.
https://doi.org/10.3389/fnint.2018.00027
Barker, A. T. (1991). An Introduction to the Basic Principles of Magnetic Nerve Stimulation. Journal of
Clinical Neurophysiology, 8(1), 26–37. https://doi.org/10.1097/00004691-199101000-00005

118
Barker, A. T., Jalinous, R., & Freeston, I. L. (1985). Non-invasive magnetic stimulation of human motor
cortex. The Lancet, 325(8437), 1106–1107. https://doi.org/10.1016/S0140-6736(85)92413-4
Barnes, C. A. (1979). Memory deficits associated with senescence: A neurophysiological and behavioral
study in the rat. Journal of Comparative and Physiological Psychology, 93(1), 74–104.
https://doi.org/10.1037/h0077579
Bathina, S., & Das, U. N. (2015). Brain-derived neurotrophic factor and its clinical implications. Archives
of Medical Science, 6, 1164–1178. https://doi.org/10.5114/aoms.2015.56342
Bazley, F. A., Hu, C., Maybhate, A., Pourmorteza, A., Pashai, N., Thakor, N. v., Kerr, C. L., & All, A. H.
(2012). Electrophysiological evaluation of sensory and motor pathways after incomplete unilateral
spinal cord contusion. Journal of Neurosurgery: Spine, 16(4), 414–423.
https://doi.org/10.3171/2012.1.SPINE11684
Belujon, P., & Grace, A. A. (2017). Dopamine System Dysregulation in Major Depressive Disorders.
International Journal of Neuropsychopharmacology, 20(12), 1036–1046.
https://doi.org/10.1093/ijnp/pyx056
Benda, J. (2021). Neural adaptation. Current Biology, 31(3), R110–R116.
https://doi.org/10.1016/j.cub.2020.11.054
Berger, T. W., Hampson, R. E., Song, D., Goonawardena, A., Marmarelis, V. Z., & Deadwyler, S. A.
(2011). A cortical neural prosthesis for restoring and enhancing memory. Journal of Neural
Engineering, 8(4), 046017. https://doi.org/10.1088/1741-2560/8/4/046017
Bergmann, T. O., Karabanov, A., Hartwigsen, G., Thielscher, A., & Siebner, H. R. (2016). Combining
non-invasive transcranial brain stimulation with neuroimaging and electrophysiology: Current
approaches and future perspectives. NeuroImage, 140, 4–19.
https://doi.org/10.1016/j.neuroimage.2016.02.012
Best, S. R. D., & Griffin, B. (2015). Combination therapy utilizing ketamine and transcranial magnetic
stimulation for treatment-resistant depression: a case report. International Journal of Neuroscience,
125(3), 232–234. https://doi.org/10.3109/00207454.2014.933834
Bestmann, S., & Krakauer, J. W. (2015). The uses and interpretations of the motor-evoked potential for
understanding behaviour. Experimental Brain Research, 233(3), 679–689.
https://doi.org/10.1007/s00221-014-4183-7
Bliss, T. V. P., & Gardner-Medwin, A. R. (1973). Long-lasting potentiation of synaptic transmission in
the dentate area of the unanaesthetized rabbit following stimulation of the perforant path. The
Journal of Physiology, 232(2), 357–374. https://doi.org/10.1113/jphysiol.1973.sp010274
Blumberger, D. M., Mulsant, B. H., Fitzgerald, P. B., Rajji, T. K., Ravindran, A. v., Young, L. T.,
Levinson, A. J., & Daskalakis, Z. J. (2012). A randomized double-blind sham-controlled
comparison of unilateral and bilateral repetitive transcranial magnetic stimulation for treatmentresistant major depression. The World Journal of Biological Psychiatry, 13(6), 423–435.
https://doi.org/10.3109/15622975.2011.579163
Blumenfeld, R. S., & Ranganath, C. (2007). Prefrontal Cortex and Long-Term Memory Encoding: An
Integrative Review of Findings from Neuropsychology and Neuroimaging. The Neuroscientist,
13(3), 280–291. https://doi.org/10.1177/1073858407299290
Boinagrov, D., Pangratz-Fuehrer, S., Suh, B., Mathieson, K., Naik, N., & Palanker, D. (2012). Upper
threshold of extracellular neural stimulation. Journal of Neurophysiology, 108(12), 3233–3238.
https://doi.org/10.1152/jn.01058.2011

119
Boonzaier, J., Petrov, P. I., Otte, W. M., Smirnov, N., Neggers, S. F. W., & Dijkhuizen, R. M. (2020).
Design and Evaluation of a Rodent-Specific Transcranial Magnetic Stimulation Coil: An In Silico
and In Vivo Validation Study. Neuromodulation: Technology at the Neural Interface, 23(3), 324–
334. https://doi.org/10.1111/ner.13025
Bortoletto, M., Veniero, D., Thut, G., & Miniussi, C. (2015). The contribution of TMS–EEG
coregistration in the exploration of the human cortical connectome. Neuroscience & Biobehavioral
Reviews, 49, 114–124. https://doi.org/10.1016/j.neubiorev.2014.12.014
Boutros, N. N., Gueorguieva, R., Hoffman, R. E., Oren, D. A., Feingold, A., & Berman, R. M. (2002).
Lack of a therapeutic effect of a 2-week sub-threshold transcranial magnetic stimulation course for
treatment-resistant depression. Psychiatry Research, 113(3), 245–254.
https://doi.org/10.1016/S0165-1781(02)00267-6
Brasil-Neto, J. P., Cohen, L. G., Panizza, M., Nilsson, J., Roth, B. J., & Hallett, M. (1992). Optimal Focal
Transcranial Magnetic Activation of the Human Motor Cortex. Journal of Clinical
Neurophysiology, 9(1), 132–136. https://doi.org/10.1097/00004691-199201000-00014
Brunelin, J., Poulet, E., Bediou, B., Kallel, L., Dalery, J., D’amato, T., & Saoud, M. (2006). Low
frequency repetitive transcranial magnetic stimulation improves source monitoring deficit in
hallucinating patients with schizophrenia. Schizophrenia Research, 81(1), 41–45.
https://doi.org/10.1016/j.schres.2005.10.009
Brunoni, A. R., & Vanderhasselt, M.-A. (2014). Working memory improvement with non-invasive brain
stimulation of the dorsolateral prefrontal cortex: A systematic review and meta-analysis. Brain and
Cognition, 86, 1–9. https://doi.org/10.1016/j.bandc.2014.01.008
Burek, M., Follmann, R., & Rosa, E. (2019). Temperature effects on neuronal firing rates and tonic-tobursting transitions. Biosystems, 180, 1–6. https://doi.org/10.1016/j.biosystems.2019.03.003
Buzsaki, G., & Draguhn, A. (2004). Neuronal oscillations in cortical networks. Science, 304(5679),
1926–1929. https://doi.org/doi.org/10.1126/science.1099745
Capocchi, G., Zampolini, M., & Larson, J. (1992). Theta burst stimulation is optimal for induction of LTP
at both apical and basal dendritic synapses on hippocampal CA1 neurons. Brain Research, 591(2),
332–336. https://doi.org/10.1016/0006-8993(92)91715-Q
Carmi, L., Alyagon, U., Barnea-Ygael, N., Zohar, J., Dar, R., & Zangen, A. (2018). Clinical and
electrophysiological outcomes of deep TMS over the medial prefrontal and anterior cingulate
cortices in OCD patients. Brain Stimulation, 11(1), 158–165.
https://doi.org/10.1016/j.brs.2017.09.004
Carmi, L., Tendler, A., Bystritsky, A., Hollander, E., Blumberger, D. M., Daskalakis, J., Ward, H.,
Lapidus, K., Goodman, W., Casuto, L., Feifel, D., Barnea-Ygael, N., Roth, Y., Zangen, A., & Zohar,
J. (2019). Efficacy and Safety of Deep Transcranial Magnetic Stimulation for ObsessiveCompulsive Disorder: A Prospective Multicenter Randomized Double-Blind Placebo-Controlled
Trial. American Journal of Psychiatry, 176(11), 931–938.
https://doi.org/10.1176/appi.ajp.2019.18101180
Castro-Alamancos, M. A., Donoghue, J. P., & Connors, B. W. (1995). Different forms of synaptic
plasticity in somatosensory and motor areas of the neocortex. Journal of Neuroscience, 15(7), 5324–
5333. https://doi.org/10.1523/jneurosci.15-07-05324.1995
Cermak, S., Meng, Q., Peng, K., Baldwin, S., Mejías-Aponte, C. A., Yang, Y., & Lu, H. (2020). Focal
transcranial magnetic stimulation in awake rats: Enhanced glucose uptake in deep cortical layers.
Journal of Neuroscience Methods, 339, 108709. https://doi.org/10.1016/j.jneumeth.2020.108709

120
Chauvette, S., Crochet, S., Volgushev, M., & Timofeev, I. (2011). Properties of slow oscillation during
slow-wave sleep and anesthesia in cats. Journal of Neuroscience, 31(42), 14998–15008.
https://doi.org/10.1523/JNEUROSCI.2339-11.2011
Chen, J., Zhou, C., Wu, B., Wang, Y., Li, Q., Wei, Y., Yang, D., Mu, J., Zhu, D., Zou, D., & Xie, P.
(2013). Left versus right repetitive transcranial magnetic stimulation in treating major depression: A
meta-analysis of randomised controlled trials. Psychiatry Research, 210(3), 1260–1264.
https://doi.org/10.1016/j.psychres.2013.09.007
Chen, R., Classen, J., Gerloff, C., Celnik, P., Wassermann, E. M., Hallett, M., & Cohen, L. G. (1997).
Depression of motor cortex excitability by low-frequency transcranial magnetic stimulation.
Neurology, 48(5), 1398–1403. https://doi.org/10.1212/WNL.48.5.1398
Chervyakov, A. V., Chernyavsky, A. Yu., Sinitsyn, D. O., & Piradov, M. A. (2015). Possible
Mechanisms Underlying the Therapeutic Effects of Transcranial Magnetic Stimulation. Frontiers in
Human Neuroscience, 9. https://doi.org/10.3389/fnhum.2015.00303
Christiansen, M. G., Howe, C. M., Bono, D. C., Perreault, D. J., & Anikeeva, P. (2017). Practical
methods for generating alternating magnetic fields for biomedical research. Review of Scientific
Instruments, 88(8), 084301. https://doi.org/10.1063/1.4999358
Christo, P., & Bottros, M. (2014). Current perspectives on intrathecal drug delivery. Journal of Pain
Research, 615. https://doi.org/10.2147/JPR.S37591
Chung, S. W., Hoy, K. E., & Fitzgerald, P. B. (2015). Theta-burst stimulation: a new form of TMS
treatment for depression? Depression and Anxiety, 32(3), 182–192. https://doi.org/10.1002/da.22335
Chunye Ren, Tarjan, P. P., & Popovic, D. B. (1995). A novel electric design for electromagnetic
stimulation-the Slinky coil. IEEE Transactions on Biomedical Engineering, 42(9), 918–925.
https://doi.org/10.1109/10.412658
Cobos Sánchez, C., Jurado García, J. J., Ruiz Cabello, M., & Fernández Pantoja, M. (2020). Design of
coils for lateralized TMS on mice. Journal of Neural Engineering, 17(3), 036007.
https://doi.org/10.1088/1741-2552/ab89fe
Cohen, D., & Cuffin, B. N. (1991). Developing a More Focal Magnetic Stimulator. Part I. Journal of
Clinical Neurophysiology, 8(1), 102–111. https://doi.org/10.1097/00004691-199101000-00013
Cohen, L. G., Roth, B. J., Nilsson, J., Dang, N., Panizza, M., Bandinelli, S., Friauf, W., & Hallett, M.
(1990). Effects of coil design on delivery of focal magnetic stimulation. Technical considerations.
Electroencephalography and Clinical Neurophysiology, 75(4), 350–357.
https://doi.org/10.1016/0013-4694(90)90113-X
Cole, S. R., & Voytek, B. (2017). Brain Oscillations and the Importance of Waveform Shape. Trends in
Cognitive Sciences, 21(2), 137–149. https://doi.org/10.1016/j.tics.2016.12.008
Cooper, Y. A., Pianka, S. T., Alotaibi, N. M., Babayan, D., Salavati, B., Weil, A. G., Ibrahim, G. M.,
Wang, A. C., & Fallah, A. (2018). Repetitive transcranial magnetic stimulation for the treatment of
drug-resistant epilepsy: A systematic review and individual participant data meta-analysis of realworld evidence. Epilepsia Open, 3(1), 55–65. https://doi.org/10.1002/epi4.12092
Coppola, G., de Pasqua, V., Pierelli, F., & Schoenen, J. (2012). Effects of repetitive transcranial magnetic
stimulation on somatosensory evoked potentials and high frequency oscillations in migraine.
Cephalalgia, 32(9), 700–709. https://doi.org/10.1177/0333102412446313
Corlier, J., Wilson, A., Hunter, A. M., Vince-Cruz, N., Krantz, D., Levitt, J., Minzenberg, M. J., Ginder,
N., Cook, I. A., & Leuchter, A. F. (2019). Changes in Functional Connectivity Predict Outcome of

121
Repetitive Transcranial Magnetic Stimulation Treatment of Major Depressive Disorder. Cerebral
Cortex, 29(12), 4958–4967. https://doi.org/10.1093/cercor/bhz035
Croarkin, P. E., Levinson, A. J., & Daskalakis, Z. J. (2011). Evidence for GABAergic inhibitory deficits
in major depressive disorder. Neuroscience & Biobehavioral Reviews, 35(3), 818–825.
https://doi.org/10.1016/j.neubiorev.2010.10.002
Croarkin, P. E., Nakonezny, P. A., Wall, C. A., Murphy, L. L., Sampson, S. M., Frye, M. A., & Port, J. D.
(2016). Transcranial magnetic stimulation potentiates glutamatergic neurotransmission in depressed
adolescents. Psychiatry Research: Neuroimaging, 247, 25–33.
https://doi.org/10.1016/j.pscychresns.2015.11.005
Curtis, C. E., & D’Esposito, M. (2003). Persistent activity in the prefrontal cortex during working
memory. Trends in Cognitive Sciences, 7(9), 415–423. https://doi.org/10.1016/S1364-
6613(03)00197-9
Davila, M. C., Hess, A., Louviere, A., & Manzardo, A. M. (2021). KETAMINE THERAPY COMBINED
WITH REPETITIVE TRANSCRANIAL MAGNETIC STIMULATION (rTMS) FOR MAJOR
DEPRESSIVE DISORDER AT A SUBURBAN TERTIARY CLINIC. Brain Stimulation: Basic,
Translational, and Clinical Research in Neuromodulation, 14(5), 1404–1405.
https://doi.org/10.1016/j.brs.2021.07.023
Davis, N. J., Tomlinson, S. P., & Morgan, H. M. (2012). The role of beta-frequency neural oscillations in
motor control. Journal of Neuroscience, 32(2), 403–404.
https://doi.org/10.1523/JNEUROSCI.5106-11.2012
de Tommaso, M., Ambrosini, A., Brighina, F., Coppola, G., Perrotta, A., Pierelli, F., Sandrini, G.,
Valeriani, M., Marinazzo, D., Stramaglia, S., & Schoenen, J. (2014). Altered processing of sensory
stimuli in patients with migraine. Nature Reviews Neurology, 10(3), 144–155.
https://doi.org/10.1038/nrneurol.2014.14
Deng, Z.-D., Lisanby, S. H., & Peterchev, A. v. (2013). Electric field depth–focality tradeoff in
transcranial magnetic stimulation: Simulation comparison of 50 coil designs. Brain Stimulation,
6(1), 1–13. https://doi.org/10.1016/j.brs.2012.02.005
di Lazzaro, V., Pilato, F., Saturno, E., Oliviero, A., Dileone, M., Mazzone, P., Insola, A., Tonali, P. A.,
Ranieri, F., Huang, Y. Z., & Rothwell, J. C. (2005). Theta-burst repetitive transcranial magnetic
stimulation suppresses specific excitatory circuits in the human motor cortex. The Journal of
Physiology, 565(3), 945–950. https://doi.org/10.1113/jphysiol.2005.087288
Ding, L., Shou, G., Yuan, H., Urbano, D., & Cha, Y.-H. (2014). Lasting modulation effects of rTMS on
neural activity and connectivity as revealed by resting-state EEG. IEEE Transactions on Biomedical
Engineering, 61(7), 2070–2080. https://doi.org/10.1109/TBME.2014.2313575
Dubin, M. (2017). Imaging TMS: antidepressant mechanisms and treatment optimization. International
Review of Psychiatry, 29(2), 89–97. https://doi.org/10.1080/09540261.2017.1283297
Dubin, M. J., Mao, X., Banerjee, S., Goodman, Z., Lapidus, K. A. B., Kang, G., Liston, C., & Shungu, D.
C. (2016). Elevated prefrontal cortex GABA in patients with major depressive disorder after TMS
treatment measured with proton magnetic resonance spectroscopy. Journal of Psychiatry and
Neuroscience, 41(3), E37–E45. https://doi.org/10.1503/jpn.150223
Duecker, F., & Sack, A. T. (2015). Rethinking the role of sham TMS. Frontiers in Psychology, 6.
https://doi.org/10.3389/fpsyg.2015.00210
Engel, A. K., & Singer, W. (2001). Temporal binding and the neural correlates of sensory awareness.
Trends in Cognitive Sciences, 5(1), 16–25. https://doi.org/10.1016/S1364-6613(00)01568-0

122
Enticott, P. G., Rinehart, N. J., Tonge, B. J., Bradshaw, J. L., & Fitzgerald, P. B. (2012). Repetitive
transcranial magnetic stimulation (rTMS) improves movement-related cortical potentials in autism
spectrum disorders. Brain Stimulation, 5(1), 30–37. https://doi.org/10.1016/j.brs.2011.02.001
Epstein, C. M., & Davey, K. R. (2002). Iron-Core Coils for Transcranial Magnetic Stimulation. Journal
of Clinical Neurophysiology, 19(4), 376–381. https://doi.org/10.1097/00004691-200208000-00010
Estrada, C., Fernández-Gómez, F. J., López, D., Gonzalez-Cuello, A., Tunez, I., Toledo, F., Blin, O.,
Bordet, R., Richardson, J. C., Fernandez-Villalba, E., & Herrero, M. T. (2015). Transcranial
magnetic stimulation and aging: Effects on spatial learning and memory after sleep deprivation in
Octodon degus. Neurobiology of Learning and Memory, 125, 274–281.
https://doi.org/10.1016/j.nlm.2015.09.011
Farmer, S. (1998). Rhythmicity, synchronization and binding in human and primate motor systems. The
Journal of Physiology, 509(1), 3–14. https://doi.org/10.1111/j.1469-7793.1998.003bo.x
Farzan, F., Vernet, M., Shafi, M. M. D., Rotenberg, A., Daskalakis, Z. J., & Pascual-Leone, A. (2016).
Characterizing and Modulating Brain Circuitry through Transcranial Magnetic Stimulation
Combined with Electroencephalography. Frontiers in Neural Circuits, 10.
https://doi.org/10.3389/fncir.2016.00073
Fehlings, M. G., Tator, C. H., Linden, R. D., & Piper, I. R. (1988). Motor and somatosensory evoked
potentials recorded from the rat. Electroencephalography and Clinical Neurophysiology, 69(1), 65–
78. https://doi.org/10.1016/0013-4694(88)90036-3
Fiáth, R., Kerekes, B. P., Wittner, L., Tóth, K., Beregszászi, P., Horváth, D., & Ulbert, I. (2016). Laminar
analysis of the slow wave activity in the somatosensory cortex of anesthetized rats. European
Journal of Neuroscience, 44(3), 1935–1951. https://doi.org/10.1111/ejn.13274
Fisicaro, F., Lanza, G., Grasso, A. A., Pennisi, G., Bella, R., Paulus, W., & Pennisi, M. (2019). Repetitive
transcranial magnetic stimulation in stroke rehabilitation: review of the current evidence and pitfalls.
Therapeutic Advances in Neurological Disorders, 12, 175628641987831.
https://doi.org/10.1177/1756286419878317
Fitzgerald, K. D., Welsh, R. C., Gehring, W. J., Abelson, J. L., Himle, J. A., Liberzon, I., & Taylor, S. F.
(2005). Error-related hyperactivity of the anterior cingulate cortex in obsessive-compulsive disorder.
Biological Psychiatry, 57(3), 287–294. https://doi.org/10.1016/j.biopsych.2004.10.038
Fries, P., Reynolds, J. H., Rorie, A. E., & Desimone, R. (2001). Modulation of oscillatory neuronal
synchronization by selective visual attention. Science, 291(5508), 1560–1563.
https://doi.org/10.1126/science.1055465
Galletly, C., Gill, S., Clarke, P., Burton, C., & Fitzgerald, P. B. (2012). A randomized trial comparing
repetitive transcranial magnetic stimulation given 3 days/week and 5 days/week for the treatment of
major depression: is efficacy related to the duration of treatment or the number of treatments?
Psychological Medicine, 42(5), 981–988. https://doi.org/10.1017/S0033291711001760
Gao, F., Wang, S., Guo, Y., Wang, J., Lou, M., Wu, J., Ding, M., Tian, M., & Zhang, H. (2010).
Protective effects of repetitive transcranial magnetic stimulation in a rat model of transient cerebral
ischaemia: a microPET study. European Journal of Nuclear Medicine and Molecular Imaging,
37(5), 954–961. https://doi.org/10.1007/s00259-009-1342-3
Gawel, K., Gibula, E., Marszalek-Grabska, M., Filarowska, J., & Kotlinska, J. H. (2019). Assessment of
spatial learning and memory in the Barnes maze task in rodents—methodological consideration.
Naunyn-Schmiedeberg’s Archives of Pharmacology, 392(1), 1–18. https://doi.org/10.1007/s00210-
018-1589-y

123
George, M. S., Ketter, T. A., & Post, R. M. (1994). Prefrontal cortex dysfunction in clinical depression.
Depression, 2(2), 59–72. https://doi.org/10.1002/depr.3050020202
George, M. S., Wassermann, E. M., Williams, W. A., Callahan, A., Ketter, T. A., Basser, P., Hallett, M.,
& Post, R. M. (1995). Daily repetitive transcranial magnetic stimulation (rTMS) improves mood in
depression. NeuroReport, 6(14), 1853–1856. https://doi.org/10.1097/00001756-199510020-00008
Guo, F., Han, X., Zhang, J., Zhao, X., Lou, J., Chen, H., & Huang, X. (2014). Repetitive Transcranial
Magnetic Stimulation Promotes Neural Stem Cell Proliferation via the Regulation of MiR-25 in a
Rat Model of Focal Cerebral Ischemia. PLoS ONE, 9(10), e109267.
https://doi.org/10.1371/journal.pone.0109267
Guse, B., Falkai, P., & Wobrock, T. (2010). Cognitive effects of high-frequency repetitive transcranial
magnetic stimulation: a systematic review. Journal of Neural Transmission, 117(1), 105–122.
https://doi.org/10.1007/s00702-009-0333-7
Hallett, M. (2007). Transcranial Magnetic Stimulation: A Primer. Neuron, 55(2), 187–199.
https://doi.org/10.1016/j.neuron.2007.06.026
Hampe, C. S., Mitoma, H., & Manto, M. (2018). GABA and Glutamate: Their transmitter role in the CNS
and pancreatic islets. U: GABA And Glutamate-New Developments In Neurotransmission Research.
Samardzic J, Ur., London, IntechOpen, 65–90. https://doi.org/10.5772/intechopen.70958
Harris, K. D., & Shepherd, G. M. G. (2015). The neocortical circuit: themes and variations. Nature
Neuroscience, 18(2), 170–181. https://doi.org/10.1038/nn.3917
Herbsman, T., Forster, L., Molnar, C., Dougherty, R., Christie, D., Koola, J., Ramsey, D., Morgan, P. S.,
Bohning, D. E., George, M. S., & Nahas, Z. (2009). Motor threshold in transcranial magnetic
stimulation: The impact of white matter fiber orientation and skull-to-cortex distance. Human Brain
Mapping, 30(7), 2044–2055. https://doi.org/10.1002/hbm.20649
Herrmann, M. J., Römmler, J., Ehlis, A.-C., Heidrich, A., & Fallgatter, A. J. (2004). Source localization
(LORETA) of the error-related-negativity (ERN/Ne) and positivity (Pe). Cognitive Brain Research,
20(2), 294–299. https://doi.org/10.1016/j.cogbrainres.2004.02.013
Herwig, U., Padberg, F., Unger, J., Spitzer, M., & Schönfeldt-Lecuona, C. (2001). Transcranial magnetic
stimulation in therapy studies: examination of the reliability of “standard” coil positioning by
neuronavigation. Biological Psychiatry, 50(1), 58–61. https://doi.org/10.1016/S0006-
3223(01)01153-2
Hong, L. E., Summerfelt, A., Buchanan, R. W., O’donnell, P., Thaker, G. K., Weiler, M. A., & Lahti, A.
C. (2010). Gamma and delta neural oscillations and association with clinical symptoms under
subanesthetic ketamine. Neuropsychopharmacology, 35(3), 632–640.
https://doi.org/10.1038/npp.2009.168
Hoogendam, J. M., Ramakers, G. M. J., & di Lazzaro, V. (2010). Physiology of repetitive transcranial
magnetic stimulation of the human brain. Brain Stimulation, 3(2), 95–118.
https://doi.org/10.1016/j.brs.2009.10.005
Horváth, C., Tóth, L. F., Ulbert, I., & Fiáth, R. (2021). Dataset of cortical activity recorded with high
spatial resolution from anesthetized rats. Scientific Data, 8(1), 1–14. https://doi.org/10.1038/s41597-
021-00970-3
Huang, G., & Mouraux, A. (2015). MEP Latencies Predict the Neuromodulatory Effect of cTBS
Delivered to the Ipsilateral and Contralateral Sensorimotor Cortex. PLOS ONE, 10(8), e0133893.
https://doi.org/10.1371/journal.pone.0133893

124
Huang, Y.-Z., Chen, R.-S., Rothwell, J. C., & Wen, H.-Y. (2007). The after-effect of human theta burst
stimulation is NMDA receptor dependent. Clinical Neurophysiology, 118(5), 1028–1032.
https://doi.org/10.1016/j.clinph.2007.01.021
Huang, Y.-Z., Edwards, M. J., Rounis, E., Bhatia, K. P., & Rothwell, J. C. (2005). Theta Burst
Stimulation of the Human Motor Cortex. Neuron, 45(2), 201–206.
https://doi.org/10.1016/j.neuron.2004.12.033
Izquierdo, A., Brigman, J. L., Radke, A. K., Rudebeck, P. H., & Holmes, A. (2017). The neural basis of
reversal learning: An updated perspective. Neuroscience, 345, 12–26.
https://doi.org/10.1016/j.neuroscience.2016.03.021
Janssen, A. M., Oostendorp, T. F., & Stegeman, D. F. (2015). The coil orientation dependency of the
electric field induced by TMS for M1 and other brain areas. Journal of NeuroEngineering and
Rehabilitation, 12(1), 47. https://doi.org/10.1186/s12984-015-0036-2
Jiang, W., Isenhart, R., Kistler, N., Lu, Z., Xu, H., Lee, D. J., Liu, C. Y., & Song, D. (2021). Low
Intensity Repetitive Transcranial Magnetic Stimulation Modulates Spontaneous Spiking Activities
in Rat Cortex. 2021 43rd Annual International Conference of the IEEE Engineering in Medicine &
Biology Society (EMBC), 6318–6321. https://doi.org/10.1109/EMBC46164.2021.9630986
Jiang, W., Isenhart, R., Liu, C. Y., & Song, D. (2023). A C-shaped miniaturized coil for transcranial
magnetic stimulation in rodents. Journal of Neural Engineering, 20(2), 026022.
https://doi.org/10.1088/1741-2552/acc097
Jiang, W., Isenhart, R., Sutherland, R., Lu, Z., Xu, H., Pace, J., Bonaguidi, M. A., Lee, D. J., Liu, C. Y.,
& Song, D. (2022). Subthreshold repetitive transcranial magnetic stimulation suppresses ketamineinduced poly population spikes in rat sensorimotor cortex. Frontiers in Neuroscience, 16.
https://doi.org/10.3389/fnins.2022.998704
Kalita, J., Bhoi, S. K., & Misra, U. K. (2017). Effect of high rate rTMS on somatosensory evoked
potential in migraine. Cephalalgia, 37(13), 1222–1230. https://doi.org/10.1177/0333102416675619
Kaskie, R. E., & Ferrarelli, F. (2018). Investigating the neurobiology of schizophrenia and other major
psychiatric disorders with Transcranial Magnetic Stimulation. Schizophrenia Research, 192, 30–38.
https://doi.org/10.1016/j.schres.2017.04.045
Ketter, T. A., George, M. S., Kimbrell, T. A., Benson, B. E., & Post, R. M. (1996). Functional Brain
Imaging, Limbic Function, and Affective Disorders. The Neuroscientist, 2(1), 55–65.
https://doi.org/10.1177/107385849600200113
Khokhar, F. A., Voss, L. J., Steyn-Ross, D. A., & Wilson, M. T. (2021). Design and Demonstration In
Vitro of a Mouse-Specific Transcranial Magnetic Stimulation Coil. IEEE Transactions on
Magnetics, 57(7), 1–11. https://doi.org/10.1109/TMAG.2021.3077976
Kim, C., Johnson, N. F., Cilles, S. E., & Gold, B. T. (2011). Common and Distinct Mechanisms of
Cognitive Flexibility in Prefrontal Cortex. The Journal of Neuroscience, 31(13), 4771–4779.
https://doi.org/10.1523/JNEUROSCI.5923-10.2011
Kim, D.-H., Georghiou, G. E., & Won, C. (2006). Improved Field Localization in Transcranial Magnetic
Stimulation of the Brain With the Utilization of a Conductive Shield Plate in the Stimulator. IEEE
Transactions on Biomedical Engineering, 53(4), 720–725.
https://doi.org/10.1109/TBME.2006.870244
Kirschen, M. P., Davis-Ratner, M. S., Jerde, T. E., Schraedley-Desmond, P., & Desmond, J. E. (2006).
Enhancement of Phonological Memory Following Transcranial Magnetic Stimulation (TMS).
Behavioural Neurology, 17(3–4), 187–194. https://doi.org/10.1155/2006/469132

125
Klomjai, W., Katz, R., & Lackmy-Vallée, A. (2015). Basic principles of transcranial magnetic stimulation
(TMS) and repetitive TMS (rTMS). Annals of Physical and Rehabilitation Medicine, 58(4), 208–
213. https://doi.org/10.1016/j.rehab.2015.05.005
Koponen, L. M., Goetz, S. M., Tucci, D. L., & Peterchev, A. v. (2020). Sound comparison of seven TMS
coils at matched stimulation strength. Brain Stimulation, 13(3), 873–880.
https://doi.org/10.1016/j.brs.2020.03.004
Koponen, L. M., Nieminen, J. O., Mutanen, T. P., Stenroos, M., & Ilmoniemi, R. J. (2017). Coil
optimisation for transcranial magnetic stimulation in realistic head geometry. Brain Stimulation,
10(4), 795–805. https://doi.org/10.1016/j.brs.2017.04.001
Koponen, L. M., Stenroos, M., Nieminen, J. O., Jokivarsi, K., Gröhn, O., & Ilmoniemi, R. J. (2020).
Individual head models for estimating the TMS-induced electric field in rat brain. Scientific Reports,
10(1), 17397. https://doi.org/10.1038/s41598-020-74431-z
Krasteva, V. T., Papazov, S. P., & Daskalov, I. K. (2002). Magnetic stimulation for non-homogeneous
biological structures. BioMedical Engineering OnLine, 1(1), 3. https://doi.org/10.1186/1475-925X1-3
Krystal, J. H., Abdallah, C. G., Sanacora, G., Charney, D. S., & Duman, R. S. (2019). Ketamine: a
paradigm shift for depression research and treatment. Neuron, 101(5), 774–778.
https://doi.org/10.1016/j.neuron.2019.02.005
Krystal, J. H., Karper, L. P., Seibyl, J. P., Freeman, G. K., Delaney, R., Bremner, J. D., Heninger, G. R.,
Bowers, M. B., & Charney, D. S. (1994). Subanesthetic effects of the noncompetitive NMDA
antagonist, ketamine, in humans: psychotomimetic, perceptual, cognitive, and neuroendocrine
responses. Archives of General Psychiatry, 51(3), 199–214.
https://doi.org/10.1001/archpsyc.1994.03950030035004
Kuwabara, S., Cappelen‐Smith, C., Lin, C. S. ‐Y., Mogyoros, I., & Burke, D. (2002). Effects of voluntary
activity on the excitability of motor axons in the peroneal nerve. Muscle & Nerve, 25(2), 176–184.
https://doi.org/10.1002/mus.10030
Lara, A. H., & Wallis, J. D. (2015). The Role of Prefrontal Cortex in Working Memory: A Mini Review.
Frontiers in Systems Neuroscience, 9. https://doi.org/10.3389/fnsys.2015.00173
Lazarevic, V., Pothula, S., Andres-Alonso, M., & Fejtova, A. (2013). Molecular mechanisms driving
homeostatic plasticity of neurotransmitter release. Frontiers in Cellular Neuroscience, 7, 244.
https://doi.org/10.3389/fncel.2013.00244
Lefebvre, L. P., Pelletier, S., & Gélinas, C. (1997). Effect of electrical resistivity on core losses in soft
magnetic iron powder materials. Journal of Magnetism and Magnetic Materials, 176(2–3), L93–
L96. https://doi.org/10.1016/S0304-8853(97)01006-8
Lenz, M., Galanis, C., Müller-Dahlhaus, F., Opitz, A., Wierenga, C. J., Szabó, G., Ziemann, U., Deller,
T., Funke, K., & Vlachos, A. (2016). Repetitive magnetic stimulation induces plasticity of inhibitory
synapses. Nature Communications, 7(1), 1–13. https://doi.org/10.1038/ncomms10020
Lenz, M., Platschek, S., Priesemann, V., Becker, D., Willems, L. M., Ziemann, U., Deller, T., MüllerDahlhaus, F., Jedlicka, P., & Vlachos, A. (2015). Repetitive magnetic stimulation induces plasticity
of excitatory postsynapses on proximal dendrites of cultured mouse CA1 pyramidal neurons. Brain
Structure and Function, 220(6), 3323–3337. https://doi.org/10.1007/s00429-014-0859-9
Lepage, M., Ghaffar, O., Nyberg, L., & Tulving, E. (2000). Prefrontal cortex and episodic memory
retrieval mode. Proceedings of the National Academy of Sciences, 97(1), 506–511.
https://doi.org/10.1073/pnas.97.1.506

126
Levkovitz, Y., Isserles, M., Padberg, F., Lisanby, S. H., Bystritsky, A., Xia, G., Tendler, A., Daskalakis,
Z. J., Winston, J. L., Dannon, P., Hafez, H. M., Reti, I. M., Morales, O. G., Schlaepfer, T. E.,
Hollander, E., Berman, J. A., Husain, M. M., Sofer, U., Stein, A., … Zangen, A. (2015). Efficacy
and safety of deep transcranial magnetic stimulation for major depression: a prospective multicenter
randomized controlled trial. World Psychiatry, 14(1), 64–73. https://doi.org/10.1002/wps.20199
Levy, W. B., & Steward, O. (1983). Temporal contiguity requirements for long-term associative
potentiation/depression in the hippocampus. Neuroscience, 8(4), 791–797.
https://doi.org/10.1016/0306-4522(83)90010-6
Li, B., Virtanen, J. P., Oeltermann, A., Schwarz, C., Giese, M. A., Ziemann, U., & Benali, A. (2017).
Lifting the veil on the dynamics of neuronal activities evoked by transcranial magnetic stimulation.
ELife, 6. https://doi.org/10.7554/eLife.30552
Li, D., & Mashour, G. A. (2019). Cortical dynamics during psychedelic and anesthetized states induced
by ketamine. NeuroImage, 196, 32–40. https://doi.org/10.1016/j.neuroimage.2019.03.076
Li, W., Yang, Y., Ye, Q., Yang, B., & Wang, Z. (2007). Effect of chronic and acute low-frequency
repetitive transcranial magnetic stimulation on spatial memory in rats. Brain Research Bulletin,
71(5), 493–500. https://doi.org/10.1016/j.brainresbull.2006.11.002
Lin, Y., Jin, J., Lv, R., Luo, Y., Dai, W., Li, W., Tang, Y., Wang, Y., Ye, X., & Lin, W.-J. (2021).
Repetitive transcranial magnetic stimulation increases the brain’s drainage efficiency in a mouse
model of Alzheimer’s disease. Acta Neuropathologica Communications, 9(1), 102.
https://doi.org/10.1186/s40478-021-01198-3
Lipton, R. B., Dodick, D. W., Silberstein, S. D., Saper, J. R., Aurora, S. K., Pearlman, S. H., Fischell, R.
E., Ruppel, P. L., & Goadsby, P. J. (2010). Single-pulse transcranial magnetic stimulation for acute
treatment of migraine with aura: a randomised, double-blind, parallel-group, sham-controlled trial.
The Lancet Neurology, 9(4), 373–380. https://doi.org/10.1016/S1474-4422(10)70054-5
Lisman, J. (2010). Working memory: the importance of theta and gamma oscillations. Current Biology,
20(11), R490–R492. https://doi.org/10.1016/j.cub.2010.04.011
Lu, M., & Ueno, S. (2009). Calculating the electric field in real human head by transcranial magnetic
stimulation with shield plate. Journal of Applied Physics, 105(7), 07B322.
https://doi.org/10.1063/1.3059614
Luscher, C., & Malenka, R. C. (2012). NMDA Receptor-Dependent Long-Term Potentiation and LongTerm Depression (LTP/LTD). Cold Spring Harbor Perspectives in Biology, 4(6), a005710–
a005710. https://doi.org/10.1101/cshperspect.a005710
M. Pohl, B., Fernando Gasca, & Hofmann, U. G. (2013). IGES and .stl file of a rat brain. In figshare.
https://doi.org/https://doi.org/10.6084/m9.figshare.823546.v1
Ma, J., Zhang, Z., Kang, L., Geng, D., Wang, Y., Wang, M., & Cui, H. (2014). Repetitive transcranial
magnetic stimulation (rTMS) influences spatial cognition and modulates hippocampal structural
synaptic plasticity in aging mice. Experimental Gerontology, 58, 256–268.
https://doi.org/10.1016/j.exger.2014.08.011
Maeda, F., Keenan, J. P., Tormos, J. M., Topka, H., & Pascual-Leone, A. (2000). Interindividual
variability of the modulatory effects of repetitive transcranial magnetic stimulation on cortical
excitability. Experimental Brain Research, 133(4), 425–430.
https://doi.org/10.1007/s002210000432
Makowiecki, K., Harvey, A. R., Sherrard, R. M., & Rodger, J. (2014). Low-intensity repetitive
transcranial magnetic stimulation improves abnormal visual cortical circuit topography and

127
upregulates BDNF in mice. Journal of Neuroscience, 34(32), 10780–10792.
https://doi.org/10.1523/jneurosci.0723-14.2014
Mansur, C. G., Fregni, F., Boggio, P. S., Riberto, M., Gallucci-Neto, J., Santos, C. M., Wagner, T.,
Rigonatti, S. P., Marcolin, M. A., & Pascual-Leone, A. (2005). A sham stimulation-controlled trial
of rTMS of the unaffected hemisphere in stroke patients. Neurology, 64(10), 1802–1804.
https://doi.org/10.1212/01.WNL.0000161839.38079.92
Martiny, K., Lunde, M., & Bech, P. (2010). Transcranial Low Voltage Pulsed Electromagnetic Fields in
Patients with Treatment-Resistant Depression. Biological Psychiatry, 68(2), 163–169.
https://doi.org/10.1016/j.biopsych.2010.02.017
Matheson, N. A., Shemmell, J. B. H., de Ridder, D., & Reynolds, J. N. J. (2016). Understanding the
Effects of Repetitive Transcranial Magnetic Stimulation on Neuronal Circuits. Frontiers in Neural
Circuits, 10. https://doi.org/10.3389/fncir.2016.00067
May, A., Hajak, G., Gänßbauer, S., Steffens, T., Langguth, B., Kleinjung, T., & Eichhammer, P. (2007).
Structural Brain Alterations following 5 Days of Intervention: Dynamic Aspects of Neuroplasticity.
Cerebral Cortex, 17(1), 205–210. https://doi.org/10.1093/cercor/bhj138
Mazzeffi, M., Johnson, K., & Paciullo, C. (2015). Ketamine in adult cardiac surgery and the cardiac
surgery intensive care unit: an evidence-based clinical review. Annals of Cardiac Anaesthesia,
18(2), 202. https://doi.org/10.4103/0971-9784.154478
McCarthy, D. A., Chen, G., Kaump, D. H., & Ensor, C. (1965). General anesthetic and other
pharmacological properties of 2-(o-chlorophenyl)-2-methylamino cyclohexanone HCl (CI-581). The
Journal of New Drugs, 5(1), 21–33. https://doi.org/10.1002/j.1552-4604.1965.tb00219.x
McDonald, W. M., Durkalski, V., Ball, E. R., Holtzheimer, P. E., Pavlicova, M., Lisanby, S. H., Avery,
D., Anderson, B. S., Nahas, Z., Zarkowski, P., Sackeim, H. A., & George, M. S. (2011). Improving
the antidepressant efficacy of transcranial magnetic stimulation: maximizing the number of
stimulations and treatment location in treatment-resistant depression. Depression and Anxiety,
28(11), 973–980. https://doi.org/10.1002/da.20885
Memarian Sorkhabi, M., & Denison, T. (2022). A neurostimulator system for real, sham, and multi-target
transcranial magnetic stimulation. Journal of Neural Engineering, 19(2), 026035.
https://doi.org/10.1088/1741-2552/ac60c9
Meng, Q., Daugherty, M., Patel, P., Trivedi, S., Du, X., Hong, E., & Choa, F.-S. (2018). High-sensitivity
and spatial resolution transient magnetic and electric field probes for transcranial magnetic
stimulator characterizations. Instrumentation Science & Technology, 46(5), 502–518.
https://doi.org/10.1080/10739149.2017.1401547
Meng, Q., Jing, L., Badjo, J. P., Du, X., Hong, E., Yang, Y., Lu, H., & Choa, F.-S. (2018). A novel
transcranial magnetic stimulator for focal stimulation of rodent brain. Brain Stimulation, 11(3), 663–
665. https://doi.org/10.1016/j.brs.2018.02.018
Meyer, R. E., & Fish, R. (2008). Pharmacology of injectable anesthetics, sedatives, and tranquilizers.
Anesthesia and analgesia of laboratory animals. 2nd ed. San Diego: Academic~….
https://doi.org/10.1016/B978-012373898-1.50006-1
Mickle, A. D., Won, S. M., Noh, K. N., Yoon, J., Meacham, K. W., Xue, Y., McIlvried, L. A., Copits, B.
A., Samineni, V. K., Crawford, K. E., Kim, D. H., Srivastava, P., Kim, B. H., Min, S., Shiuan, Y.,
Yun, Y., Payne, M. A., Zhang, J., Jang, H., … Rogers, J. A. (2019). A wireless closed-loop system
for optogenetic peripheral neuromodulation. Nature, 565(7739), 361–365.
https://doi.org/10.1038/s41586-018-0823-6

128
Mills, K. R., Boniface, S. J., & Schubert, M. (1992). Magnetic brain stimulation with a double coil: the
importance of coil orientation. Electroencephalography and Clinical Neurophysiology/Evoked
Potentials Section, 85(1), 17–21. https://doi.org/10.1016/0168-5597(92)90096-T
Miniussi, C., & Thut, G. (2010). Combining TMS and EEG Offers New Prospects in Cognitive
Neuroscience. Brain Topography, 22(4), 249–256. https://doi.org/10.1007/s10548-009-0083-8
Miranda, M., Morici, J. F., Zanoni, M. B., & Bekinschtein, P. (2019). Brain-Derived Neurotrophic Factor:
A Key Molecule for Memory in the Healthy and the Pathological Brain. Frontiers in Cellular
Neuroscience, 13. https://doi.org/10.3389/fncel.2019.00363
Misra, U. K., Kalita, J., Tripathi, G. M., & Bhoi, S. K. (2013). Is β endorphin related to migraine
headache and its relief? Cephalalgia, 33(5), 316–322. https://doi.org/10.1177/0333102412473372
Mix, A., Benali, A., Eysel, U. T., & Funke, K. (2010). Continuous and intermittent transcranial magnetic
theta burst stimulation modify tactile learning performance and cortical protein expression in the rat
differently. European Journal of Neuroscience, 32(9), 1575–1586. https://doi.org/10.1111/j.1460-
9568.2010.07425.x
Moghaddam, B., Adams, B., Verma, A., & Daly, D. (1997). Activation of glutamatergic
neurotransmission by ketamine: a novel step in the pathway from NMDA receptor blockade to
dopaminergic and cognitive disruptions associated with the prefrontal cortex. Journal of
Neuroscience, 17(8), 2921–2927. https://doi.org/10.1523/jneurosci.17-08-02921.1997
Moliadze, V., Giannikopoulos, D., Eysel, U. T., & Funke, K. (2005). Paired-pulse transcranial magnetic
stimulation protocol applied to visual cortex of anaesthetized cat: effects on visually evoked singleunit activity. The Journal of Physiology, 566(3), 955–965.
https://doi.org/10.1113/jphysiol.2005.086090
Moliadze, V., Zhao, Y., Eysel, U., & Funke, K. (2003). Effect of transcranial magnetic stimulation on
single‐unit activity in the cat primary visual cortex. The Journal of Physiology, 553(2), 665–679.
https://doi.org/10.1113/jphysiol.2003.050153
Moncrieff, J., Cooper, R. E., Stockmann, T., Amendola, S., Hengartner, M. P., & Horowitz, M. A. (2022).
The serotonin theory of depression: a systematic umbrella review of the evidence. Molecular
Psychiatry. https://doi.org/10.1038/s41380-022-01661-0
Mozzachiodi, R., & Byrne, J. H. (2010). More than synaptic plasticity: role of nonsynaptic plasticity in
learning and memory. Trends in Neurosciences, 33(1), 17–26.
https://doi.org/10.1016/j.tins.2009.10.001
Mueller, J. K., Grigsby, E. M., Prevosto, V., Petraglia, F. W., Rao, H., Deng, Z.-D., Peterchev, A. v,
Sommer, M. A., Egner, T., Platt, M. L., & others. (2014). Simultaneous transcranial magnetic
stimulation and single-neuron recording in alert non-human primates. Nature Neuroscience, 17(8),
1130–1136. https://doi.org/10.1038/nn.3751
Müller, M. (2000). Long-Term Repetitive Transcranial Magnetic Stimulation Increases the Expression of
Brain-Derived Neurotrophic Factor and Cholecystokinin mRNA, but not Neuropeptide Tyrosine
mRNA in Specific Areas of Rat Brain. Neuropsychopharmacology, 23(2), 205–215.
https://doi.org/10.1016/S0893-133X(00)00099-3
Muller, P. A., Dhamne, S. C., Vahabzadeh-Hagh, A. M., Pascual-Leone, A., Jensen, F. E., & Rotenberg,
A. (2014). Suppression of Motor Cortical Excitability in Anesthetized Rats by Low Frequency
Repetitive Transcranial Magnetic Stimulation. PLoS ONE, 9(3), e91065.
https://doi.org/10.1371/journal.pone.0091065

129
Murphy, S. C., Palmer, L. M., Nyffeler, T., Mü ri, R. M., & Larkum, M. E. (2016). Transcranial magnetic
stimulation (TMS) inhibits cortical dendrites. ELife , 5, e13598.
https://doi.org/10.7554/eLife.13598.001
Nair, A. B., & Jacob, S. (2016). A simple practice guide for dose conversion between animals and human.
Journal of Basic and Clinical Pharmacy, 7(2), 27. https://doi.org/10.4103/0976-0105.177703
Nath, T., Mathis, A., Chen, A. C., Patel, A., Bethge, M., & Mathis, M. W. (2019). Using DeepLabCut for
3D markerless pose estimation across species and behaviors. Nature Protocols, 14(7), 2152–2176.
https://doi.org/10.1038/s41596-019-0176-0
Navarro de Lara, L. I., Daneshzand, M., Mascarenas, A., Paulson, D., Pratt, K., Okada, Y., Raij, T.,
Makarov, S. N., & Nummenmaa, A. (2021). A 3-axis coil design for multichannel TMS arrays.
NeuroImage, 224, 117355. https://doi.org/10.1016/j.neuroimage.2020.117355
Neske, G. T. (2016). The Slow Oscillation in Cortical and Thalamic Networks: Mechanisms and
Functions. Frontiers in Neural Circuits, 9. https://doi.org/10.3389/fncir.2015.00088
Nicol, A. U., & Morton, A. J. (2020). Characteristic patterns of EEG oscillations in sheep (Ovis aries)
induced by ketamine may explain the psychotropic effects seen in humans. Scientific Reports, 10(1),
9440. https://doi.org/10.1038/s41598-020-66023-8
Nieminen, J. O., Koponen, L. M., & Ilmoniemi, R. J. (2015). Experimental Characterization of the
Electric Field Distribution Induced by TMS Devices. Brain Stimulation, 8(3), 582–589.
https://doi.org/10.1016/j.brs.2015.01.004
Nieminen, J. O., Pospelov, A. S., Koponen, L. M., Yrjölä, P., Shulga, A., Khirug, S., & Rivera, C. (2022).
Transcranial magnetic stimulation set-up for small animals. Frontiers in Neuroscience, 16.
https://doi.org/10.3389/fnins.2022.935268
Nikouline, V., Ruohonen, J., & Ilmoniemi, R. J. (1999). The role of the coil click in TMS assessed with
simultaneous EEG. Clinical Neurophysiology, 110(8), 1325–1328. https://doi.org/10.1016/S1388-
2457(99)00070-X
Nitsche, M. A., Cohen, L. G., Wassermann, E. M., Priori, A., Lang, N., Antal, A., Paulus, W., Hummel,
F., Boggio, P. S., Fregni, F., & Pascual-Leone, A. (2008). Transcranial direct current stimulation:
State of the art 2008. Brain Stimulation, 1(3), 206–223. https://doi.org/10.1016/j.brs.2008.06.004
Nyhus, E., & Curran, T. (2010). Functional role of gamma and theta oscillations in episodic memory.
Neuroscience & Biobehavioral Reviews, 34(7), 1023–1035.
https://doi.org/10.1016/j.neubiorev.2009.12.014
Okamoto, M., Dan, H., Sakamoto, K., Takeo, K., Shimizu, K., Kohno, S., Oda, I., Isobe, S., Suzuki, T.,
Kohyama, K., & Dan, I. (2004). Three-dimensional probabilistic anatomical cranio-cerebral
correlation via the international 10–20 system oriented for transcranial functional brain mapping.
NeuroImage, 21(1), 99–111. https://doi.org/10.1016/j.neuroimage.2003.08.026
O’Reardon, J. P., Solvason, H. B., Janicak, P. G., Sampson, S., Isenberg, K. E., Nahas, Z., McDonald, W.
M., Avery, D., Fitzgerald, P. B., Loo, C., Demitrack, M. A., George, M. S., & Sackeim, H. A.
(2007). Efficacy and Safety of Transcranial Magnetic Stimulation in the Acute Treatment of Major
Depression: A Multisite Randomized Controlled Trial. Biological Psychiatry, 62(11), 1208–1216.
https://doi.org/10.1016/j.biopsych.2007.01.018
Padberg, F. (2002). Repetitive Transcranial Magnetic Stimulation (rTMS) in Major Depression Relation
between Efficacy and Stimulation Intensity. Neuropsychopharmacology.
https://doi.org/10.1016/S0893-133X(02)00338-X

130
Paolo, M. (2015). TMS-evoked N100 responses as a prognostic factor in acute stroke. Functional
Neurology. https://doi.org/10.11138/FNeur/2015.30.2.125
Parthoens, J., Verhaeghe, J., Wyckhuys, T., Stroobants, S., & Staelens, S. (2014). Small-animal repetitive
transcranial magnetic stimulation combined with [18F]-FDG microPET to quantify the
neuromodulation effect in the rat brain. Neuroscience, 275, 436–443.
https://doi.org/10.1016/j.neuroscience.2014.06.042
Paus, T., Sipila, P. K., & Strafella, A. P. (2001). Synchronization of neuronal activity in the human
primary motor cortex by transcranial magnetic stimulation: an EEG study. Journal of
Neurophysiology, 86(4), 1983–1990. https://doi.org/10.1152/jn.2001.86.4.1983.
Penke, Z., Morice, E., Veyrac, A., Gros, A., Chagneau, C., LeBlanc, P., Samson, N., Baumgärtel, K.,
Mansuy, I. M., Davis, S., & Laroche, S. (2014). Zif268/Egr1 gain of function facilitates
hippocampal synaptic plasticity and long-term spatial recognition memory. Philosophical
Transactions of the Royal Society B: Biological Sciences, 369(1633), 20130159.
https://doi.org/10.1098/rstb.2013.0159
Perlmutter, J. S., & Mink, J. W. (2006). Deep Brain Stimulation. Annual Review of Neuroscience, 29(1),
229–257. https://doi.org/10.1146/annurev.neuro.29.051605.112824
Pogarell, O., Koch, W., Pöpperl, G., Tatsch, K., Jakob, F., Mulert, C., Grossheinrich, N., Rupprecht, R.,
Möller, H.-J., Hegerl, U., & Padberg, F. (2007). Acute prefrontal rTMS increases striatal dopamine
to a similar degree as d-amphetamine. Psychiatry Research: Neuroimaging, 156(3), 251–255.
https://doi.org/10.1016/j.pscychresns.2007.05.002
Popovych, O. v, & Tass, P. A. (2014). Control of abnormal synchronization in neurological disorders.
Frontiers in Neurology, 5, 268. https://doi.org/10.3389/fneur.2014.00268
Pradhan, B., & Rossi, G. (2020). Combining Ketamine, Brain Stimulation (rTMS) and Mindfulness
Therapy (TIMBER) for Opioid Addiction. Cureus, 12(11). https://doi.org/10.7759/cureus.11798
Pridmore, S., & Ang, G. (2008). A water-cooled transcranial magnetic stimulation coil. Brain
Stimulation, 1(1), 67. https://doi.org/10.1016/j.brs.2007.08.002
Rabiller, G., He, J.-W., Nishijima, Y., Wong, A., & Liu, J. (2015). Perturbation of brain oscillations after
ischemic stroke: a potential biomarker for post-stroke function and therapy. International Journal of
Molecular Sciences, 16(10), 25605–25640. https://doi.org/10.3390/ijms161025605
RamRakhyani, A. K., & Lazzi, G. (2014). Ferrite core non‐linearity in coils for magnetic
neurostimulation. Healthcare Technology Letters, 1(4), 87–91.
https://doi.org/10.1049/htl.2014.0087
Rastogi, P., Hadimani, R. L., & Jiles, D. C. (2016). Investigation of Coil Designs for Transcranial
Magnetic Stimulation on Mice. IEEE Transactions on Magnetics, 52(7), 1–4.
https://doi.org/10.1109/TMAG.2015.2514064
Ravazzani, P., Ruohonen, J., Tognola, G., Anfosso, F., Ollikainen, M., Ilmoniemi, R. J., & Grandori, F.
(2002). Frequency-related effects in the optimization of coils for the magnetic stimulation of the
nervous system. IEEE Transactions on Biomedical Engineering, 49(5), 463–471.
https://doi.org/10.1109/10.995685
Reis, J., Robertson, E. M., Krakauer, J. W., Rothwell, J., Marshall, L., Gerloff, C., Wassermann, E. M.,
Pascual-Leone, A., Hummel, F., Celnik, P. A., Classen, J., Floel, A., Ziemann, U., Paulus, W.,
Siebner, H. R., Born, J., & Cohen, L. G. (2008). Consensus: Can transcranial direct current
stimulation and transcranial magnetic stimulation enhance motor learning and memory formation?
Brain Stimulation, 1(4), 363–369. https://doi.org/10.1016/j.brs.2008.08.001

131
Richter, L., Neumann, G., Oung, S., Schweikard, A., & Trillenberg, P. (2013). Optimal Coil Orientation
for Transcranial Magnetic Stimulation. PLoS ONE, 8(4), e60358.
https://doi.org/10.1371/journal.pone.0060358
Ridding, M. C., & Rothwell, J. C. (2007). Is there a future for therapeutic use of transcranial magnetic
stimulation? Nature Reviews Neuroscience, 8(7), 559–567. https://doi.org/10.1038/nrn2169
Rodger, J., Mo, C., Wilks, T., Dunlop, S. A., & Sherrard, R. M. (2012). Transcranial pulsed magnetic
field stimulation facilitates reorganization of abnormal neural circuits and corrects behavioral
deficits without disrupting normal connectivity. The FASEB Journal, 26(4), 1593–1606.
https://doi.org/10.1096/fj.11-194878
Rogasch, N. C., & Fitzgerald, P. B. (2013). Assessing cortical network properties using TMS-EEG.
Human Brain Mapping, 34(7), 1652–1669. https://doi.org/10.1002/hbm.22016
Rohan, M. L., Yamamoto, R. T., Ravichandran, C. T., Cayetano, K. R., Morales, O. G., Olson, D. P.,
Vitaliano, G., Paul, S. M., & Cohen, B. M. (2014). Rapid Mood-Elevating Effects of Low Field
Magnetic Stimulation in Depression. Biological Psychiatry, 76(3), 186–193.
https://doi.org/10.1016/j.biopsych.2013.10.024
Romero, M. C., Davare, M., Armendariz, M., & Janssen, P. (2019). Neural effects of transcranial
magnetic stimulation at the single-cell level. Nature Communications, 10(1), 2642.
https://doi.org/10.1038/s41467-019-10638-7
Rösler, K. M., Hess, C. W., Heckmann, R., & Ludin, H. P. (1989). Significance of shape and size of the
stimulating coil in magnetic stimulation of the human motor cortex. Neuroscience Letters, 100(1–3),
347–352. https://doi.org/10.1016/0304-3940(89)90711-8
Rossi, S., Hallett, M., Rossini, P. M., & Pascual-Leone, A. (2009). Safety, ethical considerations, and
application guidelines for the use of transcranial magnetic stimulation in clinical practice and
research. Clinical Neurophysiology, 120(12), 2008–2039.
https://doi.org/10.1016/j.clinph.2009.08.016
Rossini, P. M., Barker, A. T., Berardelli, A., Caramia, M. D., Caruso, G., Cracco, R. Q., Dimitrijević, M.
R., Hallett, M., Katayama, Y., Lücking, C. H., Maertens de Noordhout, A. L., Marsden, C. D.,
Murray, N. M. F., Rothwell, J. C., Swash, M., & Tomberg, C. (1994). Non-invasive electrical and
magnetic stimulation of the brain, spinal cord and roots: basic principles and procedures for routine
clinical application. Report of an IFCN committee. Electroencephalography and Clinical
Neurophysiology, 91(2), 79–92. https://doi.org/10.1016/0013-4694(94)90029-9
Rotem, A., Neef, A., Neef, N. E., Agudelo-Toro, A., Rakhmilevitch, D., Paulus, W., & Moses, E. (2014).
Solving the Orientation Specific Constraints in Transcranial Magnetic Stimulation by Rotating
Fields. PLoS ONE, 9(2), e86794. https://doi.org/10.1371/journal.pone.0086794
Roth, B. J., Maccabee, P. J., Eberle, L. P., Amassian, V. E., Hallett, M., Cadwell, J., Anselmi, G. D., &
Tatarian, G. T. (1994). In vitro evaluation of a 4-leaf coil design for magnetic stimulation of
peripheral nerve. Electroencephalography and Clinical Neurophysiology/Evoked Potentials Section,
93(1), 68–74. https://doi.org/10.1016/0168-5597(94)90093-0
Rothwell, J. C. (1997). Techniques and mechanisms of action of transcranial stimulation of the human
motor cortex. Journal of Neuroscience Methods, 74(2), 113–122. https://doi.org/10.1016/S0165-
0270(97)02242-5
Rougier, N. P., Noelle, D. C., Braver, T. S., Cohen, J. D., & O’Reilly, R. C. (2005). Prefrontal cortex and
flexible cognitive control: Rules without symbols. Proceedings of the National Academy of
Sciences, 102(20), 7338–7343. https://doi.org/10.1073/pnas.0502455102

132
Ruohonen, J. (2003). Chapter 1 Background physics for magnetic stimulation (pp. 3–12).
https://doi.org/10.1016/S1567-424X(09)70204-1
Ruohonen, J., Ollikainen, M., Nikouline, V., Virtanen, J., & Ilmoniemi, R. J. (2000). Coil design for real
and sham transcranial magnetic stimulation. IEEE Transactions on Biomedical Engineering, 47(2),
145–148. https://doi.org/10.1109/10.821731
Ruohonen, J., Ravazzani, P., & Grandori, F. (1998). Functional magnetic stimulation: theory and coil
optimization. Bioelectrochemistry and Bioenergetics, 47(2), 213–219.
https://doi.org/10.1016/S0302-4598(98)00191-3
Salinas, F. S., Lancaster, J. L., & Fox, P. T. (2009). 3D modeling of the total electric field induced by
transcranial magnetic stimulation using the boundary element method. Physics in Medicine and
Biology, 54(12), 3631–3647. https://doi.org/10.1088/0031-9155/54/12/002
Salvador, R., & Miranda, P. C. (2009). Transcranial magnetic stimulation of small animals: A modeling
study of the influence of coil geometry, size and orientation. 2009 Annual International Conference
of the IEEE Engineering in Medicine and Biology Society, 674–677.
https://doi.org/10.1109/IEMBS.2009.5334070
Sanacora, G., Treccani, G., & Popoli, M. (2012). Towards a glutamate hypothesis of depression.
Neuropharmacology, 62(1), 63–77. https://doi.org/10.1016/j.neuropharm.2011.07.036
Sato, Y., Kobayashi, E., Hakamata, Y., Kobahashi, M., Wainai, T., Murayama, T., Mishina, M., & Seo,
N. (2004). Chronopharmacological studies of ketamine in normal and NMDA Ε1 receptor knockout
mice †. British Journal of Anaesthesia, 92(6), 859–864. https://doi.org/10.1093/bja/aeh144
Sauerheber, R., & Heinz, B. (2015). Temperature Effects on Conductivity of Seawater and Physiologic
Saline, Mechanism and Significance. Chemical Sciences Journal, 6(4), 10–4172.
https://doi.org/10.4172/2150-3494.1000109
Sciortino, D., Pigoni, A., Delvecchio, G., Maggioni, E., Schiena, G., & Brambilla, P. (2021). Role of
rTMS in the treatment of cognitive impairments in Bipolar Disorder and Schizophrenia: a review of
Randomized Controlled Trials. Journal of Affective Disorders, 280, 148–155.
https://doi.org/10.1016/j.jad.2020.11.001
Seamans, J. (2008). Losing inhibition with ketamine. Nature Chemical Biology, 4(2), 91–93.
https://doi.org/10.1038/nchembio0208-91
Sekino, M., Ohsaki, H., Takiyama, Y., Yamamoto, K., Matsuzaki, T., Yasumuro, Y., Nishikawa, A.,
Maruo, T., Hosomi, K., & Saitoh, Y. (2015). Eccentric figure-eight coils for transcranial magnetic
stimulation. Bioelectromagnetics, 36(1), 55–65. https://doi.org/10.1002/bem.21886
Shang, Y., Wang, X., Shang, X., Zhang, H., Liu, Z., Yin, T., & Zhang, T. (2016). Repetitive transcranial
magnetic stimulation effectively facilitates spatial cognition and synaptic plasticity associated with
increasing the levels of BDNF and synaptic proteins in Wistar rats. Neurobiology of Learning and
Memory, 134, 369–378. https://doi.org/10.1016/j.nlm.2016.08.016
Shibasaki, K., Sugio, S., Takao, K., Yamanaka, A., Miyakawa, T., Tominaga, M., & Ishizaki, Y. (2015).
TRPV4 activation at the physiological temperature is a critical determinant of neuronal excitability
and behavior. Pflügers Archiv - European Journal of Physiology, 467(12), 2495–2507.
https://doi.org/10.1007/s00424-015-1726-0
Shibasaki, K., Suzuki, M., Mizuno, A., & Tominaga, M. (2007). Effects of Body Temperature on Neural
Activity in the Hippocampus: Regulation of Resting Membrane Potentials by Transient Receptor
Potential Vanilloid 4. Journal of Neuroscience, 27(7), 1566–1575.
https://doi.org/10.1523/JNEUROSCI.4284-06.2007

133
Sibon, I., Strafella, A. P., Gravel, P., Ko, J. H., Booij, L., Soucy, J. P., Leyton, M., Diksic, M., &
Benkelfat, C. (2007). Acute prefrontal cortex TMS in healthy volunteers: Effects on brain 11CαMtrp trapping. NeuroImage, 34(4), 1658–1664. https://doi.org/10.1016/j.neuroimage.2006.08.059
Silberbauer, L. R., Spurny, B., Handschuh, P., Klöbl, M., Bednarik, P., Reiter, B., Ritter, V., Trost, P.,
Konadu, M. E., Windpassinger, M., & others. (2020). Effect of ketamine on limbic GABA and
glutamate: a human in vivo multivoxel magnetic resonance spectroscopy study. Frontiers in
Psychiatry, 920. https://doi.org/10.3389/fpsyt.2020.549903
Silvanto, J., & Pascual-Leone, A. (2008). State-Dependency of Transcranial Magnetic Stimulation. Brain
Topography, 21(1), 1–10. https://doi.org/10.1007/s10548-008-0067-0
Sleigh, J., Harvey, M., Voss, L., & Denny, B. (2014). Ketamine–More mechanisms of action than just
NMDA blockade. Trends in Anaesthesia and Critical Care, 4(2–3), 76–81.
https://doi.org/10.1016/j.tacc.2014.03.002
Slopsema, J. P., Peña, E., Patriat, R., Lehto, L. J., Gröhn, O., Mangia, S., Harel, N., Michaeli, S., &
Johnson, M. D. (2018). Clinical deep brain stimulation strategies for orientation-selective pathway
activation. Journal of Neural Engineering, 15(5), 056029. https://doi.org/10.1088/1741-
2552/aad978
Smith, J. E., & Peterchev, A. v. (2018). Electric field measurement of two commercial active/sham coils
for transcranial magnetic stimulation. Journal of Neural Engineering, 15(5), 054001.
https://doi.org/10.1088/1741-2552/aace89
Sohn, M. K., Song, H.-J., & Jee, S. (2014). Plasticity Associated Changes in Neurophysiological Tests
Following Non Invasive Brain Stimulation in Stroke Rat Model. Korean Journal of Clinical
Neurophysiology, 16(2), 62. https://doi.org/10.14253/kjcn.2014.16.2.62
Sommer, M., Alfaro, A., Rummel, M., Speck, S., Lang, N., Tings, T., & Paulus, W. (2006). Half sine,
monophasic and biphasic transcranial magnetic stimulation of the human motor cortex. Clinical
Neurophysiology, 117(4), 838–844. https://doi.org/10.1016/j.clinph.2005.10.029
Song, D., Wang, H., Tu, C. Y., Marmarelis, V. Z., Hampson, R. E., Deadwyler, S. A., & Berger, T. W.
(2013). Identification of sparse neural functional connectivity using penalized likelihood estimation
and basis functions. Journal of Computational Neuroscience, 35(3), 335–357.
https://doi.org/10.1007/s10827-013-0455-7
Spaldin, N. A. (2010). Magnetic materials: fundamentals and applications (2nd ed.). Cambridge
university press.
Staley, K. J., & Dudek, F. E. (2006). Interictal Spikes and Epileptogenesis. Epilepsy Currents, 6(6), 199–
202. https://doi.org/10.1111/j.1535-7511.2006.00145.x
Steeds, H., Carhart-Harris, R. L., & Stone, J. M. (2015). Drug models of schizophrenia. Therapeutic
Advances in Psychopharmacology, 5(1), 43–58. https://doi.org/10.1177/2045125314557797
Stefan, K., Kunesch, E., Benecke, R., Cohen, L. G., & Classen, J. (2002). Mechanisms of enhancement of
human motor cortex excitability induced by interventional paired associative stimulation. The
Journal of Physiology, 543(2), 699–708. https://doi.org/10.1113/jphysiol.2002.023317
Stenroos, M., & Koponen, L. M. (2019). Real-time computation of the TMS-induced electric field in a
realistic head model. NeuroImage, 203, 116159. https://doi.org/10.1016/j.neuroimage.2019.116159
Stoney, S. D., Thompson, W. D., & Asanuma, H. (1968). Excitation of pyramidal tract cells by
intracortical microstimulation: effective extent of stimulating current. Journal of Neurophysiology,
31(5), 659–669. https://doi.org/10.1152/jn.1968.31.5.659

134
Strafella, A. P. (2003). Striatal dopamine release induced by repetitive transcranial magnetic stimulation
of the human motor cortex. Brain, 126(12), 2609–2615. https://doi.org/10.1093/brain/awg268
Strafella, A. P., Paus, T., Barrett, J., & Dagher, A. (2001). Repetitive Transcranial Magnetic Stimulation
of the Human Prefrontal Cortex Induces Dopamine Release in the Caudate Nucleus. The Journal of
Neuroscience, 21(15), RC157–RC157. https://doi.org/10.1523/JNEUROSCI.21-15-j0003.2001
Strahm, C., Min, K., Boos, N., Ruetsch, Y., & Curt, A. (2003). Reliability of perioperative SSEP
recordings in spine surgery. Spinal Cord, 41(9), 483–489. https://doi.org/10.1038/sj.sc.3101493
Sun, K., Feng, S., Jiang, Q., Li, X., Li, Y., Fan, R., An, Y., & Wang, J. (2020). Intergranular insulating
reduced iron powder-carbonyl iron powder/SiO2-Al2O3 soft magnetic composites with high
saturation magnetic flux density and low core loss. Journal of Magnetism and Magnetic Materials,
493, 165705. https://doi.org/10.1016/j.jmmm.2019.165705
Sun, W., Mao, W., Meng, X., Wang, D., Qiao, L., Tao, W., Li, L., Jia, X., Han, C., Fu, M., Tong, X., Wu,
X., & Wang, Y. (2012). Low-frequency repetitive transcranial magnetic stimulation for the
treatment of refractory partial epilepsy: A controlled clinical study. Epilepsia, 53(10), 1782–1789.
https://doi.org/10.1111/j.1528-1167.2012.03626.x
Tan, T., Xie, J., Liu, T., Chen, X., Zheng, X., Tong, Z., & Tian, X. (2013). Low-frequency (1Hz)
repetitive transcranial magnetic stimulation (rTMS) reverses Aβ1–42-mediated memory deficits in
rats. Experimental Gerontology, 48(8), 786–794. https://doi.org/10.1016/j.exger.2013.05.001
Tang, A. D., Bennett, W., Bindoff, A. D., Bolland, S., Collins, J., Langley, R. C., Garry, M. I., Summers,
J. J., Hinder, M. R., Rodger, J., & Canty, A. J. (2021). Subthreshold repetitive transcranial magnetic
stimulation drives structural synaptic plasticity in the young and aged motor cortex. Brain
Stimulation, 14(6), 1498–1507. https://doi.org/10.1016/j.brs.2021.10.001
Tang, A. D., Bennett, W., Hadrill, C., Collins, J., Fulopova, B., Wills, K., Bindoff, A., Puri, R., Garry, M.
I., Hinder, M. R., Summers, J. J., Rodger, J., & Canty, A. J. (2018). Low intensity repetitive
transcranial magnetic stimulation modulates skilled motor learning in adult mice. Scientific Reports,
8(1), 4016. https://doi.org/10.1038/s41598-018-22385-8
Tang, A. D., Lowe, A. S., Garrett, A. R., Woodward, R., Bennett, W., Canty, A. J., Garry, M. I., Hinder,
M. R., Summers, J. J., Gersner, R., Rotenberg, A., Thickbroom, G., Walton, J., & Rodger, J. (2016).
Construction and Evaluation of Rodent-Specific rTMS Coils. Frontiers in Neural Circuits, 10.
https://doi.org/10.3389/fncir.2016.00047
Tang, A. D., Thickbroom, G., & Rodger, J. (2017). Repetitive Transcranial Magnetic Stimulation of the
Brain. The Neuroscientist, 23(1), 82–94. https://doi.org/10.1177/1073858415618897
Taylor, P. C. J., Walsh, V., & Eimer, M. (2008). Combining TMS and EEG to study cognitive function
and cortico–cortico interactions. Behavioural Brain Research, 191(2), 141–147.
https://doi.org/10.1016/j.bbr.2008.03.033
Teo, J. T. H., Swayne, O. B., & Rothwell, J. C. (2007). Further evidence for NMDA-dependence of the
after-effects of human theta burst stimulation. Clinical Neurophysiology, 118(7), 1649–1651.
https://doi.org/10.1016/j.clinph.2007.04.010
Terao, Y., Ugawa, Y., Sakai, K., Uesaka, Y., Kohara, N., & Kanazawa, I. (2009). Transcranial
stimulation of the leg area of the motor cortex in humans. Acta Neurologica Scandinavica, 89(5),
378–383. https://doi.org/10.1111/j.1600-0404.1994.tb02650.x
Thomson, A. C., Kenis, G., Tielens, S., de Graaf, T. A., Schuhmann, T., Rutten, B. P. F., & Sack, A. T.
(2020). Transcranial Magnetic Stimulation-Induced Plasticity Mechanisms: TMS-Related Gene
Expression and Morphology Changes in a Human Neuron-Like Cell Model. Frontiers in Molecular
Neuroscience, 13. https://doi.org/10.3389/fnmol.2020.528396

135
Tischler, H., Wolfus, S., Friedman, A., Perel, E., Pashut, T., Lavidor, M., Korngreen, A., Yeshurun, Y., &
Bar-Gad, I. (2011). Mini-coil for magnetic stimulation in the behaving primate. Journal of
Neuroscience Methods, 194(2), 242–251. https://doi.org/10.1016/j.jneumeth.2010.10.015
Tofts, P. S., & Branston, N. M. (1991). The measurement of electric field, and the influence of surface
charge, in magnetic stimulation. Electroencephalography and Clinical Neurophysiology/Evoked
Potentials Section, 81(3), 238–239. https://doi.org/10.1016/0168-5597(91)90077-B
Tokimura, H., Ridding, M. C., Tokimura, Y., Amassian, V. E., & Rothwell, J. C. (1996). Short latency
facilitation between pairs of threshold magnetic stimuli applied to human motor cortex.
Electroencephalography and Clinical Neurophysiology/Electromyography and Motor Control,
101(4), 263–272. https://doi.org/10.1016/0924-980X(96)95664-7
Tsuji, T., & Rothwell, J. C. (2002). Long lasting effects of rTMS and associated peripheral sensory input
on MEPs, SEPs and transcortical reflex excitability in humans. The Journal of Physiology, 540(1),
367–376. https://doi.org/10.1113/jphysiol.2001.013504
Uddin, L. Q. (2021). Cognitive and behavioural flexibility: neural mechanisms and clinical
considerations. Nature Reviews Neuroscience, 22(3), 167–179. https://doi.org/10.1038/s41583-021-
00428-w
Ueno, S., Matsuda, T., & Hiwaki, O. (1990). Localized stimulation of the human brain and spinal cord by
a pair of opposing pulsed magnetic fields. Journal of Applied Physics, 67(9), 5838–5840.
https://doi.org/10.1063/1.345982
Ueno, S., & Sekino, M. (2021). Figure-Eight Coils for Magnetic Stimulation: From Focal Stimulation to
Deep Stimulation. Frontiers in Human Neuroscience, 15.
https://doi.org/10.3389/fnhum.2021.805971
Ueno, S., Tashiro, T., & Harada, K. (1988). Localized stimulation of neural tissues in the brain by means
of a paired configuration of time‐varying magnetic fields. Journal of Applied Physics, 64(10), 5862–
5864. https://doi.org/10.1063/1.342181
Ueyama, E., Ukai, S., Ogawa, A., Yamamoto, M., Kawaguchi, S., Ishii, R., & Shinosaki, K. (2011).
Chronic repetitive transcranial magnetic stimulation increases hippocampal neurogenesis in rats.
Psychiatry and Clinical Neurosciences, 65(1), 77–81. https://doi.org/10.1111/j.1440-
1819.2010.02170.x
Vadivelu, N., Schermer, E., Kodumudi, V., Belani, K., Urman, R. D., & Kaye, A. D. (2016). Role of
ketamine for analgesia in adults and children. Journal of Anaesthesiology, Clinical Pharmacology,
32(3), 298. https://doi.org/10.4103/0970-9185.168149
Vahabzadeh-Hagh, A. M., Muller, P. A., Gersner, R., Zangen, A., & Rotenberg, A. (2012). Translational
Neuromodulation: Approximating Human Transcranial Magnetic Stimulation Protocols in Rats.
Neuromodulation: Technology at the Neural Interface, 15(4), 296–305.
https://doi.org/10.1111/j.1525-1403.2012.00482.x
Valls-Solé, J., Pascual-Leone, A., Wassermann, E. M., & Hallett, M. (1992). Human motor evoked
responses to paired transcranial magnetic stimuli. Electroencephalography and Clinical
Neurophysiology/Evoked Potentials Section, 85(6), 355–364. https://doi.org/10.1016/0168-
5597(92)90048-G
van Wijk, B. C. M., Beek, P. J., & Daffertshofer, A. (2012). Neural synchrony within the motor system:
what have we learned so far? Frontiers in Human Neuroscience, 6, 252.
https://doi.org/10.3389/fnhum.2012.00252

136
Veilleux-Lemieux, D., Castel, A., Carrier, D., Beaudry, F., & Vachon, P. (2013). Pharmacokinetics of
Ketamine and Xylazine in Young and Old Sprague–Dawley Rats. Journal of the American
Association for Laboratory Animal Science, 52(5), 567–570.
Vlachos, A., Müller-Dahlhaus, F., Rosskopp, J., Lenz, M., Ziemann, U., & Deller, T. (2012). Repetitive
magnetic stimulation induces functional and structural plasticity of excitatory postsynapses in mouse
organotypic hippocampal slice cultures. Journal of Neuroscience, 32(48), 17514–17523.
https://doi.org/10.1523/jneurosci.0409-12.2012
Voigt, J. D., Leuchter, A. F., & Carpenter, L. L. (2021). Theta burst stimulation for the acute treatment of
major depressive disorder: A systematic review and meta-analysis. Translational Psychiatry, 11(1),
330. https://doi.org/10.1038/s41398-021-01441-4
Wang, B., Grill, W. M., & Peterchev, A. v. (2018). Coupling Magnetically Induced Electric Fields to
Neurons: Longitudinal and Transverse Activation. Biophysical Journal, 115(1), 95–107.
https://doi.org/10.1016/j.bpj.2018.06.004
Wang, H., & Gao, H. (2023). Protective effects of repetitive transcranial magnetic stimulation against 6-
OHDA-induced Parkinson’s symptoms in a mice model: the key role of miR-409-3p/PDHB axis.
International Journal of Neuroscience, 133(5), 578–585.
https://doi.org/10.1080/00207454.2021.1938037
Wang, T., Guo, M., Wang, N., Zhai, H., Wang, Z., & Xu, G. (2023). Effects of theta burst stimulation on
the coherence of local field potential during working memory task in rats. Brain Research, 1813,
148408. https://doi.org/10.1016/j.brainres.2023.148408
Weiler, M., Moreno-Castilla, P., Starnes, H. M., Melendez, E. L. R., Stieger, K. C., Long, J. M., & Rapp,
P. R. (2021). Effects of repetitive Transcranial Magnetic Stimulation in aged rats depend on pretreatment cognitive status: Toward individualized intervention for successful cognitive aging. Brain
Stimulation, 14(5), 1219–1225. https://doi.org/10.1016/j.brs.2021.08.008
Welniak–Kaminska, M., Fiedorowicz, M., Orzel, J., Bogorodzki, P., Modlinska, K., Stryjek, R.,
Chrzanowska, A., Pisula, W., & Grieb, P. (2019). Volumes of brain structures in captive wild-type
and laboratory rats: 7T magnetic resonance in vivo automatic atlas-based study. PLOS ONE, 14(4),
e0215348. https://doi.org/10.1371/journal.pone.0215348
Wilson, M. T., Tang, A. D., Iyer, K., McKee, H., Waas, J., & Rodger, J. (2018). The challenges of
producing effective small coils for transcranial magnetic stimulation of mice. Biomedical Physics &
Engineering Express, 4(3), 037002. https://doi.org/10.1088/2057-1976/aab525
Wu, Q., Xu, X., Zhai, C., Zhao, Z., Dai, W., Wang, T., & Shen, Y. (2022). High-frequency repetitive
transcranial magnetic stimulation improves spatial episodic learning and memory performance by
regulating brain plasticity in healthy rats. Frontiers in Neuroscience, 16.
https://doi.org/10.3389/fnins.2022.974940
Xie, Y., Zhang, T., & Chen, A. C. N. (2015). Repetitive Transcranial Magnetic Stimulation for the
Recovery of Stroke Patients With Disturbance of Consciousness. Brain Stimulation, 8(3), 674–675.
https://doi.org/10.1016/j.brs.2015.01.406
Yang, H.-Y., Liu, Y., Xie, J.-C., Liu, N.-N., & Tian, X. (2015). Effects of repetitive transcranial magnetic
stimulation on synaptic plasticity and apoptosis in vascular dementia rats. Behavioural Brain
Research, 281, 149–155. https://doi.org/10.1016/j.bbr.2014.12.037
Yang, J., Liang, R., Wang, L., Zheng, C., Xiao, X., & Ming, D. (2021). Repetitive Transcranial Magnetic
Stimulation (rTMS) Improves the Gait Disorders of Rats Under Simulated Microgravity Conditions
Associated With the Regulation of Motor Cortex. Frontiers in Physiology, 12.
https://doi.org/10.3389/fphys.2021.587515

137
Yeh, N., & Rose, N. S. (2019). How Can Transcranial Magnetic Stimulation Be Used to Modulate
Episodic Memory?: A Systematic Review and Meta-Analysis. Frontiers in Psychology, 10.
https://doi.org/10.3389/fpsyg.2019.00993
Yu, K., Niu, X., Krook-Magnuson, E., & He, B. (2021). Intrinsic functional neuron-type selectivity of
transcranial focused ultrasound neuromodulation. Nature Communications, 12(1), 2519.
https://doi.org/10.1038/s41467-021-22743-7
Yu, X., Qian, C., Chen, D., Dodd, S. J., & Koretsky, A. P. (2014). Deciphering laminar-specific neural
inputs with line-scanning fMRI. Nature Methods, 11(1), 55–58. https://doi.org/10.1038/nmeth.2730
Yukimasa, T., Yoshimura, R., Tamagawa, A., Uozumi, T., Shinkai, K., Ueda, N., Tsuji, S., & Nakamura,
J. (2006). High-Frequency Repetitive Transcranial Magnetic Stimulation Improves Refractory
Depression by Influencing Catecholamine and Brain-Derived Neurotrophic Factors.
Pharmacopsychiatry, 39(2), 52–59. https://doi.org/10.1055/s-2006-931542
Yunokuchi, K., & Cohen, D. (1991). Developing a More Focal Magnetic Stimulator. Part II. Journal of
Clinical Neurophysiology, 8(1), 112–120. https://doi.org/10.1097/00004691-199101000-00014
Zangen, A., Moshe, H., Martinez, D., Barnea‐Ygael, N., Vapnik, T., Bystritsky, A., Duffy, W., Toder, D.,
Casuto, L., Grosz, M. L., Nunes, E. V., Ward, H., Tendler, A., Feifel, D., Morales, O., Roth, Y.,
Iosifescu, D. V., Winston, J., Wirecki, T., … George, M. S. (2021). Repetitive transcranial magnetic
stimulation for smoking cessation: a pivotal multicenter double‐blind randomized controlled trial.
World Psychiatry, 20(3), 397–404. https://doi.org/10.1002/wps.20905
Zhang, N., Xing, M., Wang, Y., Tao, H., & Cheng, Y. (2015). Repetitive transcranial magnetic
stimulation enhances spatial learning and synaptic plasticity via the VEGF and BDNF–NMDAR
pathways in a rat model of vascular dementia. Neuroscience, 311, 284–291.
https://doi.org/10.1016/j.neuroscience.2015.10.038
Zhi-De Deng, & Peterchev, A. v. (2011). Transcranial magnetic stimulation coil with electronically
switchable active and sham modes. 2011 Annual International Conference of the IEEE Engineering
in Medicine and Biology Society, 1993–1996. https://doi.org/10.1109/IEMBS.2011.6090561
Zhu, H., Xu, G., Li, Y., Fu, R., Yin, X., Xu, B., & Ding, C. (2021). Immediate and cumulative effects of
high-frequency repetitive transcranial magnetic stimulation on cognition and neuronal excitability in
mice. Neuroscience Research, 173, 90–98. https://doi.org/10.1016/j.neures.2021.05.015
Zimmermann, K. P., & Simpson, R. K. (1996). “Slinky” coils for neuromagnetic stimulation.
Electroencephalography and Clinical Neurophysiology/Electromyography and Motor Control,
101(2), 145–152. https://doi.org/10.1016/0924-980X(95)00227-C
Zmeykina, E., Mittner, M., Paulus, W., & Turi, Z. (2020). Weak rTMS-induced electric fields produce
neural entrainment in humans. Scientific Reports, 10(1), 1–16. https://doi.org/10.1038/s41598-020-
68687-8

Abstract (if available)

Abstract Animal models are essential for investigating the neurobiological mechanisms of transcranial magnetic stimulation (TMS). However, the lack of miniaturized coils hinders TMS studies in small animals, since most commercial coils are designed for humans and are thus incapable of focal stimulation in small animals. To address this, we developed a novel miniaturized C-shaped TMS coil. The resulting magnetic and electric fields were characterized through measurements and modeling. The coil’s efficacy in neuromodulation was validated with concurrent electrophysiological recordings. Additionally, the effect of TMS and ketamine on brain oscillations was explored. A novel form of synchronized activities, poly population spikes (PPS), was discovered as the biomarker of ketamine in local field potentials. TMS effectively and reversibly suppressed PPS. This result suggests that ketamine and TMS have convergent but opposing effects on cortical oscillations and circuits. We further investigated both the chronic and acute effects of TMS on learning and memory in rats performing the Barnes maze task. Chronic TMS initially disrupted spatial learning during acquisition but subsequently enhanced cognitive flexibility in reversal learning and strengthened spatial memories of previous target locations during probe trials. However, the acute effects of TMS were not evident on spatial learning and memory. This novel miniaturized coil and its associated experimental paradigms facilitate the integration of TMS, standard electrophysiology, and pharmacological manipulations. It provides a platform for investigating the mechanisms of TMS from the single neuron to behavioral level in small animal models, as well as for optimizing therapeutic strategies for neurological and neuropsychiatric disorders.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Ultrasound neuromodulation and its applications for noninvasive vision restoration

PDF

Next generation neural interfaces for neural modulation and prosthesis

PDF

Contextual interference in motor skill learning: an investigation of the practice schedule effect using transcranial magnetic stimulation (TMS)

PDF

On the electrophysiology of multielectrode recordings of the basal ganglia and thalamus to improve DBS therapy for children with secondary dystonia

PDF

Parylene-based biomems sensors for multiple physiological systems

PDF

Fabrication and packaging of three-dimensional Parylene C neural interfaces

PDF

Robust and generalizable knowledge acquisition from text

PDF

Novel beamforming techniques for robust contrast enhancement in ultrasound imaging

PDF

Theory and design of magnetic induction-based wireless underground sensor networks

PDF

Electronics design and in vivo evaluation of a wirelessly rechargeable fetal micropacemaker

PDF

Adaptive octree meshing of the extracellular space in neural simulations

PDF

Data-driven modeling of the hippocampus & the design of neurostimulation patterns to abate seizures

PDF

Therapeutic electrical stimulation strategies for neuroregeneration and neuroprotection of retinal neurons

PDF

Decoding memory from spatio-temporal patterns of neuronal spikes

PDF

Stimulation strategies to improve efficiency and temporal resolution of epiretinal prostheses

PDF

Towards high-performance low-cost AMS designs: time-domain conversion and ML-based design automation

PDF

A wireless implantable MEMS micropump system for site-specific anti-cancer drug delivery

PDF

Simulating electrical stimulation and recording in a multi-scale model of the hippocampus

PDF

Cortical and subcortical responses to electrical stimulation of rat retina

PDF

Liquid flow rate sensing via solution resistance for in vivo applications

Asset Metadata

Creator Jiang, Wenxuan (author)

Core Title Design and application of a C-shaped miniaturized coil for transcranial magnetic stimulation in rodents

School Viterbi School of Engineering

Degree Doctor of Philosophy

Degree Program Biomedical Engineering

Degree Conferral Date 2023-12

Publication Date 11/01/2023

Defense Date 09/11/2023

Publisher Los Angeles, California (original), University of Southern California (original), University of Southern California. Libraries (digital)

Tag electrophysiology,ketamine,learning and memory,neuromodulation,OAI-PMH Harvest,rodent models,transcranial magnetic stimulation

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Song, Dong (committee chair), Hashemi, Hossein (committee member), Liu, Charles (committee member), Meng, Ellis (committee member), Zhou, Qifa (committee member)

Creator Email jian393@usc.edu,jiangwenxuan1993@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC113762980

Unique identifier UC113762980

Identifier etd-JiangWenxu-12447.pdf (filename)

Legacy Identifier etd-JiangWenxu-12447

Document Type Dissertation

Format theses (aat)

Rights Jiang, Wenxuan

Internet Media Type application/pdf

Type texts

Source 20231103-usctheses-batch-1104 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu