University of Southern California Dissertations and Theses

Contextual modulation of sensory processing via the pulvinar nucleus

(USC Thesis Other)

Contextual modulation of sensory processing via the pulvinar nucleus

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content CONTEXTUAL MODULATION OF SENSORY PROCESSING
VIA THE PULVINAR NUCLEUS
by
Qi Fang
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
(Neuroscience)
May 2021
Copyright 2021 Qi Fang
ii
Acknowledgements
This thesis would not have been possible without the constant support from my friends, family,
colleagues and community over these years. I would like to express my most sincere thanks to
them.
Firstly, to my supervisors, Dr. Li Zhang and Dr. Huizhong Tao, I am really impressed by your real
talent in science and persistence in the academia. Thank you for your genuine enthusiasm for
science, for mentoring me in all aspects of data collection, analysis and preparing the manuscripts,
and for your kind guidance and brilliant ideas on my projects.
To my labmates, thank you for creating such a cheerful, friendly, helpful and supportive working
environment. Your suggestions and advice have been invaluable to this work.
To my thesis committee, Dr. Jeannie Chen and Dr. Samuel Andrew Hires, thank you for your
guidance and assistance and for always having an open door to discuss my thesis work.
To my guidance committee, Dr. Jeannie Chen, Dr. Samuel Andrew Hires, and Dr. Alexandre
Bonnin, thank you for having you on my written and oral qualifications and for your insightful
questions and suggestions on my projects.
To my fellow members of NGP 2013, it was a pleasure to have been part of such a diverse, talented
and inspiring group. Thank you for all the enlightening conversations and activities.
To the USC and ZNI neuroscience community, thank you for being such a collaborative and
supportive workplace, where ideas, techniques and skills are always freely shared.
iii
To my dear friends near and far, thank you for motivating, supporting, inspiring during my PhD
study.
To my parents, thank you for your commitment to education. Working in scientific research is a
such privilege that I would not have been fortunate enough to pursue without your support.

iv
Table of Contents
Acknowledgements ......................................................................................................................... ii
Table of Figures ............................................................................................................................ vii
Abstract .......................................................................................................................................... ix
Chapter 1: Introduction ................................................................................................................... 1
1.1 Feedforward pathways of sensory processing .................................................................... 2
1.2 More than feedforward pathways of sensory processing .................................................... 4
1.3 First- and higher-order thalamic nuclei............................................................................... 6
1.4 Connectivity of pulvinar/LP ............................................................................................... 8
1.5 Subdivisions of pulvinar/LP ............................................................................................. 10
1.6 Function of pulvinar/LP .................................................................................................... 13
1.6.1 Vision ....................................................................................................................... 13
1.6.2 Motor........................................................................................................................ 14
1.6.3 Innate defensive behavior ........................................................................................ 15
1.6.4 Selective attention .................................................................................................... 15
1.6.5 Decision making ...................................................................................................... 16
Chapter 2: A differential circuit via colliculo-pulvinar pathway enhances feature selectivity in
visual cortex through surround suppression ................................................................................. 18
2.1 Introduction ....................................................................................................................... 18
2.2 Results ............................................................................................................................... 20
2.2.1 LP silencing impairs visual discrimination task performance in mice .................... 20
2.2.2 LP silencing reduces functional selectivity in L2/3 of V1 ....................................... 21
2.2.3 Optogenetic silencing of LP reduces visual selectivity in L2/3 ............................... 26
2.2.4 Activation of LP improves visual processing in L2/3 ............................................. 29
2.2.5 LP axons innervate L1 inhibitory neurons in V1 ..................................................... 33
2.2.6 The LP-V1 projection can modulate V1 processing ................................................ 35
2.2.7 The sSC drives LP to modulate V1 feature selectivity ............................................ 39
2.2.8 LP helps to maintain V1 orientation selectivity in a noisy background .................. 43
2.3 Discussion ......................................................................................................................... 46
2.4 Material and methods ........................................................................................................ 51
v
2.4.1 Animal...................................................................................................................... 51
2.4.2 Head fixation and surgery ........................................................................................ 51
2.4.3 Behavioral task......................................................................................................... 52
2.4.4 In vivo electrophysiological recording ..................................................................... 53
2.4.5 Viral injection .......................................................................................................... 54
2.4.6 Histology and imaging ............................................................................................. 56
2.4.7 Visual and LED stimulation..................................................................................... 56
2.4.8 In vivo optogenetic manipulation ............................................................................. 58
2.4.9 Pharmacological silencing of brain regions ............................................................. 58
2.4.10 Slice preparation and recording ............................................................................... 59
2.4.11 Data analysis ............................................................................................................ 61
2.4.12 Statistics ................................................................................................................... 64
Chapter 3: Contextual and Cross-Modality Modulation of Auditory Cortical Processing through
Pulvinar ......................................................................................................................................... 65
3.1 Introduction ....................................................................................................................... 65
3.2 Results ............................................................................................................................... 67
3.2.1 Bidirectional modulation of frequency tuning and SNR in A1 L2/3 by LP ............ 67
3.2.2 LP exerts a thresholding effect on A1 L2/3 responses ............................................ 68
3.2.3 LP modulation of A1 is mainly mediated by the LP-A1 projection ........................ 74
3.2.4 LP axons produce a disynaptic inhibitory effect on A1 L2/3 neurons .................... 77
3.2.5 LP plays a role in noise-related contextual modulation of A1 responses ................ 79
3.2.6 SC can drive the LP-mediated modulation of A1 responses ................................... 83
3.2.7 LP mediates visual stimuli induced modulation of auditory responses in A1 ......... 85
3.3 Discussion ......................................................................................................................... 87
3.4 Material and methods ........................................................................................................ 90
3.4.1 Experimental animals............................................................................................... 90
3.4.2 Viral and neural tracer injection .............................................................................. 90
3.4.3 Histology and imaging ............................................................................................. 91
3.4.4 Optogenetic and pharmacological manipulation ..................................................... 92
3.4.5 Slice preparation and recording ............................................................................... 93
3.4.6 Sound and visual stimulation ................................................................................... 94
vi
3.4.7 In vivo electrophysiology ......................................................................................... 95
3.4.8 Data analysis ............................................................................................................ 96
3.4.9 Statistics ................................................................................................................... 97
References ..................................................................................................................................... 99

vii
Table of Figures
Figure 1. Two major visual pathways ............................................................................................. 4
Figure 2. Wiring pattern of first-order and higher-order thalamic nuclei ....................................... 7
Figure 3. The subdivisions and projection pattern of the rodent LP ............................................. 11
Figure 4. The subdivisions and projection pattern of the primate pulvinar .................................. 12
Figure 5. Visual response properties of V1 neurons after pharmacologically silencing LP ......... 24
Figure 6. Summary of changes in response properties in V1 before and after LP silencing with
bupivacaine ................................................................................................................................... 26
Figure 7. Effects of LP silencing with ArchT activation on V1 response properties ................... 28
Figure 8. Summary of changes in response properties in V1 without and with ArchT inactivation
of LP.............................................................................................................................................. 29
Figure 9. Changes of visual response properties of V1 neurons with optogenetic activation of LP
....................................................................................................................................................... 31
Figure 10. Summary of changes in response properties in V1 without and with ChR2 activation
of LP.............................................................................................................................................. 32
Figure 11. LP axons innervate V1 L1 inhibitory neurons ............................................................ 34
Figure 12. The LP to V1 projection primarily accounts for the LP modulation of V1 responses 37
Figure 13. PM was not affected when activating LP-V1 axons ................................................... 39
Figure 14. SC input drives LP to modulate V1 responses ............................................................ 41
Figure 15. Effects of SC and V1 silencing on LP visual responses .............................................. 42
Figure 16. Summary of changes in response properties in V1 before and after silencing SC with
bupivacaine ................................................................................................................................... 43
viii
Figure 17. LP helps to maintain V1 orientation selectivity in the face of increasing background
noise .............................................................................................................................................. 45
Figure 18. Changes of firing rates to flashing grating after silencing LP ..................................... 46
Figure 19. A working model for LP modulation of visual processing in V1 ............................... 50
Figure 20. Cortical depth of recorded neurons ............................................................................. 54
Figure 21. Effects of bidirectional manipulation of LP activity on A1 response properties ........ 70
Figure 22. LED illumination had no effect on auditory response................................................. 72
Figure 23. LP activity manipulation did not change CF of TRF .................................................. 73
Figure 24. Effects on BW20 ......................................................................................................... 73
Figure 25. No effects in L4 of A1 ................................................................................................. 74
Figure 26. Bidirectional activity manipulations of the LP-A1 projection .................................... 76
Figure 27. Projection from LP to A1 ............................................................................................ 77
Figure 28. Effects of LP-A1 terminal manipulations on the TRF ................................................ 77
Figure 29. A1 cell types innervated by LP axons ......................................................................... 79
Figure 30. LP plays a role in noise-related contextual modulation of A1 responses.................... 82
Figure 31. SC can drive LP-mediated modulation of A1 responses ............................................. 84
Figure 32. SC projection to LP ..................................................................................................... 85
Figure 33. LP neurons are activated strongly by visual looming stimuli ..................................... 86
Figure 34. Visual looming stimuli modulate A1 auditory responses via LP ................................ 86

ix
Abstract
We interact with the external world using senses. Sensory processing is essential for our survival.
Encoding and decoding of sensory information allows us to understand the external world and take
actions to it when necessary. External sensory stimuli are collected by our sensory organs and sent
to the brain. These ever-changing signals are transduced and encoded to establish an internal,
dynamic and optimized representation of the external world, which is further shaped by internal
brain state, previous experience, and learning and memory, and sometimes distorted under illusory
or pathological conditions. This internal modelling of external world is then decoded to serve as
the basis to execute higher cognitive function, such as conceptualization and decision making, and
to guide our behaviors and actions. After decades of inter-disciplinary investigations and
tremendous efforts, we begin to elucidate the strategies of encoding and decoding sensory stimuli
in context or noisy background. Most of the sensory information sent to the cortex is routed via
the thalamus first. While a small portion of the thalamus (first-order thalamic nuclei) are dedicated
to faithfully convey sensory the information to the primary sensory cortex, the vast majority of the
thalamus are higher-order nuclei whose functionality is poorly understood. Here I will present two
studies on the most prominent higher-order thalamic nucleus, the pulvinar nucleus (or the lateral
posterior nucleus in rodents), trying to uncover its function in contextual and cross-modal sensory
processing.
In the first study, we investigated the modulation of visual processing via the pulvinar nucleus. In
the mammalian visual system, information from the retina streams into parallel bottom-up
pathways. It remains unclear how these pathways interact to contribute to contextual modulation
of visual cortical processing. By optogenetic inactivation and activation of mouse lateral posterior
x
nucleus (LP) of thalamus, a homolog of pulvinar, or its projection to primary visual cortex (V1),
we found that LP contributes to surround suppression of layer (L) 2/3 responses in V1 by driving
L1 inhibitory neurons. This results in subtractive suppression of visual responses and an overall
enhancement of orientation, direction, spatial, and size selectivity. Neurons in V1-projecting LP
regions receive bottom-up input from the superior colliculus (SC) and respond preferably to non-
patterned visual noise. The noise-dependent LP activity allows V1 to ‘‘cancel’’ noise effects and
maintain its orientation selectivity under varying noise background. Thus, the retina-SC-LP-V1
pathway forms a differential circuit with the canonical retino-geniculate pathway to achieve
context-dependent sharpening of visual representations.
In the second study, we explored whether pulvinar can modulate auditory processing in a similar
manner and examined its role in the cross-modal integration. Indeed, bidirectional activity
modulations of LP or its projection to the primary auditory cortex (A1) in awake mice reveal that
LP improves auditory processing in A1 supragranular-layer neurons by sharpening their receptive
fields and frequency tuning, as well as increasing the signal-to-noise ratio (SNR). This is achieved
through a subtractive-suppression mechanism, mediated largely by LP-to-A1 axons preferentially
innervating specific inhibitory neurons in layer 1 and superficial layers. LP is strongly activated
by specific sensory signals relayed from the SC, contributing to the maintenance and enhancement
of A1 processing in the presence of auditory background noise and threatening visual looming
stimuli, respectively. Thus, a multisensory bottom-up SC-pulvinar-A1 pathway plays a role in
contextual and cross-modality modulation of auditory cortical processing.
These two studies together suggest that the pulvinar nucleus modulates cortical sensory processing
of various modalities and enhances the saliency of sensory stimuli in a noisy background or a
xi
multisensory confounding context. They will deepen our understanding of higher-order thalamic
nuclei in sensory processing and shed light on future studies to explore how the pulvinar nucleus
corroborate with other higher-order thalamic nuclei and cortical regions to optimize our sensory
perception and decision making.
1
Chapter 1: Introduction
Sensory processing is fundamental for our survival. We interact with the external world using
senses. External sensory stimuli, including visual scenes, sounds, touch, taste, odor, etc., are
collected by our sensory organs (eyes, ears, skin, tongue, nose, respectively). However, most of
the information is not sent to the cortex directly but routed via the thalamus. These ever-changing
signals are transduced and encoded to establish an internal, dynamic and optimal representation of
the external world, which is further shaped by internal brain state, previous experience, and
learning and memory, and sometimes distorted in hallucinations caused by drug intoxication or
schizophrenia. This internal modelling of external world is then decoded to serve as the basis to
execute higher cognitive function, such as conceptualization and decision making, and to guide
our behaviors and actions. After decades of inter-disciplinary investigations and tremendous
efforts, we begin to elucidate the encoding and decoding strategies of sensory stimuli and the
underlying neural correlates or pathways at synaptic, cellular, network, system, and ultimately
cognitive and behavioral levels.
In this introduction chapter, I will first briefly overview the function and composition of the visual
pathways, with an emphasis on the two types of thalamic nuclei, the first-order nuclei and the
higher-order nuclei. Then I will focus on the largest and most important higher-order visual
thalamic nucleus, the pulvinar nucleus in primates or the lateral posterior nucleus in rodents and
introduce the anatomy and known functions of this structure.

2
1.1 Feedforward pathways of sensory processing
While each of the five principal senses (vision, hearing, somatic sensation, gustation, and olfaction)
engages distinct neural pathways, they share some common features of sensory processing. All
physical (light or mechanical waves) or chemical inputs are collected by specialized receptors on
the peripheral sensory epithelium and transduced/converted into neural activities, relayed in
subcortical structures, and further processed and integrated in the cortical areas to produce an
internal representation of the world. Such peripheral to central feedforward pathway is found in
every sensory modality, and we will elaborate the visual pathways in the following sections, as
this sense plays a privileged role in human perception (Figure 1).
The feedforward pathway of visual processing begins in the eye. Light as photon influx hits the
back of the eyeball, the retina, and is transduced into electrical signals by photoreceptors (rods and
cones) at the outer nuclear layer (ONL). These signals are received by bipolar cells in the inner
nuclear layer (INL) and sent to the retinal ganglion cells (RGCs) in the ganglion cell layer (GCL).
There, local visual features such as motion direction and edges are encoded in the spike trains
(Azeredo da Silveira and Roska, 2011; Dhande and Huberman, 2014; Dhande et al., 2015). The
RGC axonal bundle leaves the retina and enters the brain through the optic tract, with two major
targets. In the canonical image-forming pathway, the retinal inputs innervate the dorsal part of the
lateral geniculate nucleus of the thalamus (LGd), and LGd neurons in turn project to layer (L)4 of
the primary visual cortex (V1), the major recipient layer of thalamocortical inputs. In the parallel,
other RGC axons proceed to the superior colliculus (SC). SC is a highly conserved structure in the
midbrain, indispensable for eye movement, arm reaching, orienting behavior, etc. (Ito and
Feldheim, 2018). Interestingly, the relative proportion of RGC projecting to LGd and SC is
3
significantly different across species. In primates, 90% of the retinal axons target LGd (Perry and
Cowey, 1984). In contrast, 90% of the axons leaving the retina travel to SC (Ellis et al., 2016;
Hofbauer and Dräger, 1985; Linden and Perry, 1983). Note that no direct connection is present
between SC and V1, and information leaving the SC is relayed in the pulvinar nucleus of the
thalamus before reaching V1.
The visual information relayed by the thalamus is subsequently processed in visual cortices
organized in a functional hierarchy. The first stage is V1, where local features, including eye
dominance, edge, orientation, motion direction, and color, are extracted from visual scenes.
Information from V1 is then routed to a series of higher-order visual areas (HVA) where local
features are integrated to generate more abstract and global features. In the primates, these areas
are organized into two functionally distinct streams (Goodale and Milner, 1992; Mishkin and
Ungerleider, 1982). The dorsal or “where/action” stream stretches from V1 in the occipital lobe
forward into the parietal regions through V2, middle temporal area (MT), and medial superior
temporal (MST) area. It is sensitive to the motion information of objects and involved in the
planning and execution of the visually guided behaviors, such as arm reaching and grasping. On
the contrary, the ventral or “what/perception” stream goes through V1, V2, V3, V4 to areas of the
inferior temporal (IT) lobe, where the color, shape, size, and texture information of objects is
processed for object recognition and visual perception. These two parallel pathways are inter-
connected and ultimately reach the frontal lobe where the motion and form information of objects
are integrated to serve as the basis of decision making and motor planning. Consistently, along the
ventral and dorsal pathways, neurons exhibit increasingly larger receptive fields and more
sophisticated tuning properties.
4

Figure 1. Two major visual pathways
Visual information is collected and firstly processed in the retina lines the back of the eye on the
inside. Retinofugal axons go to the lateral geniculate nucleus (LGN) in the canonical imaging-
forming pathway and the superior colliculus (SC) in the non-imaging-forming pathway. LGN is
the major source of direct input for primary visual cortex (V1). V1 projects forward to higher order
visual cortical areas (HVA) and backward to the lateral posterior nucleus of the thalamus (LP). LP
receives strong inputs from SC and also sends feedforward projections to the HVA. Meanwhile,
LP sends modulatory inputs to layer (L)1 and deep layers of V1. LGN and LP both receive
modulatory feedback from V1 L6. Driving and modulatory projections are designated by solid and
dashed gray lines, respectively. Adapted from Fridman (2019).

1.2 More than feedforward pathways of sensory processing
Brain regions are rarely connected in a unidirectional manner. Instead, they mostly form direct or
indirect reciprocal connections. This holds true in the visual processing. Besides the canonical
ascending retina-visual thalamus-V1-HVA pathway, there are substantial and robust feedback
projections from HVA to V1 and thalamus and from V1 to thalamus. In the case of V1, L1 neurons
receives top-down feedback inputs from HVA and cingulate cortex (D’Souza et al., 2019; Ji et al.,
2015; Marques et al., 2018). Interestingly, L1 is also innervated by axons from higher order (HO)
5
thalamic nuclei (Rockland, 2019; Roth et al., 2016). The pyramidal tract (PT) neurons in deep L5
project to subcortical structures such as the midbrain, brainstem, and spinal cord (Harris and Mrsic-
Flogel, 2013; Harris and Shepherd, 2015). Concurrently, they send collaterals to the striatum and
HO nuclei (e.g. pulvinar). In comparison, the corticothalamic (CT) neurons in L6 project back to
first order (FO) thalamic nuclei (e.g. LGd) (Harris and Mrsic-Flogel, 2013; Harris and Shepherd,
2015). Note that SC and HO nuclei can directly project to HVA, bypassing V1 (Baldwin et al.,
2017; Bourne and Morrone, 2017; Kinoshita et al., 2019).
In addition to the excitatory pyramidal neurons (PRYs) that project across cortical layers and
cortical areas, local inhibitory neurons are equally important for sensory processing, although they
are less in number (~20%). These interneurons are mutually connected and heavily innervate the
excitatory neurons in their vicinity. Classified according to the molecular identity, parvalbumin
expressing (PV), somatostatin expressing (SOM) and vasoactive intestinal peptide expressing
(VIP) neurons account for the vast majority of inhibitory neurons. PV neurons target the peri-soma
region of PRYs and powerfully suppress neural activity. SOM neurons preferentially synapse on
the distal dendrites of PRYs. VIP neurons mainly inhibit PV and SOM neurons and can cause
disinhibition when activated (Pfeffer et al., 2013). The excitatory and inhibitory neurons work
together to sculpture neuronal activity and expand the dynamic range in response to the sensory
stimuli. Note that L1 is populated with exclusively inhibitory neurons, as well as long-range axons
and apical dendrites of L2/3 and L5 neurons.
Ultimately, during active interrogation, the sensory processing is further modulated by sensory
history (adaptation, facilitation, prediction, etc.), internal brain states (attention, emotion,
motivation, etc.) and locomotor activity (eye and head movement in primates, whisking and
6
sniffing in rodents, and locomotion) to enhance the salience of the signal while filtering out the
noise in the ambiguous context. This contextual and multisensory modulation of sensory
processing allows us to make the best use of sensory stimuli to understand and interact with the
external world for our survival and well-being.

1.3 First- and higher-order thalamic nuclei
The thalamus connects peripheral sensory organs and the cortex. Accumulating evidence suggests
that the thalamus is more than a passive relay center for sensory information; instead, it is an active
player in perception and cognition (Halassa and Sherman, 2019). Classical electron microscopy
has revealed common innervation pattern across thalamic nuclei of multiple modalities, that is the
glutamatergic synapses formed onto the thalamus are either “drivers” or “modulators” (Figure 2).
The driver inputs form synapses that are large in size but small in number and can largely drive
the spiking response of postsynaptic thalamic neurons. On the contrary, the synapses formed by
modulator inputs are small but large in number and have a modest impact on the response of
postsynaptic neurons (Sherman and Guillery, 1998). The FO and HO nuclei are primarily
determined by whether the driving force is from the periphery or cortex (Sherman and Guillery,
2002). LGd is a typical FO nucleus, as it receives driving inputs directly from the retina and
projects to L4 of V1. In contrast, pulvinar is considered a HO nucleus, in that it mainly innervates
L4 of HVA instead (and also projects to L1 and deep layers of V1) and receives robust feedback
from V1 L5 that can drive spiking responses. In fact, thalamic nuclei are mostly HO nuclei (Rubio-
Garrido et al., 2009).
7
In rodents, major HO nuclei include the lateral posterior nucleus (LP) for vision (Baldwin et al.,
2017; Zhou et al., 2017), the higher-order auditory thalamus surrounding the first-order ventral
division of the medial geniculate nucleus (MGv) for audition (Pardi et al., 2020), the medial
posterior nucleus (POm) for mechanical sense (Audette et al., 2019; Gharaei et al., 2020;
Jouhanneau et al., 2014; Zhang and Bruno, 2019), and the mediodorsal nucleus (MD) for motor
function (Vertes et al., 2015). The rodent LP is thought to be the functional homologue of the
pulvinar nucleus in the primates (Baldwin et al., 2017; Zhou et al., 2017). The functionality of HO
nuclei is still poorly understood. This thesis work aimed to elucidate the role of HO nuclei,
pulvinar/LP in particular, in the contextual and multi-modal modulation of visual and auditory
cortical processing. The following sections will summarize the known connectivity, subdivisions,
and function of pulvinar/LP.

Figure 2. Wiring pattern of first-order and higher-order thalamic nuclei
First-order nuclei receive strong driving inputs from the periphery and projects to L4 of the primary
sensory cortex. On the contrary, higher-order nuclei receive driving inputs from L5 of the primary
sensory cortex and projects to L4 of secondary sensory areas and meanwhile sends modulatory
8
inputs to L1 and deep layers of the primary sensory cortex. Both first-order and higher-order
thalamic nuclei receive modulatory inputs from L6 of the primary sensory cortex. Adapted from
Lee and Sherman (2008).

1.4 Connectivity of pulvinar/LP
As this thesis work focused on the rodent sensory processing, we firstly compile the afferent
projections coming to and efferent projections leaving from LP; we will compare the rodent LP
and the primate pulvinar in the next section.
LP is an excitatory HO thalamic nucleus in essence—98% of the LP neurons are excitatory
projection neurons, with merely 2% local inhibitory interneurons (Evangelio et al., 2018). LP
receives substantial inputs from a variety of cortical regions, including V1, HVA, posterior parietal
cortex, anterior cingulate cortex, and orbital cortex (Bennett et al., 2019; Juavinett et al., 2020;
Kamishina et al., 2009; Olavarria, 1979; Scholl et al., 2020; de Souza et al., 2020; Takahashi, 1985;
Zhou et al., 2018). LP is also heavily innervated by the subcortical inputs from the superficial and
intermediate layers of the SC (Beltramo and Scanziani, 2019; Bennett et al., 2019; Donnelly et al.,
1983; Gale and Murphy, 2014; Trojanowski and Jacobson, 1975). LP in turn projects to all these
cortical areas it receives inputs from. Moreover, LP innervates the amygdala and striatum (Wei et
al., 2015; Zhou et al., 2018). With the strong inputs from SC and the mutual connections with the
cortex, LP is well positioned to serve as a pivot for providing contextual and multi-modal
modulation to V1.
As a HO nucleus, LP exhibits laminar-specific innervation pattern to visual areas. LP axons
terminate in L1 and deep layers of V1, but target L4 of HVA (Herkenham, 1980; Roth et al., 2016;
9
Zhou et al., 2018). Furthermore, LP axons that project to V1 L1 send collaterals to L4 of HVA
(Nakamura et al., 2015). This pattern suggests that LP projections to V1 and HVA are
fundamentally different. Projections to L4 tend to be the “drivers”, and thus LP is likely to drive
spiking responses in HVA, whereas projections to L1 are mostly “modulators”, presumably
allowing LP to have a modulatory impact on the responses in V1. This hypothesis is supported by
previous experimental results that HO nuclei can evoke responses in higher sensory areas when
the primary sensory cortex is inactivated (Casanova and Molotchnikoff, 1990; Tohmi et al., 2014).
LP in turn is innervated by deep layers of V1. In specific, the L5 PT neurons form large synapses
onto LP neurons, providing the driving inputs to LP, while the L6 CT projections to LP are
modulatory, manifested in the relatively smaller synapses at the terminals (Sherman and Guillery,
1998). These two connections also display distinct electrophysiological properties in vitro
(Sherman and Guillery, 1998). Note that in awake mice, the L6 CT neurons differentially modulate
LGd and LP through parallel excitatory and inhibitory pathways in a dynamic and context-
dependent manner (Kirchgessner et al., 2020).
In addition to the cortical inputs, LP also receives subcortical projections from the excitatory wide-
field vertical cells (WFV) in the superficial and intermediate SC (Gale and Murphy, 2014). These
neurons exhibit large dendritic abors and receptive fields and respond the best to small and slowly
moving objects, indicating their role as motion detectors (Gale and Murphy, 2014; Ito and
Feldheim, 2018). Recent studies also found that WFV cells are involved in the prey capture and
innate defensive behavior (Hoy et al., 2019). As WFV cells per se are activated by V1 corticotectal
inputs (Masterson et al., 2019), LP is directly and indirectly modulated under V1 command.
Finally, LP receives direct but relatively weak inputs from the RGCs, the non-image-forming
10
melanopsin ‐ expressing intrinsically photosensitive RGCs in particular (Allen et al., 2016),
indicating that LP can be engaged in the circadian rhythm.

1.5 Subdivisions of pulvinar/LP
LP is not a homogenous structure. Early studies divide LP into the rostral-medial (Prm), caudal-
medial (Prm) and lateral part (Pl) based on cytoarchitecture and immunostaining (Zhou et al.,
2017). More recently, LP is divided into functionally subregions in accordance with neural
connectivity (Beltramo and Scanziani, 2019; Bennett et al., 2019) (Figure 3).
The two major projections from V1 and SC target two spatially distinct but overlapping LP
subregions along the anterior-posterior axis: the anterior corticorecepient zone and the posterior
tectorecepient zone, respectively. In specific, the anterior-ventral LP (aLP) is mainly mutually
connected with V1, the lateromedial area (LM) and the dorsal stream of visual pathway, including
the anterolateral area (AL), rostrolateral area (RL), anteromedial area (AM), and posteromedial
area (PM) (Bennett et al., 2019). On the other hand, the posterior-dorsal LP (pLP) primarily
receives inputs from SC and is interconnected with the ventral stream of visual pathway, including
the postrhinal area (POR) and laterointermediate area (LI) (Beltramo and Scanziani, 2019; Bennett
et al., 2019); the medial pLP also projects to the lateral amygdala (Zhou et al., 2018). Interestingly,
two coarse scale but complete retinotopic map (mirroring images to each other) are present in the
aLP and pLP, respectively (Bennett et al., 2019). A third small subregion, the meidal LP (mLP),
is dedicated for the projections to anterior cingulate and orbital cortex (Bennett et al., 2019).

11

Figure 3. The subdivisions and projection pattern of the rodent LP
Rodent LP is divided into three functionally distinct regions. The posterior LP (pLP) receives
inputs from SC and is interconnected with the ventral HVA, including postrhinal area (POR) and
laterointermediate area (LI). It also projects to amygdala. The anterior LP (aLP) is mutually
connected with V1, the lateromedial area (LM) and the dorsal stream of visual pathway, including
the anterolateral area (AL), rostrolateral area (RL), anteromedial area (AM), and posteromedial
area (PM). The medial LP (mLP) make reciprocal connections with orbital and cingulate cortex.
Adapted from Bennett et al. (2019) and Fridman (2019).

The pulvinar nucleus is the largest and most sophisticated thalamic nucleus in the primates. The
general projection pattern is conserved across species that pulvinar/LP projects to L1 of V1 and
L4 of HVA. Moreover, the spatial layout of LP subdivisions is also largely preserved in the primate
pulvinar. Based on cytoarchitecture, the pulvinar can be divided into three major areas: the lateral,
inferior, and dorsal/medial pulvinar, which are functionally similar to aLP, pLP, and mLP,
respectively. The lateral pulvinar has reciprocal connections with V1 and early visual areas, while
12
the inferior pulvinar receives inputs from superficial and intermediate SC, as well as visual cortices.
The dorsal/medial pulvinar similarly projects cingulate and orbital cortex. Nonetheless, it is worth
noting that there are subtle differences between the pulvinar and LP. First, the projection to
amygdala is from pLP in rodents but dorsal pulvinar in primates. More importantly, the di-synaptic
SC-LP-amygdala pathway is present only in the rodents but not primates (Wei et al., 2015). Second,
although primate inferior pulvinar receives substantial inputs from SC, its visual responses are
primarily driven by cortical inputs (Bender, 1983). In contrast, SC silencing abolishes most activity
in rodent pLP (Bennett et al., 2019). Third, ventral and dorsal visual areas preferentially target aLP
and pLP, respectively, while no clear segregation of these inputs is observed in lateral and inferior
pulvinar (Figure 4).

Figure 4. The subdivisions and projection pattern of the primate pulvinar
The primate pulvinar consists of three regions. The inferior pulvinar (Inf) receive inputs from SC.
Lateral pulvinar (Lat) and Inf are interconnected with V1 as well as higher visual areas along the
ventral and dorsal streams, including V4, middle temporal area (MT), and inferior temporal area
(IT). The dorsal medial pulvinar (Med) is connected with associative cortices, such as the posterior
13
parietal cortex, prefrontal cortex, and cingulate cortex and projects to the amygdala. Note the
similarity between the layout of rodent LP and primate pulvinar. Adapted from Pessoa and
Adolphs (2010) and Fridman (2019).

1.6 Function of pulvinar/LP
Given its extensive connections with a wide range of cortical and subcortical areas, it is not
surprising that pulvinar/LP profoundly contributes to various brain functions, from basic visual
processing to blindsight. Previous results on the functionality of pulvinar/LP are typically obtained
by traditional lesion studies and electrophysiological recordings. With the recent development of
transgenic mouse lines and optogenetic/chemogenetic approaches, we begin to further reveal the
neural circuits underlying the functionality of pulvinar/LP.
1.6.1 Vision
Pulvinar/LP is considered a HO visual thalamus as it shows robust evoked spiking responses to
visual stimulation. In anesthetized mouse LP, neurons exhibit diverse temporal response profile to
full-field bright light steps. The majority of neurons show transient responses to the onset and
offset of light steps, while other neurons respond sustained or delayed responses (Allen et al.,
2016). In addition to the general responsiveness, neurons in LP and pulvinar with large visual
receptive fields organized into retinotopic maps. Consistently, they are more broadly tuned to
orientation and direction than V1 neurons (Durand et al., 2016; Roth et al., 2016). LP neurons are
also responsive to looming stimuli (expanding dark disks) that trigger innate defensive behavior
(Wei et al., 2015). A recent study took a closer look at the axonal terminals of LP to V1 and found
that LP axons have larger receptive fields compared with LGd axons and V1 neurons. LP axons
are also significantly less selective to orientation or direction than V1 neurons, similar to LGd
14
axons (Roth et al., 2016). The exact visual response properties of LP neurons are highly dependent
on the subdivision. For instance, pLP neurons prefer small flashing spots and is sensitive to the
motion contrast of the figure versus ground in the visual stimuli. In contrast, aLP and mLP neurons
are biased towards larger spots and insensitive to the differential motion (Bennett et al., 2019).
1.6.2 Motor
The primate pulvinar is well documented to be engaged in the goal-directed action behavior. A
classical study has shown that the pulvinar activity is correlated with the intentional arm reaching
and eye movements towards attended objects (Acuña et al., 1983). Unilateral inactivation of
pulvinar causes disrupts in the visually guided reaching of the contralateral arm and reduces
contralateral saccades (Wilke et al., 2010, 2013). Similarly, a human patient with selective damage
to the mediodorsal pulvinar caused by stroke is reported to have difficulty in arm reaching and
grasping, but with normal general visual and motor function (Wilke et al., 2018). Interestingly,
human patients or primates with damaged V1 have “blindsight”, that is, residual vision guides
goal-directed eye or forelimb movements towards targets in the blind field without awareness.
Inactivation of the neurons or SC axons in the ventrolateral pulvinar greatly impairs the visually
guided saccades to the blind field in blindsight monkeys (Kinoshita et al., 2019), suggesting the
SC-pulvinar pathway contributes to blindsight. In rodents, a recent study has demonstrated that LP
signals the discrepancy of actual optic flow and predicted optic flow based on running speed (Roth
et al., 2016). A subset of LP axonal boutons in V1 respond specifically when the actual optic flow
and running speeding are de-coupled, i.e., visuomotor mismatch, while others respond
independently to either optic flow or locomotion. Together, these data suggest that pulvinar/LP
combines motor command, intentional signals, and predictive coding to optimize the motor output,
even in the absence of V1 inputs.
15
1.6.3 Innate defensive behavior
As pulvinar/LP projects to the lateral amygdala, it is naturally to assume it is involved in emotion-
related behavior. Indeed, neurons in the pulvinar are sensitive to threatening snake images
(Almeida et al., 2015; Gomes et al., 2018; Le et al., 2016; Van Le et al., 2013; Soares et al., 2017),
and pulvinar lesions significantly disrupt fear processing (Bertini et al., 2018; Ward et al., 2005).
A recent computational study using human neuroimaging data also demonstrates that the
subcortical SC-pulvinar-amygdala pathway can facilitate fear recognition (McFadyen et al., 2019).
In rodents, researchers have further identified a di-synaptic pathway from SC to lateral amygdala
via pLP (Wei et al., 2015). Activation of this pathway directly triggers innate freezing behavior in
the absence of looming stimuli (Wei et al., 2015).
1.6.4 Selective attention
As the primate pulvinar is extensively interconnected with higher cortical areas that are engaged
in selective attention, it is not surprising that the pulvinar is shown to be related and required for
attention by recordings and lesion studies. One mechanism of attention is to selectively route
behaviorally relevant information through cortical networks. This is accomplished by the
synchronized neuronal activity across cortical regions (Fiebelkorn and Kastner, 2020; Kastner et
al., 2020). A previous study reported that the pulvinar synchronized neural activity and regulated
information transmission according to attention allocation when monkeys were performing a
visuospatial attention task (Saalmann et al., 2012). In addition, inactivation of ventrolateral
pulvinar decreased attentional effects on the response level and gamma synchrony in V4 and
caused severe behavioral deficits (Zhou et al., 2016). Meanwhile, pulvinar inactivation increased
low-frequency cortical oscillations, an indicator of inattention or sleep (Zhou et al., 2016).
Moreover, the mediodorsal pulvinar is found to coordinate the coupling of macaque frontal eye
16
fields (FEF) and lateral intraparietal area (LIP) during rhythmic spatial attention (Fiebelkorn et al.,
2019).
In line with these data from electrophysiological recordings, lesion studies have demonstrated that
pulvinar damage can lead to spatial neglect in monkeys and humans, i.e., difficulty in perceiving,
attended to, and reacting in the affected visual field (Arend et al., 2008a; Karnath et al., 2002;
Petersen et al., 1987; Wilke et al., 2010; Zihl and von Cramon, 1979). Interestingly, pulvinar lesion
can also lead to auditory neglect and problems in understanding or expressing speech (Hugdahl et
al., 1991; Maeshima and Osawa, 2018). Also, pulvinar lesion can cause deficits in ignoring
distractors (Desimone and Duncan, 1995). Besides, the pulvinar is considered to be involved in
the visual awareness (Pessoa and Adolphs, 2010), as recordings and neuroimaging studies found
that pulvinar activity is enhanced when animals or human subjects report detecting a visual
stimulus (Padmala et al., 2010; Wilke et al., 2009).
1.6.5 Decision making
The pulvinar is also indispensable for the goal-directed decision making, as suggested by its
connection with the frontal lobe (Dominguez-Vargas et al., 2017; Jaramillo et al., 2018; Wang and
Pleger, 2020; Yang and Burwell, 2020). To excel in such reward-related associative learning tasks,
animals should be able to evaluate the valence of the incoming stimuli and predict forthcoming
rewarding events. Indeed, researchers have found that the pulvinar neurons exhibit both early and
late responses during the learning process. The early responses appeared shortly after the stimulus
(cue) onset and resisted extinction, while the late responses increased during the delay period and
peaked right before the delivery of reward (Komura et al., 2001). These data indicate the pulvinar
can encode the valence of a stimulus to predict and anticipate the reward.
17
Interestingly, the pulvinar is also reported to encode confidence of a subject. Researchers have
found that during a visual categorization “opt-out” task, pulvinar activities decreased when
monkeys chose the “escape” option. Moreover, pulvinar silencing significantly increased the
“escape” rate, suggesting a decreased confidence level, suggesting that the pulvinar signals a
subject’s certainty or confidence of visual categorization (Komura et al., 2013).

18
Chapter 2: A differential circuit via colliculo-pulvinar
pathway enhances feature selectivity in visual cortex
through surround suppression

2.1 Introduction
In the mammalian visual system, sensory information from the retina undergoes a hierarchy of
processing by streaming into two major bottom-up pathways (Nassi and Callaway, 2009). One is
through the dorsal lateral geniculate nucleus (dLGN), the first-order thalamic nucleus, to reach the
primary visual cortex (V1) and subsequent higher visual cortical areas to generate visual
perception. The other is through the superior colliculus (SC) in the midbrain to generate certain
visually guided behaviors (Basso and May, 2017; Shang et al., 2015; Wei et al., 2015; Yilmaz and
Meister, 2013; Zingg et al., 2017). While visual cortical feedback modulates SC activity via
corticocollicular projections (Liang et al., 2015; Zhao et al., 2014) and thus influences visually
induced behaviors, e.g. looming stimulus induced defensive freezing (Yilmaz and Meister, 2013;
Zingg et al., 2017), SC activity in turn may have impacts on visual cortical processing (Ahmadlou
et al., 2018; Beltramo and Scanziani, 2019; Ogino and Ohtsuka, 2000; Tohmi et al., 2014). The
lateral posterior nucleus (LP) of thalamus, which is generally considered as the rodent homologue
of the primate pulvinar (Zhou et al., 2017), appears well-poised to bridge SC and visual cortex and
to play a role in mediating the SC impact on visual cortex. This possibility is strongly suggested
by the anatomical evidence that among thalamic nuclei, the LP receives the strongest projection
from SC

(Gale and Murphy, 2014; Lane et al., 1997; Redgrave et al., 1993; Stepniewska et al.,
2000; Wei et al., 2015; Zingg et al., 2017), and that it projects to both primary and secondary visual
19
cortices (Beltramo and Scanziani, 2019; Bennett et al., 2019; Ganz et al., 2012; Juavinett et al.,
2020; Kaas and Lyon, 2007; Nakamura et al., 2015; Oh et al., 2014; Roth et al., 2016; Shipp, 2001;
Wong et al., 2009; Zhou et al., 2018).
Previous studies of pulvinar have mostly been focused on higher visual functions such as visual
attention, perceptual suppression as well as planning and selection of visually guided eye and hand
movements (Dominguez-Vargas et al., 2017; Grieve et al., 2000; Saalmann et al., 2012; Soares et
al., 2017; Wilke et al., 2009, 2010; Zhou et al., 2016). Recently, there has been evidence that
LP/pulvinar transmits moving visual information as well as motor-related signals to secondary or
higher visual cortices, contributing to the visual response properties there (Beltramo and Scanziani,
2019; Bennett et al., 2019; Berman and Wurtz, 2011; Tohmi et al., 2014). However, the role of
LP/pulvinar in fundamental information processing in V1 has largely remained unclear. Different
from the dLGN which relays ascending visual information mainly to layer 4 of V1, the excitatory
projections of LP to V1 primarily terminate in layer 1 and deep layers, while its projections to
secondary visual cortices (V2) mainly terminate in layer 4 (Herkenham, 1980; Roth et al., 2016;
Zhou et al., 2017). Cortical layer 1 is only populated by inhibitory neurons (Jiang et al., 2013) and
has been implicated in the modulation of V1 responses in superficial layers by long-range
projections from other cortical areas (Jiang et al., 2013; Zhang et al., 2014; Zhou et al., 2014).
This innervation pattern of LP axons in V1 raises a hypothesis that LP might play a modulatory
role in influencing V1 responses, rather than an excitatory “driver” role as suggested in a previous
study of pulvinar in galagos (Purushothaman et al., 2012). Consistent with this idea, functional
studies of LP neurons and their axon terminals in layer 1 of V1 have shown that these neurons
exhibit extremely broad receptive fields (RFs) (Allen et al., 2016; Bender, 1982; Berman and
20
Wurtz, 2011; Chalupa et al., 1983; Durand et al., 2016; Roth et al., 2016), suggesting that LP is
able to provide contextual information to modulate V1 responses (Roth et al., 2016).
In this study, by applying pharmacological and optogenetic manipulations in awake mice, we
explored the functional contribution of LP to visual processing in V1 and the potential underlying
circuit basis. We demonstrate that LP activity enhances fundamental visual processing functions
via a feedforward, surround-suppression mechanism mediated by layer 1 inhibitory neurons. Our
study further suggests that the parallel retina-SC-LP-V1 and retina-geniculate-V1 visual pathways
form a bottom-up “differential” circuit by which V1 selectivity can be largely maintained despite
varying visual noise background.

2.2 Results
2.2.1 LP silencing impairs visual discrimination task performance in mice
To study potential functional contributions of LP, we first employed a visual discrimination task.
We trained mice to lick water for reward upon detection of a Go signal (a visual grating pattern of
vertical orientation) but refrain from licking upon detection of a No-Go signal (grating of
horizontal orientation) (Figure 5A). The animals could learn this task over one-week training, as
manifested by the increasing correct performance rate (hit + correct rejection) over training
sessions before reaching a plateau level (Figure 5B, top). In the well-trained animals, silencing
LP bilaterally either by local infusion of muscimol or by expressing DREADDi receptors and
administering the DREADD ligand, CNO (see Star Methods), significantly lowered the correct
performance rate, whereas injections of CNO in GFP control mice had no effect (Figure 5B). The
21
poorer performance resulted from increased false alarm rate, while the miss rate was unchanged
(Figure 6A), suggesting that the ability to distinguish the visual patterns was compromised after
LP silencing, but the animals were still engaged in the behavioral task. While this impairment of
performance may be attributed to deficits in visual attention after silencing LP (Zhou et al., 2016),
it is also possible that LP activity directly contributes to normal information processing in V1,
which remains to be carefully investigated.
2.2.2 LP silencing reduces functional selectivity in L2/3 of V1
To understand how exactly LP activity influences visual information processing, we carried out in
vivo single-cell loose-patch recordings in awake mouse V1 and examined visual responses of the
same neurons before and after manipulating LP activity. The mouse was head-fixed but allowed
to run freely on a flat rotatable plate (Figure 5C, top left), following our previous studies (Chou
et al., 2018; Xiong et al., 2015). We first silenced LP by slowly infusing 100 nl of 0.5%
bupivacaine (Lee et al., 2008) via an implanted cannula (Figure 5C, bottom left; Figure 6A).
Bupivacaine effectively eliminated both evoked and spontaneous spikes of LP neurons for at least
30 min (Figure 5C, right) (see 2.4).
For each recorded neuron, we applied full-field moving gratings of 12 directions (0° to 330° with
30° steps) to measure orientation selectivity and direction selectivity (see 2.4). As shown by the
polar graphs of two example L2/3 neurons and the average normalized orientation tuning curves,
after silencing LP, orientation selectivity was weakened due to increases of evoked response level
at all orientations (Figure 5D). Such change was evident in most of L2/3 neurons we recorded, as
shown by a reduced global orientation selectivity index (gOSI) (Figure 5E and Figure 6F) and
conventional orientation selectivity index (OSI) (Neill and Stryker, 2008; Tan et al., 2011) (Figure
22
6F). Direction selectivity was also weakened, as shown by a lowered global direction selectivity
index (gDSI) (Figure 5F and Figure 6G) and conventional direction selectivity index (DSI)
(Figure 6G). Notably, the orientation preference was not altered after LP silencing (Figure 5G).
In the meantime, the change in evoked firing rate after LP silencing was the same for preferred
and orthogonal orientations (Figure 5H), suggesting that the change of orientation tuning can be
described as a subtractive effect.
We also measured the overall visual response level by presenting a series of full-screen, flashing
dense white-noise patterns. The visual response level was significantly increased after LP
silencing (Figure 5I and Figure 6E; see control in Figure 6C), consistent with the grating results.
The spontaneous firing rate was increased from 2.07 ± 1.18 Hz to 3.03 ± 2.15 Hz (mean ± SD, n
= 27 cells from 6 mice, p = 0.029, paired t-test) after LP silencing. Therefore, our results suggest
that the overall effect of LP activity on V1 responses in L2/3 is suppressive, which results in an
enhancement of orientation and direction selectivity (Figure 6F-G), i.e. improvement of visual
processing.
In another cohort of animals, we examined size tuning of V1 neurons by presenting gratings
(centered at the RF center, moving to the optimal direction) of different sizes (ranging from 3° to
51° in radius). As shown by an example L2/3 neuron (Figure 5J), before LP silencing the cell
had an optimal response to 10° size and its responses to larger stimuli were suppressed. In other
words, the cell exhibited size tuning and surround suppression (Nienborg et al., 2013; Sugawara
and Nikaido, 2014). After LP was silenced, the responses were increased, especially for larger
stimulus sizes (Figure 5J). In other words, surround suppression was weakened. We quantified
the strength of surround suppression with a surround suppression index (SSI) (Sugawara and
23
Nikaido, 2014), which ranges from 0 (no suppression) to 1 (complete suppression by the largest
stimulus). The SSI of recorded L2/3 neurons was significantly reduced after silencing LP (Figure
5K and Figure 6H). In particular, in several cells tested, size tuning was completely lost, i.e. SSI
became zero (Figure 5K). As shown by the population average of size tuning curves from all the
cells, the increase of evoked firing rate after LP silencing was relatively higher for larger than
smaller stimulus sizes (Figure 5K, inset). At the population level, the distribution of optimal sizes
was consistent with what has been reported before (Nienborg et al., 2013; Sugawara and Nikaido,
2014; Vaiceliunaite et al., 2013) (Figure 6D). We also directly measured spatial receptive fields
(RFs) using a reverse correlation method (Jones and Palmer, 1987). We found that the size (in
radius) of the RF subfield of dominant contrast (On or Off) increased after LP silencing (Figure
5L-M and Figure 6I). These results indicate that LP activity contributes to surround suppression
and increases the size and spatial selectivity of V1 L2/3 neurons.
Interestingly, all the above effects observed in L2/3 (increased response levels, reduced orientation
and direction selectivity, as well as reduced size and spatial selectivity after LP silencing) were
absent in L4. Basically, there were no changes in visual selectivity, response level or RF subfield
size in L4 neurons after silencing LP (Figure 5N-W and Figure 61J-N). This absence of effects
on L4 neurons argues against the possibility that the drug infusion had affected the dLGN, which
would result in a reduction of response level in L4 (Reinhold et al., 2015) . Rather, it suggests that
the dLGN was intact and that the influence of LP activity on V1 neurons was specific to superficial
layers.
24

Figure 5. Visual response properties of V1 neurons after pharmacologically silencing LP
(A) Paradigm of visual discrimination task. The mouse was trained to lick to receive water reward
within a 1-s response window 0.5 s after presenting one of two visual cues. Behavioral
performance was quantified with a correct rate (the percentage of hit plus correct rejection trials).
(B) Top, correct rate over training days for two example animals. Arrows indicate LP silencing
via bilateral infusion of muscimol or CNO on the last day, as well as control saline infusion on the
day before last. Bottom, summary of correct rates one day before (saline) and on the day of LP
silencing. Data points for the same animal are connected by a line. ***p < 0.001; N.S., not
significant, paired t-test, n = 4 mice for muscimol silencing, 4 DREADDi-expressing mice, and 5
GFP-expressing control mice. (C) Left, awake recording with the animal implanted with a cannula
for drug infusion (top) and an example fluorescence image showing the spread of drug infusion
25
(bottom). Scale bar: 500 μm. Right, visually evoked (top) and spontaneous (bottom) firing rates of
recorded LP neurons before and 30 min after drug infusion. ***p < 0.001, paired t-test, n = 12
cells from 2 mice. Bar = SD. (D) Polar plots of evoked firing rates at different directions of moving
gratings (full-field) for two example L2/3 neurons (top inset, axis limit: 10 Hz) and average
normalized orientation tuning curves (with Gaussian fit) of the recorded L2/3 population (n = 28
cells from 5 mice) before (black) and after (red) silencing LP with bupivacaine. Bar = SEM. (E)
The gOSI values of recorded L2/3 neurons after versus before silencing LP. The dashed line is the
unity line. Solid symbols label cells showing significant changes after drug administration
(permutation test with bootstrapping, p < 0.05). ***p < 0.001, paired t-test, n = 28 cells from 5
mice. (F) The gDSI values after versus before silencing LP. ***p < 0.001, paired t-test, n = 28
cells from 5 mice. (G) Preferred orientation after versus before silencing LP (p = 0.37, paired t-
test, n = 28 cells from 5 mice). (H) Changes of evoked firing rate at the preferred and orthogonal
orientations. Data points for the same cell are connected with a line. Bar = SD. N.S., not significant,
p = 0.40, paired t-test, n = 28 cells from 5 mice. (I) Response levels to flash white-noise patterns
after versus before LP silencing. ***p < 0.001, paired t-test, n = 27 cells from 6 mice. (J) Firing
rates (averaged by trials) evoked by increasing stimulus sizes (in radius) before (black) and after
(red) silencing LP for an example L2/3 neuron. Bar = SD. (K) Surround suppression index (SSI)
for recorded L2/3 neurons after versus before silencing LP. **p = 0.002, paired t-test, n = 18 cells
from 4 mice. Inset, population average of normalized size tuning curves aligned by the peak
response before (dark gray) and after (red) silencing LP. ***p < 0.001, **p < 0.01, *p < 0.05,
paired t-test, n = 18 cells from 4 mice. Bar = SEM. (L) Spatial RF revealed by the spike-triggered
average of sparse noise stimuli using reverse correlation for an example L2/3 neuron before (top)
and after (bottom) silencing LP. Left, raw RF subfield. Right, subfield fitted with 2D elliptical
Gaussian. White and red ovals mark the boundary of the subfield (defined by the 2σX and 2σY of
the Gaussian) before and after silencing LP, respectively. Scale bar: 8°. (M) RF subfield sizes (in
radius) of recorded L2/3 neurons after versus before silencing LP. ***p < 0.001, paired t-test, n =
22 cells from 3 mice. (N –W) No effects on L4 neurons. Data are displayed in same manners as in
(D –M). Axis limit: 20 Hz in (N) (top inset). Statistics: gOSI in (O) (p = 0.61, paired t-test, n = 19
cells from 5 mice), gDSI in (P) (p = 0.68, paired t-test, n = 19 cells from 5 mice), preferred
orientation in (Q) (p = 0.85, paired t-test, n = 19 cells from 5 mice), changes of evoked response
level at preferred versus orthogonal orientation in (R) (p = 0.25, paired t-test, n = 19 cells from 5
mice), overall visual response level in (S) (p = 0.87, paired t-test, n = 18 cells from 6 mice), SSI
in (U) (p = 0.65, Wilcoxon signed-rank test, n = 17 cells from 4 mice), and RF subfield size in (W)
(p = 0.51, paired t-test, n = 18 cells from 3 mice).

26

Figure 6. Summary of changes in response properties in V1 before and after LP silencing
with bupivacaine
(A) Changes in false alarm rate (left, percentage of licking trials in total trials with No-Go signal)
and miss rate (right, percentage of no-licking trials in total trials with Go signal) in mice before
and after LP silencing with pharmacological and chemogenetic approaches. ***P < 0.001, N.S.,
not significant, paired t-test, n = 4 mice in muscimol silencing group and n = 4 mice in DREADDi
+ CNO silencing group. (B) Left, experimental paradigm: injecting bupivacaine into LP and
recording from V1. Right, overlying color showing spread of bupivacaine in 12 animals. Pink color
indicates the spread in each individual animal. (C) Distribution of percentage changes in visual
response level before and after saline (gray, n = 12 cells) versus bupivacaine (red, n = 25 cells)
injection in LP. P < 0.001, Kolmogorov–Smirnov test. (D) Distribution of optimal grating radius
which evoked the maximum response (n = 99 L2/3 cells). (E –I) Mean evoked firing rate, gOSI/OSI,
gDSI/DSI, SSI, and dominant subfield size for V1 L2/3 neurons respectively before and after LP
silencing. **P < 0.01, ***P < 0.001. Bar = SEM. (J–N) Similar as (E –I) but plotted for V1 L4
neurons. N.S., not significant.

2.2.3 Optogenetic silencing of LP reduces visual selectivity in L2/3
In order to have a better spatial and temporal control of LP activity, we used an adeno-associated
viral (AAV) vector to express an inhibitory opsin, archaerhodopsin (ArchT), in LP (Figure 7A-B
and Figure 8A). Green LED light (530 nm) was delivered via an implanted optic fiber to silence
LP neurons, which covered the entire duration of the visual stimulus presented. LED-on and LED-
27
off trials were interleaved. We then compared visual responses between the LED-on and LED-off
(control) trials in the same recorded neurons. Firstly, we shed LED light on LP and recorded
spiking activity in LP or dLGN (see 2.4). We found that visual responses in LP were nearly
abolished in LED-on trials, whereas those in dLGN were not affected (Figure 7C). These results
verified the effectiveness and spatial specificity of optogenetic LP silencing. Next, we recorded
from V1 neurons. Essentially, the optogenetic silencing of LP reproduced the results of
pharmacological silencing with bupivacaine: orientation and direction selectivity was weakened
without changing preference (Figure 7D-H and Figure 8D-E), the overall visual response level
was increased (Figure 7I and Figure 8C; see control in Figure 8B), and size tuning/surround
suppression was weakened (Figure 7J-K and Figure 8F). These effects were observed only in
L2/3, but not in L4 (Figure 7L-S and Figure 8G-J). In addition, spontaneous firing rate was
increased from 2.00 ± 1.10 Hz in the LED-off condition to 3.56 ± 2.51 Hz in the LED-on condition
(n = 26 cells from 4 mice, p = 0.0097, paired t-test). The optogenetic silencing experiments further
demonstrate that LP is engaged by visual input to improve V1 processing in L2/3.
28

Figure 7. Effects of LP silencing with ArchT activation on V1 response properties
(A) Schematic recording setup. Green LED light was applied via an implanted optic fiber. (B) Left,
schematic recording in V1 while optogenetically silencing LP. Right, an example image showing
the expression of ArchT in LP. White dashed line marks the boundaries of dLGN and LP. Scale
bar: 500 m. (C) Visually evoked firing rates of neurons in LP (left, ***p < 0.001, paired t-test, n
= 13 cells from 2 mice) and in dLGN (right, N.S., not significant, p = 0.36, paired t-test, n = 13
cells from 2 mice) and without (OFF) and with (ON) LED light. (D) Polar plots of orientation
tuning for two example L2/3 neurons without (black) and with (green) LED inactivation of LP
neurons (top, axis limit is 20 Hz) and average tuning curves for the recorded L2/3 population (n =
26 cells from 4 mice). (E) The gOSI values of recorded L2/3 neurons with versus without LED
stimulation. Solid symbols mark neurons with significant changes (permutation test with
bootstrapping, p < 0.05). ***p < 0.001, paired t-test, n = 26 cells from 4 mice. (F) The gDSI values.
***p < 0.001, paired t-test, n = 26 cells from 4 mice. (G) Preferred orientation with versus without
LED stimulation. N.S., not significant, p = 0.16, paired t-test, n = 26 cells from 4 mice. (H)
Changes of evoked firing rate at preferred and orthogonal orientations. N.S., p = 0.27, paired t-test,
n = 26 cells from 4 mice. Bar = SD. (I) Response levels to flash noise patterns with versus without
LED stimulation. ***p < 0.001, paired t-test, n = 30 cells from 6 mice. (J) SSI values with versus
29
without LED stimulation. ***p < 0.001, paired t-test, n = 27 cells from 6 mice. (K) Population
average of normalized size tuning curves of L2/3 neurons without (dark gray) and with (green)
LED stimulation. ***p < 0.001, **p < 0.01, *p < 0.05, paired t-test, n = 27 cells from 6 mice. Bar
= SEM. (L –S) For L4 neurons. Data are displayed similarly to (D –K). Statistics: gOSI in (M) (p
= 0.11, paired t-test, n = 19 cells from 4 mice), gDSI in (N) (p = 0.85, paired t-test, n = 19 cells
from 4 mice), preferred orientation in (O) (p = 0.49, paired t-test, n = 19 cells from 4 mice), changes
of evoked response level at preferred and orthogonal orientations in (P) (p = 0.084, paired t-test, n
= 19 cells from 4 mice), response level in (Q) (p = 0.54, Wilcoxon signed-rank test,, n = 18 cells
from 6 mice), and SSI in (R) (p = 0.25, Wilcoxon signed-rank test, n = 18 cells from 6 mice).

Figure 8. Summary of changes in response properties in V1 without and with ArchT
inactivation of LP
(A) Left, experimental paradigm: injecting AAV-ArchT into LP, shining light on LP, and
recording from V1. Right, summary of spread of ArchT expression in LP (n = 10 animals). (B)
Distribution of percentage changes in visual response level in mice injected with AAV-GFP in LP
(control, n = 14 cells) versus those injected with AAV-ArchT in LP (experimental group, n = 26
cells) without and with green LED stimulation. P < 0.001, Kolmogorov–Smirnov test. (C –F) Mean
evoked firing rate, gOSI/OSI, gDSI/DSI, and SSI for V1 L2/3 neurons respectively before and
after LP silencing. ***P < 0.001. Bar = SEM. (G –J) Similar as (C –F) but plotted for V1 L4
neurons. N.S., not significant.

2.2.4 Activation of LP improves visual processing in L2/3
30
To further understand the direct effect of LP activity on V1 responses, we injected AAV encoding
channelrhodopsin2 (ChR2) into LP, and applied blue LED light pulses (473 nm, 20-ms pulse
duration, 20 Hz) to activate LP while recording in V1 (Figure 9A and Figure 10A). The
optogenetic stimulation effectively evoked spiking responses of LP neurons while left the dLGN
unaffected (Figure 9B). In V1 L2/3, we observed effects opposite to the inactivation of LP:
orientation and direction selectivity was strengthened (Figure 9C-G and Figure 10D-E), the
visual response level was reduced (Figure 9H and Figure 10C; see control in Figure 10B), and
size tuning/surround suppression was enhanced (Figure 9I-J and Figure 10F). In addition,
spontaneous firing rate was reduced from 2.58 ± 1.71 Hz in the LED-off condition to 1.42 ± 0.68
Hz in the LED-on condition (n = 26 cells from 6 mice, p = 0.0043, paired t-test). Furthermore,
similar to the inactivation of LP, these effects on V1 responses were only observed in L2/3 but not
L4 neurons (Figure 9K-R and Figure 10G-J). The optical stimulation did not result in significant
eye movements (Figure 10K). These optogenetic activation experiments further support the
conclusion that LP activity exerts a suppressive effect on the responses of V1 L2/3 neurons, which
results in enhanced surround suppression and improvements of tuning selectivity of these neurons.
The overall suppressive effect of LP could be described as subtractive rather than divisive, as the
orientation tuning curve of L2/3 neurons appeared to be shifted down when activating LP (Figure
9C,G) while shifted up when inactivating LP (Figure 7D,H).
31

Figure 9. Changes of visual response properties of V1 neurons with optogenetic activation
of LP
(A) Left, schematic recording in V1 while optogenetically activating LP. Right, example image
showing the expression of ChR2 in LP. Scale bar: 400 m. (B) Firing rates of neurons in LP (left,
***p < 0.001, paired t-test, n = 12 cells from 2 mice) and dLGN (right, N.S., not significant, p =
0.18, paired t-test, n = 16 cells from 2 mice) without (OFF) and with (ON) LED activation of LP.
(C) Polar plots of orientation tuning for two example L2/3 neurons without (black) and with (blue)
LED stimulation of ChR2-expressing LP neurons (top inset, axis limit is 15 Hz) and average tuning
curves for the recorded L2/3 population (n = 26 cells from 6 mice). (D) The gOSI values of
recorded L2/3 neurons with versus without LED stimulation. Solid symbols mark neurons with
significant changes (permutation test with bootstrapping, p < 0.05). ***p < 0.001, paired t-test, n
= 26 cells from 6 mice. (E) The gDSI values. ***p < 0.001, paired t-test, n = 26 cells from 6 mice.
(F) Preferred orientation with versus without LED stimulation. N.S., p = 0.21, paired t-test, n = 26
cells from 6 mice. (G) Changes of evoked firing rate at preferred and orthogonal orientations. N.S.,
p = 0.14, paired t-test, n = 26 cells from 6 mice. Bar = SD. (H) Response levels to flash noise
patterns with versus without LED stimulation. *p = 0.014, paired t-test, n = 20 cells from 4 mice.
32
(I) SSI values with versus without LED stimulation. **p = 0.0032, paired t-test, n = 34 cells from
8 mice. (J) Population average of normalized size tuning curves of L2/3 neurons without (dark
gray) and with (blue) LED stimulation. ***p < 0.001, **p < 0.01, *p < 0.05, paired t-test, n = 34
cells from 8 mice. Bar = SEM. (K –R) For L4 neurons. Data are displayed similarly to (C –J). Axis
limit in (K) is 15 Hz. Statistics: gOSI in (L) (p = 0.95, paired t-test, n = 16 cells from 6 mice),
gDSI in (M) (p = 0.91, paired t-test, n = 16 cells from 6 mice), preferred orientation in (N) (p =
0.64, paired t-test, n = 16 cells from 6 mice), changes of evoked response level at preferred and
orthogonal orientations in (O) (p = 0.89, paired t-test, n = 16 cells from 6 mice), response level in
(P) (p = 0.84, paired t-test, n = 18 cells from 4 mice), and SSI in (Q) (p = 0.25, paired t-test, n =
18 cells from 8 mice).

Figure 10. Summary of changes in response properties in V1 without and with ChR2
activation of LP
(A) Left, experimental paradigm: injecting AAV-ChR2 into LP, shining light on LP, and recording
from V1. Right, summary of spread of ChR2 expression (n = 14 animals). (B) Distribution of
percentage changes in visual response level in mice injected with AAVGFP in LP (control, n = 16
cells) versus those injected with AAV-ChR2 in LP (experimental group, n = 31 cells) without and
with blue LED stimulation. P < 0.001, Kolmogorov–Smirnov test. (C –F) Mean evoked firing rate,
gOSI/OSI, gDSI/DSI, and SSI for V1 L2/3 neurons respectively before and after LP activation. *P
< 0.05, **P < 0.01, ***P < 0.001. Bar = SEM. (G –J) Similar as (C –F) but plotted for V1 L4
33
neurons. N.S., not significant. (K) Average horizontal (left) and vertical (right) eye movements
aligned to the onset of optical activation of LP.

2.2.5 LP axons innervate L1 inhibitory neurons in V1
We next examined LP projections to V1. By injecting a retrograde dye CTB in V1 (Figure 11A),
we observed that labeled neurons were distributed nearly across the entire LP in the anterior-
posterior axis, although more neurons in the rostral part of LP project to V1 compared to its caudal
part (Figure 11B). This rostral-caudal gradient is consistent with recent anatomical results
(Beltramo and Scanziani, 2019; Bennett et al., 2019). LP contains predominantly (> 95%)
excitatory neurons (Evangelio et al., 2018). Since these LP projections to visual cortex are
excitatory (Roth et al., 2016; Zhou et al., 2018), how do they produce suppressive effects on V1
L2/3 neurons? By injecting AAV-GFP in LP, we observed that in V1 the anterogradely labeled
LP axons were densely distributed in layer 1 besides some distributions in deep layers (Figure
11C), consistent with previous results (Bennett et al., 2019; Roth et al., 2016; Zhou et al., 2018).
Therefore, they may directly innervate L1 neurons which are inhibitory (Jiang et al., 2013). To
test this idea, in slice preparations we made whole-cell recordings from V1 pyramidal cells or L1
GABAergic neurons, labeled by crossing GAD2-Cre mice with the Ai14 tdTomato reporter, while
optically stimulating ChR2-expressing LP axons (Figure 11D). Cuts were made in the tissue along
boundaries between V1 and V2, as to prevent potential feedback inputs from V2 to V1. With TTX
and 4-AP present in the bath solution, we found that blue LED light evoked monosynaptic
excitatory postsynaptic currents (EPSCs) in L1 GABAergic neurons, which could be blocked by
CNQX (Figure 11E). A cell is considered innervated if the peak amplitude of EPSCs is larger
than 3SD above the baseline fluctuation (see 2.4). Such EPSC response with a relatively large
amplitude (257.6 ± 192.9 pA, mean ± SD, n = 13 cells from 5 mice) was observed in a majority of
34
recorded L1 inhibitory neurons (Figure 11F, red), indicating that indeed LP axons can directly
innervate L1 inhibitory neurons. In comparison, most pyramidal neurons in L2/3 or L4 were not
innervated by LP axons, and those innervated by LP axons only exhibited weak EPSCs (Figure
11F, black). We also examined parvalbumin (PV) and somatostatin (SOM) inhibitory neurons in
L2/3 and L4, which were labeled by crossing PV-Cre or SOM-Cre mice with Ai14 reporters
respectively. These inhibitory neurons were essentially not innervated by LP axons (Figure 11G).
In the normal bath solution without TTX and 4-AP, we observed that blue light activation of LP
axons in V1 induced a hyperpolarizing membrane potential response (-4.0 ± 2.8 mV, n = 7 cells
from 3 mice) in L2/3 pyramidal neurons, which reduced the number of action potentials induced
by current injections (Figure 11H-I). These results suggest that activation of LP-V1 axons can
generate a net suppressive effect on L2/3 pyramidal cells through driving at least partially L1
inhibitory neurons.

Figure 11. LP axons innervate V1 L1 inhibitory neurons
35
(A) Injection of CTB in V1 to retrogradely label neurons in LP. (B) Fluorescence images showing
distribution of CTB-labeled neurons at different coronal sectional levels (relative to bregma). Scale
bar: 200 μm. (C) Images showing the expression of AAV-encoded GFP in LP (left) and GFP-
labeled LP axons in V1 (right). Scale bar: 500 (left) / 200 (right) μm. (D) Illustration of slice
recording paradigm: expressing ChR2 in LP and recording from neurons in V1 of GAD2-
Cre::Ai14 mice. Red dots represent tdTomato-labeled inhibitory neurons. (E) Average EPSC
traces recorded in an example V1 L1 neuron evoked by blue light stimulation of LP axons before
(top) and after (bottom) applying CNQX. Recording was made in the presence of TTX and 4-AP,
with the cell clamped at -70 mV. Blue arrow marks the onset of light pulse (5 ms). Scale bar: 50
pA, 50 ms. (F) Scatter plot of peak amplitudes of light-evoked monosynaptic EPSCs against the
cells’ cortical depths. Red dots represent GABAergic neurons (in L1); black dots represent
pyramidal neurons. (G) Summary of percentage of innervation of V1 neurons of different types in
different layers by LP axons. Ratios on top indicate the number of responsive cells over the total
number of recorded cells. (H) Left, superimposed traces of hyperpolarizing membrane potential
responses recorded in an example L2/3 pyramidal neuron to activation of LP-V1 axons. Blue dot
indicates the brief blue light pulse (5 ms). Resting membrane potential (-60 mV) is marked. Scale
bar: 1 mV, 100 ms. Right, average hyperpolarizing voltage (mean ± SD) evoked by a pulse of light
activation of LP-V1 axons in L2/3 pyramidal neurons (n = 7 from 3 mice). (I) Left, traces of
membrane potential responses to a square current injection without (upper) and with (lower)
stimulation of LP-V1 axons (10 pulses of blue light, 20-ms pulse duration) in the same L2/3
pyramidal neuron. Different trials are labeled by different colors. Note that action potentials were
abolished in the LED-on condition. Scale bar: 20 mV, 200 ms. Right, average number of action
potentials (per trial) evoked by the current injection without and with optic stimulation (10 pulses)
of LP-V1 axons. ***p < 0.001, paired t-test, n = 6 cells from 3 mice.

2.2.6 The LP-V1 projection can modulate V1 processing
We wondered whether the effect of LP activity on V1 observed in vivo could be mediated by the
LP to V1 projection, since LP more strongly projects to V2 than V1 (Zhou et al., 2018). To address
this issue, we injected AAV encoding Cre-dependent halorhodopsin (NpHR3.0) in LP of Vglut2-
Cre mice (Figure 12A). We applied yellow LED light (589 nm) to the surface of V1 to suppress
presynaptic release at the LP axon terminals (Mahn et al., 2016), most strongly in L1. This light
stimulation reduced visual responses of L1 neurons, as shown by the loose-patch recording from
L1 neurons (Figure 12B-C), demonstrating the effectiveness of the terminal silencing.
Consistently, we observed effects similar to silencing LP neuron cell bodies: the visual response
level was elevated and orientation/direction/size selectivity was weakened in L2/3 (Figure 12D-
36
H). Spontaneous firing rate was increased from 2.03 ± 0.96 Hz in the LED-off condition to 2.83
± 2.09 Hz in the LED-on condition, but this increase had not reached a significant level (n = 10
cells from 4 mice, p = 0.27, paired t-test). No effects were observed in L4 (Figure 12I-L).
Conversely, we activated the LP-V1 projection by expressing ChR2 in LP and shining blue LED
light on the surface of V1 (Figure 12M). In line with the slice recording data showing
monosynaptic innervation of V1 L1 neurons by LP axons (Figure 11F), blue LED stimulation
effectively evoked spiking responses of V1 L1 neurons in vivo (Figure 12N-O). In L2/3 neurons,
effects we observed were opposite to the optogenetic silencing of LP-V1 axon terminals: the visual
response level was reduced, and orientation/direction/size selectivity was enhanced (Figure 12P-
T). Spontaneous firing rate was reduced from 2.44 ± 1.53 Hz in the LED-off condition to 1.44 ±
0.64 Hz in the LED-on condition (n = 12 cells from 5 mice, p = 0.040, paired t-test). Again, no
effects were observed L4 (Figure 12U-X).
37

Figure 12. The LP to V1 projection primarily accounts for the LP modulation of V1
responses
(A) Top, experimental paradigm: injecting AAV-DIO-eNpHR3.0-EYFP into LP of Vglut2-Cre
mice and shining yellow LED light on the surface of V1. Bottom, an example image showing the
expression of eNpHR3.0-EYFP in LP. Scale bar: 400 μm. (B) Peristimulus spike time histograms
(PSTHs) of an example V1 L1 neuron to flash visual noise patterns without (left) and with (right)
inactivation of LP-V1 axons. The shaded area marks the duration yellow LED stimulation. (C)
Visually evoked firing rates of L1 neurons without (OFF) and with (ON) the optical inactivation
of LP-V1 axons. **p = 0.0039, Wilcoxon signed rank test, n = 9 cells from 2 mice. (D) Response
levels of L2/3 neurons to moving gratings without (OFF) and with (ON) the optical inactivation
of LP-V1 axon terminals. ***p < 0.001, paired t-test, n = 10 cells from 4 mice. (E) Polar plots of
orientation tuning for two example L2/3 neurons (top inset, axis limit is 10Hz) and average tuning
38
curves of the recorded L2/3 population without (black) and with (orange) the optical inactivation
of L1-V1 axon terminals. (F) The gOSI values of recorded L2/3 neurons. ***p < 0.001, paired t-
test, n = 10 cells from 4 mice. (G) The gDSI values of recorded L2/3 neurons. **p = 0.0072, paired
t-test, n = 10 cells from 4 mice. (H) The SSI values of recorded L2/3 neurons. ***p = 0.0010,
paired t-test, n = 9 cells from 4 mice. (I-L) For L4 neurons. Data are displayed similarly to (D-G).
Axis limit in (J) is 20Hz. Statistics: p = 0.79 (I), 0.90 (K) and 0.96 (L), paired t-test, n = 10 cells
for 4 mice. (M) Experimental paradigm: injecting AAV-ChR2 in LP and shining blue LED light
on V1 surface. (N) PSTH of an example V1 L1 neuron to the optical activation of LP-V1 axons
terminals. Blue bar marks the duration of LED stimulation. (O) Firing rates of V1 L1 neurons
without and with the optical activation of LP-V1 axon terminals. ***p < 0.001, paired t-test, n = 8
cells from 3 mice. (P) Visually evoked firing rates of L2/3 neurons to gratings without and with
the optical activation LP-V1 axon terminals. ***p < 0.001, paired-t test, n = 12 cells from 5 mice.
(Q) Polar plots of orientation tuning for two example L2/3 neurons and average tuning curves of
the recorded L2/3 population without (black) and with (blue) the optical activation of LP-V1 axon
terminals. Axis limit is 20Hz. (R) The gOSI values. ***p < 0.001, Wilcoxon signed rank test, n =
12 cells from 5 mice. (S) The gDSI values. **p = 0.0067, paired t-test, n =12 cells from 5 mice.
(T) The SSI values. **p = 0.0072, paired t-test, n =8 cells from 5 mice. (U –X) For L4 neurons.
Data are displayed similarly to (P-S). Axis limit in (V) is 10Hz. Statistics: p = 0.70 (U), 0.18 (W)
and 0.63 (X), paired t-test, n = 10 cells from 5 mice. Bar = SD, except for (E), (J), (Q), (V).

To examine whether scattered LED light might affect LP axons in V2, we recorded in the
posteromedial (PM) region of V2, which is the most proximate to the monocular zone of V1 we
recorded from among secondary visual areas (Wang et al., 2011). LED stimulation of LP-V1
axons terminals did not affect either spontaneous or evoked activity of superficial-layer PM
neurons (Figure 13). This indicates that the effects of the LED stimulation observed in V1 L2/3
are unlikely attributed to feedback projections from V2 to V1. Together, our results indicate that
the LP to V1 projection can primarily account for the modulatory effect of LP activity on V1 L2/3
neurons.
39

Figure 13. PM was not affected when activating LP-V1 axons
Summary of the evoked (left, N.S., not significant, P = 0.41, paired t-test, n = 15 cells) and
spontaneous (right, P = 0.92, paired t-test, n = 15 cells) firing rates of superficial layer PM neurons
without (OFF) and with (ON) optical activation of LP-V1 terminals. N = 2 animals.

2.2.7 The sSC drives LP to modulate V1 feature selectivity
Previously it has been shown that LP receives strong input from the superficial layer of SC (sSC)
(Baldwin et al., 2011, 2013; Masterson et al., 2009). In this study, by using an anterograde
transsynaptic labeling approach we recently developed (Zingg et al., 2017), we further
demonstrated that the retinorecipient SC neurons, which were located in sSC (Figure 14A),
projected to LP across the rostral-caudal axis (Figure 14B). The projection to the caudal part of
LP was apparently stronger than that to the rostral part, reminiscent of recent results (Beltramo
and Scanziani, 2019; Bennett et al., 2019). In LP regions both receiving prominent SC input and
projecting to V1 (between bregma -2.0 mm and -2.5 mm, see Figure 11B, 14B), we found that
consistent with a bottom-up SC-LP-V1 hierarchy, onsets of spiking responses of LP neurons to
flash stimuli were slower than SC but faster than V1 (Figure 15A). When the sSC was silenced
with muscimol (Figure 155B), visual responses of LP neurons were reduced by about 75%
whereas those of dLGN neurons did not exhibit significant changes (Figure 15C, left). In
comparison, LP responses were reduced by only about 15% after silencing V1 (Figure 15C, right).
40
These results demonstrate that neurons in V1-projecting LP regions can be strongly driven by
bottom-up visual input from SC.
To further test whether silencing sSC could lead to effects similar to silencing LP, we recorded
single-cell responses in V1 before and after silencing the sSC with bupivacaine (Figure 16A).
Indeed, we observed effects similar to silencing LP: orientation and direction selectivity was
weakened without changing preference (Figure 14C-G and Figure 16C-D), the visual response
level was enhanced (Figure 14H and Figure 16B), size-tuning/surround suppression was reduced
(Figure 14I-J and Figure 16E), and the dominant RF subfield was enlarged (Figure 14K-L and
Figure 16F). Similarly, these effects were observed only in L2/3, but not in L4 (Figure 14M-V
and Figure 16G-K). Together, our results indicate that the LP modulation of V1 L2/3 responses
can be driven by the bottom-up visual input from sSC.
41

Figure 14. SC input drives LP to modulate V1 responses
(A) Left, strategy for labeling retinorecipient SC neurons: first injection of AAV1-Cre in the eye
and second injection of Cre-dependent GFP virus in SC. Right, GFP-labeled retinorecipient SC
neurons were locate in the superficial layer of SC (sSC). Scale bar: 500 μm. (B) Tracing of axons
from GFP-labeled retinorecipient SC neurons at different coronal sectional levels. Scale bar: 500
μm. (C) Polar plots of orientation tuning for two example L2/3 neurons (axis limit is 20Hz) and
average tuning curve of the recorded L2/3 population (n = 24 cells from 4 mice) before (black)
and after (red) silencing sSC with bupivacaine. (D) The gOSI values of L2/3 neurons after versus
before silencing sSC. *p = 0.015, paired t-test, n = 24 cells from 4 mice. (E) The gDSI values. **p
= 0.0036, paired t-test, n = 24 cells from 4 mice. (F) Preferred orientation. N.S., p = 0.86, paired
t-test, n = 24 cells from 4 mice. (G) Changes of evoked firing rates at preferred and orthogonal
orientations. N.S., p = 0.88, paired t-test, n = 24 cells from 4 mice. (H) Visual response levels.
***p < 0.001, paired t-test, n = 26 cells from 4 mice. (I) The SSI values after versus before
silencing sSC. **p = 0.0023, paired t-test, n = 20 cells from 6 mice. (J) Population average of size
tuning curves of L2/3 neurons before (dark gray) and after (red) silencing sSC. ***p < 0.001, **p
42
< 0.01, *p < 0.05, paired t-test, n = 20 cells from 6 mice. (K) RF subfield of an example L2/3
neuron before and after silencing sSC. Scale bar: 8°. (L) RF subfield sizes after versus before
silencing sSC. **p = 0.0038, Wilcoxon signed-rank test, n = 19 cells from 4 mice. (M –V) For L4
neurons. Data are displayed similarly to (C –L). Axis limit in (M) is 20Hz. Statistics: gOSI in (N)
(p = 0.21, paired t-test, n = 16 cells from 4 mice), gDSI in (O) (p = 0.30, paired t-test, n = 16 cells
from 4 mice), preferred orientation in (P) (p = 0.68, paired t-test, n = 16 cells from 4 mice), changes
of response at preferred and orthogonal orientations in (Q) (p = 0.45, paired t-test, n = 16 cells
from 4 mice), response level in (R) (p = 0.62, Wilcoxon signed-rank test, n = 18 cells from 4 mice),
SSI in (S) (p = 0.86, paired t-test, n = 18 cells from 6 mice), and RF subfield size in (V) (p = 0.74,
paired t-test, n = 21 cells from 4 mice). Bar = SD except in (C), (J), (M), (T).

Figure 15. Effects of SC and V1 silencing on LP visual responses
(A) Onset latencies of spike responses of SC, LP and V1 neurons to flash stimuli. ***P < 0.001,
one-way ANOVA with post hoc Bonferroni's multiple comparisons test, n = 14, 12, 10 cells from
2 animals, respectively. (B) Recording LP neuron responses with a silicon probe and infusion of
muscimol into sSC. Right, image showing the spread of fluorescent muscimol. Scale bar: 400 μm.
(C) Left, changes of evoked firing rates (normalized, tested with flash stimuli) of LP (***P < 0.001,
paired t-test, n = 12 cells from 2 animals) and dLGN (P = 0.70, paired t-test, n = 12 cells from 2
animals) neurons after silencing sSC with muscimol. Right, changes of evoked firing rates of LP
neurons (normalized) after silencing V1 with muscimol. *P = 0.023, paired t-test, n = 11 cells from
2 animals.

43

Figure 16. Summary of changes in response properties in V1 before and after silencing SC
with bupivacaine
(A) Left, experimental paradigm: injecting bupivacaine into SC and recording from V1. Right,
summary of spread of bupivacaine (n = 8 animals). (B –F) Mean evoked firing rate, gOSI/OSI,
gDSI/DSI, SSI and dominant subfield size of V1 L2/3 neurons respectively before and after SC
silencing. *P < 0.05, **P < 0.01, ***P < 0.001. Bar = SEM. (G –K) Similar as (B –F) but plotted
for V1 L4 neurons. N.S., not significant.

2.2.8 LP helps to maintain V1 orientation selectivity in a noisy background
The above experiments have shown that LP activity helps enhance feature selectivity in V1. We
wondered under what circumstances this LP modulation might be particularly useful. Previous
studies have shown that visually evoked responses in V1 are reduced when visual stimuli are
presented with a noisy background (Macknik and Livingstone, 1998). Here, we noticed that LP
neurons responded more strongly to noise patterns than gratings (Figure 17A-B). In addition, they
exhibited much weaker orientation and direction selectivity than V1 neurons (Figure 17C-D),
consistent with previous reports (Durand et al., 2016; Roth et al., 2016). These observations
prompted us to speculate that LP might be more engaged when visual background is noisy. To
test this idea, animals were presented with full-screen gratings of different orientations embedded
44
in ever-changing white-noise background of different contrasts (Figure 17E, insets). We recorded
spike responses of LP neurons as well as of V1 L2/3 neurons before and after silencing LP. In the
control condition, we observed that LP responses to gratings increased (Figure 17E,G), whereas
the V1 responses decreased (Figure 17F,H), with increasing noise contrasts. Within a range of
low to moderate noise levels, V1 orientation selectivity was largely unchanged, but with the noise
level further increased it was then gradually reduced (Figure 17I, black). This indicates that high
noise background can be detrimental to orientation selectivity of L2/3 neurons. After LP was
silenced, orientation selectivity of L2/3 neurons was reduced at all noise levels (Figure 17I, red),
accompanied the changes of spontaneous firing rates, as well as evoked responses to the preferred
and orthogonal orientation (Figure 18). In particular, it was most severely compromised at a mid-
level noise contrast, so that increasing the noise level had an accelerated damaging effect on the
orientation tuning. Our results thus suggest that LP plays a role in maintaining V1 orientation
selectivity in the face of varying visual background noise.
45

Figure 17. LP helps to maintain V1 orientation selectivity in the face of increasing
background noise
(A) Schematic recording in LP with a multichannel probe. (B) Evoked firing rates to moving
gratings of optimal orientation/direction vs. to flash noise patterns (full screen). Dash line is the
unity line. ***p < 0.001, Wilcoxon signed-rank test, n = 16 cells from 2 mice. (C) Polar plot of
orientation tuning for an example LP neuron and a V1 L2/3 neuron (top inset, axis limit is 8 Hz
and 15 Hz, respectively) and average normalized tuning curves for recorded LP (light grey, n = 13
from 2 mice) and V1 L2/3 (dark, n = 16 from 2 mice) neurons. (D) Comparisons of gOSI and gDSI
between LP and V1 neurons. ***p < 0.001, Mann-Whitney rank sum test. (E) Top inset, example
flashing grating patterns (full screen) embedded in a noise background of different contrasts. Note
that the grating pattern is less visible with higher noise levels. Bottom, PSTHs of responses of a
LP neuron to the same grating pattern with increasing noise background. (F) PSTHs of responses
of a V1 L2/3 neuron to the same stationary flashing grating patterns with increasing noise
background. (G) Summary of firing rates of LP neurons to stationary flashing grating patterns with
increasing noise levels. n = 13 cells from 4 mice. Bar = SEM. (H) Summary of firing rates of V1
L2/3 neurons to stationary flashing grating patterns with increasing noise levels. n = 15 cells from
3 mice. Bar = SEM. (I) Summary of gOSI values of V1 L2/3 neurons at different noise contrasts
before (black) and after (red) silencing LP with muscimol. Star indicates a significant difference
in gOSI as compared with the one-level lower noise contrast in the same condition. ***p < 0.001,
**p < 0.01, paired t-test. The pound sign indicates a significant difference between conditions at
46
the same noise contrast.
###
p < 0.001,
##
p < 0.01,
#
p < 0.05, paired t-test, n = 13 (control) and 13
(LP silencing) cells from 4 mice, respectively. Bar = SD.

Figure 18. Changes of firing rates to flashing grating after silencing LP
(A) Summary of the changes of the evoked firing rates to the preferred orientation of the stationary
flashing grating in V1 L2/3 neurons at different noise contrasts before (black) and after (red)
silencing LP with muscimol. (B) Similar as (A) but plotted for evoked firing rates to the orthogonal
orientation of the flashing grating. (C) Similar as (A) but plotted for the spontaneous rates to the
flashing grating. Star indicates a significant difference in firing rates as compared with the one-
level lower noise contrast in the same condition. ***p < 0.001, **p < 0.01, *p < 0.05, paired t-test.
The pound sign indicates a significant difference between conditions at the same noise contrast.
###p < 0.001, ##p < 0.01, #p < 0.05, paired t-test, n = 13 (control) and 13 (LP silencing) cells from
4 mice, respectively. Bar = SD.

2.3 Discussion
In this study, we have discovered a functional contribution of LP to V1 processing: LP provides a
net suppressive signal to V1 superficial layers, which reduces the visual response level of
pyramidal neurons there and significantly enhances their feature selectivity including orientation,
direction, spatial and size tuning selectivity. This suppressive effect contributing to surround
suppression can be described as “subtractive” rather than “divisive”, as the V1 orientation tuning
curve is shifted down by increasing LP activity while shifted up by silencing LP activity. Such
modulation of orientation/direction and size tuning is consistent with the functional properties of
47
LP neurons as well as of their axons in V1: they are coarsely retinotopic, exhibiting extremely
large receptive fields (Allen et al., 2016; Bender, 1982; Berman and Wurtz, 2011; Chalupa et al.,
1983; Durand et al., 2016; Roth et al., 2016), and are largely not orientation/direction selective
(Figure 17D; Durand et al., 2016; Roth et al., 2016). The apparent surround suppression driven
by LP is likely achieved via the feedforward activation of V1 L1 inhibitory neurons by excitatory
LP axons, which then generates a thresholding-like effect (Jiang et al., 2013; Liang et al., 2018;
Priebe and Ferster, 2008) and results in an enhancement of tuning selectivity of L2/3 pyramidal
neurons, the output neurons of V1 projecting to higher cortical areas (Glickfeld and Olsen, 2017).
LP receives top-down inputs from various cortical regions including primary and secondary visual
cortices (Bennett et al., 2019; Ganz et al., 2012; Kamishina et al., 2009; Roth et al., 2016; Shipp,
2001; Tohmi et al., 2014; Ungerleider et al., 2014; Zhou et al., 2017; Zingg et al., 2017), suggesting
that top-down inputs may be able to modulate V1 processing through LP. On the other hand, LP
also receives input from SC, a route likely relaying bottom-up visual information (Baldwin et al.,
2011, 2013; Bennett et al., 2019; Gale and Murphy, 2014; Masterson et al., 2009; Roth et al., 2016;
Zingg et al., 2017), and some direct input from melanopsin-expressing retinal ganglion cells (Allen
et al., 2016). An immediate question is whether the LP modulation of V1 processing is driven by
bottom-up visual input. If so, LP can participate in the normal information processing in V1 and
be an integral part of visual processing centers. Recently, an extensive anatomical and functional
study demonstrates that SC projects preferentially to the posterior compartment of LP (pLP) and
that pLP preferentially projects to ventral-stream V2 areas, while an anterior compartment of LP
(aLP) preferentially projects to V1 and dorsal-stream V2 areas (Bennett et al., 2019). Consistently,
visual responses in the postrhinal cortex (POR), a ventral-stream V2 area, have been found to be
largely dependent on inputs from SC (Beltramo and Scanziani, 2019). In the current study, we are
48
interested in LP regions that receive input from the visual SC and also project to V1. Based on the
coordinates, it appears that our recording sites cover a central region of LP which impinges on
both the pLP and aLP described in the above study (Bennett et al., 2019). We find that the visual
responses of LP neurons in this region can be strongly driven by the bottom-up input relayed from
SC, as their responses are delayed relative to SC and are greatly reduced after silencing SC. This
indicates that the visual input from SC can drive LP to modulate V1 responses. Indeed, silencing
SC results in similar changes in visual response properties of V1 L2/3 neurons as silencing LP.
Furthermore, our experiments with optogenetic manipulations of LP-V1 axon terminals (Figure
12) suggest that the modulation can be largely attributed to the direct LP-V1 projection, although
they do not exclude the possibility that LP can also modulate V1 activity indirectly through
feedback projections from higher cortical areas, e.g. V2 (Zhou et al., 2018), as LP activity can also
drive visual responses in V2 areas (Beltramo and Scanziani, 2019; Bennett et al., 2019). Together,
we have identified a colliculo-pulvinar pathway for modulating visual response properties in V1,
which has been poorly studied before.
Since LP neurons do not exhibit specific feature selectivity (Figure 17D, Durand et al., 2016; Roth
et al., 2016), they do not directly contribute to V1 selectivity, but rather their activity sharpens V1
selectivity by generating broad and unselective net suppression in L2/3 neurons (Li et al., 2012;
Liu et al., 2011). Since LP neurons are quite responsive to common visual stimuli, their
modulation of V1 responses could be ubiquitous. In other words, LP contributes to V1 processing
and likely visual perception, and can be viewed as an important component of the central visual
system. Our behavioral experiments showing that silencing of LP impairs visual discrimination
performance support the idea that LP activity in the normal condition contributes to visual
discrimination. It is worth noting that the LP effect on V1 responses may be highly dependent on
49
anesthesia/wakefulness states, since the type of modulation we have observed here was not
reported in previous studies in anesthetized animals (Ahmadlou et al., 2018; Purushothaman et al.,
2012; Tohmi et al., 2014).
Our results in this study suggest that LP can contribute to contextual modulation of V1 processing
through its direct impact on V1 circuits. Contextual modulation, in particular when surround
suppression is concerned, is thought to be mediated by combined feedforward, intracortical
horizontal, and feedback circuit mechanisms (Angelucci et al., 2017). Subcortical neurons such
as dLGN cells already exhibit a certain level of surround suppression (Alitto and Usrey, 2008;
Bonin et al., 2005), which can be relayed to the cortex via the feedforward thalamocortical
connectivity. In layer 2/3 of V1, somatostatin (SOM) inhibitory neurons strongly integrate
horizontal inputs from L2/3 excitatory neurons driven by L4 inputs (Sugawara and Nikaido, 2014),
which allows SOM neuron responses to grow with increasing stimulus sizes and to contribute to
surround suppression of L2/3 excitatory neurons (Nienborg et al., 2013; Sugawara and Nikaido,
2014). Finally, feedback inputs from higher cortical areas also contribute to surround suppression
of V1 neurons (Hupé et al., 1998; Nassi et al., 2013; Nurminen et al., 2018), and SOM neurons
could be involved (Angelucci et al., 2017; Zhang et al., 2014). These different circuit mechanisms
may operate at different time and spatial scales (Angelucci et al., 2017). Our findings in this study
have thus revealed a novel source for driving surround suppression in V1 L2/3, LP/pulvinar, which
is independent of the canonical retinogeniculate pathway. Since this extrageniculate pathway is
mainly driven by feedforward input from SC, it may contribute to surround suppression in V1 L2/3
at a faster time scale than feedback projections.
50
Compared to patterned visual stimuli, LP neurons seem to be more sensitive to non-patterned
visual noise and their firing rates are monotonically modulated by the noise level. This property
would lead to noise-dependent LP-driven suppression in V1 L2/3, which helps to maintain V1
orientation selectivity within a range of low to moderate background noise levels. When LP is
silenced, increasing the noise level much more quickly impairs V1 orientation selectivity. In other
words, via the LP-mediated feed-forward contextual modulation, the detrimental effect of
background noise on V1 selectivity can be “cancelled” to a certain degree. Such “noise-cancelling”
effect may be important for animals to quickly detect predators in visually noisy environments.
Together, our results suggest that a bottom-up retina-SC-LP-V1(L1) visual pathway, running in
parallel with the canonical retina-geniculate-V1 ascending visual pathway, forms a “differential”
circuit which provides a suppressive but sharpening effect in V1 superficial layers (Figure 19).
This may represent an important neural circuit strategy for enhancing sensory information
processing and achieving contextual modulations in general.

Figure 19. A working model for LP modulation of visual processing in V1
51
The SC-LP-V1 pathway can drive “feedforward” suppression in L2/3 pyramidal neurons via L1
inhibitory neurons and together with the canonical retino-geniculate-V1 pathway forms a
differential circuit to “cancel” noise effects. Gray triangles represent excitatory neurons, gray
circles inhibitory neurons. Arrows represent excitatory projections. Red bar represents a
suppressive effect.

2.4 Material and methods
2.4.1 Animal
The surgeries and experiments were performed in the Zilkha Neurogenetic Institute (ZNI) at the
University of Southern California (USC). Animal Care and Use Committee of USC approved all
procedures used in this study. Male and female wild-type C57BL/6J and transgenic Vglut2-
Cre::Ai14 mice (The Jackson Laboratory) mice of 2-4 months old were used in this study. The
animals were housed in the ZNI vivarium with 12h light/dark cycles.
2.4.2 Head fixation and surgery
Three days before electrophysiological recordings, a screw was glued to the skull surface of the
mouse with acrylic dental cement under the anesthesia with 1.5% isoflurane (v/v). The screw was
clamped tightly with a metal head post on the recording setup to achieve head fixation. After
recovery from anesthesia, the mouse was trained to get accustomed to the head fixation and to run
freely on a plastic rotatable plate (Liang et al., 2015). On the day of recording, the mouse was
again anesthetized with isoflurane. A craniotomy window of 0.2 mm × 0.2 mm was made over V1
(ML +2.7 mm, AP -3.45 mm), PM (ML +1.5 mm, AP -4.00 mm), LP (AP -2.2 to -2.3 mm, ML
+1.4 mm, DV -2.35 mm), dLGN (AP -2.15 mm, ML +2.5 mm, DV -2.8 mm), or SC (AP -3.75
mm, ML +0.6 mm, DV -1.35 mm). A durotomy was further performed to allow the insertion of a
glass pipette or silicon probe, or the implantation of a drug cannula or optic fiber. The recorded
V1 was on the same side of LP or SC being manipulated. After the surgery, the exposed cortex
52
was covered with a silicon elastomer (Kwik-CAST, WPI). The mouse was fully recovered from
anesthesia before recording sessions.
2.4.3 Behavioral task
To assess the functionality of visual perception, mice were trained to perform a head-fixed Go/No-
Go visual discrimination task (Lee et al., 2012). The visual cues are two full screen static sinusoidal
grating patterns (0° and 90° in orientation, spatial frequency = 0.04 cpd, full contrast). The inter-
stimulus-interval was randomly chosen between 10-12 s. The cues were presented for 0.5 s
according to a pseudorandom sequence, followed by a delay period of 0.5 s. The reward (30 nl
water) was given after the delay period when the 90° (vertical) grating (Go signals) was presented,
and no reward was delivered when the 0° (horizontal) grating (No-Go signals) was present. After
initial two days of water deprivation, mice were trained to lick to the Go cues but not to lick upon
the No-Go cues. A trial was terminated 1 s after the water reward. No punishment was given when
the animals licked to the No-Go cues. The licking behavior was videotaped by a camera placed
closed to the mouth of the animal for online monitoring and offline analysis. The hit and correct
rejection trials were pooled to determine the correct rate. The animals were considered to have
learned the task when the correct rate had been above 75% for two concessive days. One day
before the last training day, 100 nl saline was infused into LP, and on the last day 100 nl of
muscimol was bilaterally infused into LP. In another cohort of mice, AAV encoding inhibitory
DREADD receptors (hM4Di) was bilaterally injected into LP two weeks before the beginning of
training sessions. On the last training day, the DREADDi agonist, CNO, was administered (i.p.,
1mg/kg) 40 min before the behavior test to chemogenetically silence LP. The correct rates one day
before (saline) and on the day of LP silencing (either through muscimol injection or chemogenetic
53
inactivation) were compared. For control, we injected CNO in well-trained GFP-expressing
animals and compared their performance before and after the CNO injection.
2.4.4 In vivo electrophysiological recording
All recordings were performed in a dark room with weak background illumination. A single
recording session was limited to at most 4hr. Between sessions, the mouse was allowed to rest and
receive drops of 5% sucrose. The silicon seal was removed before recordings. Loose-patch
recordings were performed following our previous studies (Ibrahim et al., 2016; Liang et al., 2015).
A patch pipette (5-7 MΩ) was filled with artificial cerebral spinal fluid (ACSF; 126 mM NaCl, 2.5
mM KCl, 1.25 mM Na2PO4, 26 mM NaHCO3, 1 mM MgCl2, 2 mM CaCl2, and 10 mM glucose).
A seal of 0.1-0.5 GΩ was achieved on the soma. Spikes from the patched neuron were recorded
with an Axopatch 200B amplifier (Molecular Devices) under voltage-clamp mode, with clamping
voltage adjusted to achieve a zero baseline. The electric signals were sampled at 20 kHz sampling
rate and passed through a 300-3000 Hz band-pass filter. The relatively large opening of the pipettes
(2 μm) highly biased the recordings in favor of excitatory pyramidal neurons (Liang et al., 2015;
Liu et al., 2010). The location of V1 was verified from the correct pattern of a retinotopic map in
the exposed region, and all the recordings were performed in the monocular region of the V1
(Drager, 1975). The cortex was pre-penetrated before recording to minimize the tissue dimpling.
The cortical layers were assigned according to the travel distance of the pipette orthogonal to the
pia surface of V1, which was found to match well with the actual cortical depth determined form
post hoc histological results in experiments where the pipette was coated with DiI (Figure 20A).
The recorded L2/3 neurons had a range of depths, 170-350 μm from the pia, and L4 neurons, 375-
500 μm from the pia (Figure 20B). This layer assignment was based on our previous study that
Scnn1a expressing neurons were at 374-510 μm from the pia (Li et al., 2012). To record from
54
neurons in LP, dLGN or PM, a 64-channel silicone probe (NeuroNexus) was used. Signals were
recorded by an Open-Ephys system (Open Ephys) at 30 kHz sampling rate. Raw unfiltered traces
were saved for offline spike sorting and analysis. Typically, each animal went through 1-2
recording sessions. Recordings were performed during quiet wakefulness, and pharmacological or
optogenetic manipulations of brain activity did not change the general locomotion status of the
animals. We also monitored the pupil position during the recording sessions and found that optical
stimulation did not trigger significant eye movements (Figure 10K).

Figure 20. Cortical depth of recorded neurons
(A) Scatter plot of pipette travel distance (insertion angle corrected) and cortical depth determined
from histology. P = 0.86, paired t-test, n = 12 cells. (B) Distribution of recording depths for L2/3
(n = 207) and L4 (n = 142) neurons recorded in this study.

2.4.5 Viral injection
The injection of virus was performed on a stereotaxic apparatus as previously reported (Chou et
al., 2018; Liang et al., 2015; Xiong et al., 2015). The mouse was anesthetized with 1.5% (v/v)
isoflurane. A 0.2 mm x 0.2 mm craniotomy was performed to expose the cortex above LP (AP -
1.5 mm, ML +1.4 mm, DV -2.35 mm) or SC (AP -3.75 mm, ML +0.6 mm, DV -1.35 mm). The
dura mater was then removed. The virus was delivered via a beveled glass micropipette with a tip
opening 20-30 μm in diameter, which was attached a microsyringe pump system. The following
55
adeno-associated viruses (AAVs) encoding ChR2, ArchT, eNpHR3.0 and h4MDi were used to
optogenetically or chemogenetically manipulate neuronal activity: AAV1-CaMKIIa-hChR2-
EYFP (1.7×10
13
GC/ml, UPenn vector core, Addgene 26969), AAV9-EF1a-DIO-eNpHR3.0-
EYFP (1.7×10
13
GC/ml, UPenn vector core, Addgene 26966) and AAV1-CAG-ArchT-GFP
(1.7×10
13
GC/ml, UNC vector core, Addgene 29777), and AAV5-hSyn-hM4Di(Gi)-mCherry
(3×10
12
GC/ml, UPenn vector core, Addgene 50475). 50 nl of the viral solution was injected into
LP at a rate of 15 nl/min. AAV9-EF1a-DIO-eNpHR3.0-EYFP was injected in the Vglut2-ires-Cre
knock-in mice (The Jackson Laboratory) to specifically silence the axonal terminals of excitatory
LP projection neurons in V1. The pipette then stayed in the LP for 5 min before withdrawal to
prevent leakage. For anterograde tracing of neural projections and control experiments for chemo-
or optogenetics, a volume of 50 nl AAV1-CB7-CI-eGFP-WPRE-rBG solution (1.7×10
13
GC/ml,
UPenn vector core, Addgene 105542) was injected to SC and LP. To anterogradely label the axonal
outputs of SC neurons that receive direct retina inputs, we adopted a two-step viral tracing
approach (Zingg et al., 2017). 50 nl of AAV2/1-hSyn-Cre-WPRE-hGH (UPenn Vector Core,
2.510
13
GC/mL, Addgene 105553) was first injected into the eye of Ai14 mice. After 3 days, a
second load of AAV2/1-CAG-FLEX-eGFP-WPRE-bGH (UPenn vector core, 1.7 10
13
GC/mL,
Addgene 51502) was injected into the contralateral SC (50 nl). The spacing between two injections
allowed sufficient time for the clearance of any remaining AAV-Cre virus that may have spread
across the pial surface. For retrograde labelling of V1-projecting LP neurons, 50 nl of cholera toxin
subunit B, Alexa 488 (CTB-488, 0.5% solution in PBS, ThermoFisher) was injected into V1. The
scalp was re-sutured, and the mouse was administrated with 0.1 mg/kg buprenorphine
subcutaneously and sent back to home cages. The recordings and imaging were performed 3 weeks
after viral injection to allow the expression of the virus and the recovery of the mouse from surgery.
56
2.4.6 Histology and imaging
After experiments, the mouse was deeply anesthetized and transcardially perfused with 4%
paraformaldehyde (PFA) in phosphate-buffered saline (PBS). The brain tissue was then harvested
and sliced into 150 μm coronal sections using a vibratome (Leica, VT1000s). Fluorescence images
were acquired using a confocal microscope (Olympus FluoView FV1000) to verify the location of
virus expression and drug injection. Images of both the injection site and projection from SC to
LP or LP to V1 were collected and imaged under a 4  objective. Regions with axonal labeling
were further imaged under a 10  objective to get clearer projection patterns.
2.4.7 Visual and LED stimulation
Visual stimuli were generated in Matlab (MATLAB) with the Psychophysics Toolbox Version 2
(Brainard, 1997) and were presented on a ViewSonic VA705b monitor (1920 × 1440 pixels, 33.9
cm wide, 27.2 cm high, 60 Hz refresh rate, mean luminance 41 cd/m
2
) mounted on a flexible arm.
The monitor was adjusted to be 20 cm away from the eye, at 45° azimuth, 25° elevation, and thus
subtending 80° azimuth × 70° elevation of mouse’s visual field, corresponding to the monocular
region of V1. The monitor was gamma-corrected to achieve linear luminance. To pre-map V1 or
measure the receptive field (RF) center of a patched neuron, static bright (58 cd/m
2
) and dark (24
cd/m
2
) squares (each square 10° × 10°) of an 8 × 6 grid covering the entire screen were presented
individually in a pseudorandom sequence on a gray background of mean luminance. Each square
was displayed for 200 ms with 5-10 repeats, and the inter-stimulus interval was 440 ms. To
measure the overall visual response level, a set of dense white-noise stimuli were presented. Each
frame of the stimuli consisted of a grid of 20 × 20 squares (each square 4° × 3°) intensities of
which were determined by an m-sequence. Each pattern was presented for 200 ms and 20-50
patterns were presented according to response level and fidelity. To measure orientation and
57
direction tuning, a set of drifting sinusoidal gratings were presented in pseudorandom order. The
stimulus set consisted of gratings of 12 directions (0°-330°, 30° per step, 6 orientations). To drive
as many V1 neurons as possible, the spatial frequency of the gratings was chosen to be 0.04 cycles
per second (cpd) and temporal frequency to be 2 cycles per second (Hz) based on previous research
(Niell and Stryker, 2008). Each grating drifted for 1.5 s and another grating appeared, which
remained to be static for 3 s before moving. To measure size tuning, drifting gratings of the optimal
orientation but of varying sizes were presented at RF center. The size (diameter) of each grating
stimulus was chosen from an array of values as 6°, 9°, 14°, 20°, 30°, 46°, 68° and 102° according
to a pseudorandom sequence. To prevent from direction adaptation, gratings with preferred and
null directions were interleaved. To precisely map spatial RFs, sparse noise stimuli, composed of
static bright and dark squares (each square 3° × 3°) was used. The squares were presented
individually on a gray screen within a 16 × 16 grid centered at RF center according to an m-
sequence. Each square was displayed for 32 ms (i.e., updated every other frame), with zero inter-
stimulus interval between 2 consecutive squares. To test V1 processing with a visually noisy
background, we generated a series of visual noise patterns. As control, static sinusoidal grating
patterns of 4 phases (0, π/2, π, 3π /2) and 6 orientations ranging from 0° to 150° (step = 30°) with
50% contrast (pixel values ranging evenly from 0.25 to 0.75, with 0 means black and 1 white,
respectively) was generated. A series of noise squares with increasing contrast (17%, 33%, 50%
and 75%, pixel values evenly distributed in ranges = [-0.08, +0.08], [-0.17, +0.17], [-0.25, +0.25],
[-0.375, 0.375]) were added onto the grating patterns, so that the grating patterns were increasingly
blurred and invisible among the noise. The pixel values of the merged images were cutoff at [0.25,
0.75] to ensure the mean luminance and the range of luminance of the pixels were the same as the
original grating pattern. Each pixel of the noise patterns was 4° in visual degree, and each pattern
58
was presented 200 ms for 10-20 repetitions in a pseudorandom order. For optogentic manipulations,
two sequences were generated independently for drifting grating stimuli, with trials with and
without LED light stimulation interleaved. The inter-stimulus interval between consecutive LED-
ON and LED-OFF trials was 10 s to allow light-gated channels fully recover from desensitization.
2.4.8 In vivo optogenetic manipulation
To photo-manipulate LP neurons, an optic fiber (200 μm, Thorlabs) was inserted through the
cortex close to the surface of LP (ipsilateral to the V1 recorded) at the depth of 2.1 mm. The fiber
was then secured with acrylic dental cement. Right before recording sessions, the fiber was
connected to a LED source (Thorlabs) to deliver blue (473 nm, 10 mW), green (530 nm, 10 mW)
or yellow light (589 nm, 10 mW). No labeled neural structures other than LP were observed within
800 μm depth from the tip of the fiber. In the LED-ON sessions, light stimulation preceded the
onset of visual stimulation by 500ms, covered the entire duration of visual stimulation and
terminated at 300ms after the offset of visual stimuli. Blue light pulses (20 ms in duration, 20 Hz)
and continuous green light were applied to examine the effect of activating (via ChR2) and
silencing (via ArchT) LP on cortical visual processing, respectively. For terminal manipulation,
optic fiber was implanted over V1. Blue light pulses and continuous yellow light were given to
activate (via ChR2) and silence (via eNpHR3.0) LP axonal terminals in V1, respectively.
2.4.9 Pharmacological silencing of brain regions
Bupivacaine (a voltage-gated sodium channel blocker, 0.5%, dissolved in ACSF containing Alexa
conjugated dextran) was used to silence targeted brain regions. To effectively silence LP
(ipsilateral to the V1 recorded), a cannula was implanted 2.5 mm below the pia and 100 nl
bupivacaine was slowly infused. For silencing superficial SC (ipsilateral to the V1 recorded), a
59
cannula was implanted at the depth of 0.75 mm and a volume of 100 nl the drug was slowly
administrated. The silencing effect was verified by comparing spontaneous and evoked spiking
responses before and after drug infusion. We found that spiking activity was silenced at 5 min
after the beginning of the infusion for at least 30 min and was recovered after 1.2 hr (data not
shown). This method allowed us not only to temporally and reversibly silence LP when recording
from one neuron, but also to record from multiple neurons during the entire recording session. In
another set of multi-channel recording experiments, we infused 100 nl of fluorescence conjugated
muscimol (1.5 mM, Life Technologies), an agonist of GABAA receptors, into sSC and V1 for
silencing while recording from LP or dLGN. In behavioral tests, muscimol was bilaterally infused
into LP for silencing for a longer period of time (> 2hr).
2.4.10 Slice preparation and recording
To confirm the functional connectivity from LP to V1, GAD2-Cre, PV-Cre or SOM-Cre crossed
with Ai14 mice and injected with AAV1-CaMKIIa-hChR2-EYFP in LP were used for slice
recording. Three weeks following the injections, animals were decapitated following urethane
anesthesia and the brain was rapidly removed and immersed in an ice-cold dissection buffer
(composition: 60 mM NaCl, 3 mM KCl, 1.25 mM NaH2PO4, 25 mM NaHCO3, 115 mM sucrose,
10 mM glucose, 7 mM MgCl2, 0.5 mM CaCl2; saturated with 95% O2 and 5% CO2; pH = 7.4).
Coronal slices at 350 µm thickness were sectioned by a vibrating microtome (Leica VT1000s) and
recovered for 30 min in a submersion chamber filled with warmed (35 °C) ACSF (composition:119
mM NaCl, 26.2 mM NaHCO3, 11 mM glucose, 2.5 mM KCl, 2 mM CaCl2, 2 mM MgCl2, and 1.2
mM NaH2PO4, 2 mM Sodium Pyruvate, 0.5 mM VC). Cuts were made in the slices along the
boundaries of secondary visual areas flanking V1 (identified based on fluorescence pattern of LP
axons) before recording to prevent potential feedback inputs to V1. EYFP
+
fibers, L1 inhibitory
60
neurons or PV/SOM inhibitory neurons with tdTomato expression were visualized under a
fluorescence microscope (Olympus BX51 WI). Patch pipettes (~ 4-5 MΩ resistance) filled with a
cesium-based internal solution (composition: 125 mM cesium gluconate, 5 mM TEA-Cl, 2 mM
NaCl, 2 mM CsCl, 10 mM HEPES, 10 mM EGTA, 4 mM ATP, 0.3 mM GTP, and 10 mM
phosphocreatine; pH = 7.25) were used for whole-cell voltage-clamp recordings. Synaptic currents
were recorded with an Axopatch 200B amplifier (Molecular Devices) under voltage clamp mode
at a holding voltage of –70 mV for excitatory currents or 0 mV for inhibitory currents, filtered at
2 kHz and sampled at 10 kHz. Tetrodotoxin (TTX, 1 μM) and 4-aminopyridine (4-AP, 1 mM)
were added to the external solution for recording monosynaptic responses only to blue light
stimulation (10 ms pulse, 3 mW power, 10-30 trials, delivered via a mercury Arc lamp gated with
an electronic shutter). A neuron was considered innervated if the peak amplitude of the
monosynaptic response was larger than 3SD above the baseline fluctuation. To test the excitatory
nature of the LP to V1 input, a glutamate receptor antagonist, 6-cyano-7-nitroquinoxaline-2,3-
dione (CNQX, 20 μM) was further added to the bath solution. For current clamp recordings, the
potassium-based internal solution was used (130 mM K-gluconate, 2 mM KCl, 1 mM CaCl2, 4
mM MgATP, 0.3 mM GTP, 8 mM phosphocreatine, 10 mM HEPES, 11 mM EGTA, pH = 7.25).
Membrane potentials were recorded under current-clamp mode. To examine the effect of LP-V1
terminal activation on the postsynaptic membrane potential of V1 L2/3 neurons, a brief blue light
pulse (10 ms pulse duration, 3 mW power, 10-30 trials) was applied. A positive current (100 pA,
500 ms) was injected into the cell to induce spiking. A train of blue light pulses (10 pulses at 20
Hz, 25 ms pulse duration, 10-30 trials) was applied concurrently with the current injection. Trials
with and without blue light pulses were interleaved. The postsynaptic membrane potentials and
firing rates without and with LP-V1 terminal activation were compared.
61
2.4.11 Data analysis
Data analysis was performed using customized scripts written in Matlab (MATLAB) and Labview
(National Instruments). Raw waveforms were saved online, and spikes were detected offline using
a thresholding algorithm.
Response level. To quantify the overall response level of V1 neurons, the number of spikes evoked
by dense white noise flashes was counted within the 70-270 ms window after the onset of the
visual stimulus. The spike number was averaged across trials to derive firing rate. The spontaneous
firing rate was calculated from the time window 200 ms before the stimulus onset when a static
grating pattern was present on the screen, and then subtracted from the evoked firing rate. The
response levels before and after drug application or between LED-ON and LED-OFF trials were
compared.
Orientation and direction tuning. To derive tuning curves, the spike number evoked by drifting
gratings was counted within the 70-1570 ms window after stimulus onset. The responses were
organized according to the direction of moving stimuli and averaged across repetitions.
Spontaneous firing rate calculated within a 500 ms window before the stimulus onset was
subtracted. Orientation tuning curves were obtained by averaging responses to two opposite
directions (i.e. 0° and 180°, 30° and 210°, etc). To further quantify tuning strength, a global
orientation (gOSI) and direction selectivity index (gDSI) were determined by dividing vector
summation with the sum of scalar values as follows:
𝑔 𝑂 𝑆𝐼 =
Σ ∥ 𝑅 (𝜃 )𝑒 2𝑖𝜃
∥
Σ𝑅 (𝜃 )

62
𝑔𝐷𝑆𝐼 =
Σ ∥ 𝑅 (𝜃 )𝑒 𝑖𝜃
∥
Σ𝑅 (𝜃 )

θ is the direction of moving grating. R( θ) is the response to direction θ. 𝑖 = √−1. Both gOSI and
gDSI ranges from 0 to 1. gOSI = 0 (or gDSI = 0) means that the neuron responds equally to all
orientations (or directions). gOSI = 1 (or gDSI = 1) implies the neuron responds exactly to just one
orientation (or direction). gOSI and gDSI values before and after drug application or between
LED-ON and LED-OFF trials were compared. The conventional OSI was calculated as:
𝑂𝑆𝐼 =
𝑅 𝑝𝑟𝑒𝑓 − 𝑅 𝑜𝑟𝑡 ℎ
𝑅 𝑝𝑟𝑒𝑓 + 𝑅 𝑜𝑟𝑡 ℎ
,
where 𝑅 𝑝𝑟𝑒𝑓 and 𝑅 𝑜𝑟𝑡 ℎ
are the responses to the preferred and the orthogonal orientation (90° from
the preferred orientation), respectively. Similarly, the conventional DSI was calculated as:
𝐷𝑆𝐼 =
𝑅 𝑝𝑟𝑒𝑓 − 𝑅 𝑛𝑢𝑙𝑙 𝑅 𝑝𝑟𝑒𝑓 + 𝑅 𝑛𝑢𝑙𝑙 ,
where 𝑅 𝑝𝑟𝑒𝑓 and 𝑅 𝑛𝑢𝑙𝑙 are the responses to the preferred and the null direction (180° from the
preferred direction), respectively.
Size tuning. To characterize the neuronal response to gratings of increasingly larger sizes,
responses to gratings of optimal direction within a window of 70-1570 ms after the stimulus onset
were organized by grating size and averaged across trials. A neuron was defined as size-tuned if
the largest stimuli did not elicit the most robust response. The tuning strength was further
quantified by surround suppression index (SSI), which determined as (Sugawara and Nikaido,
2014):
63
𝑆𝑆𝐼 =
𝑅 𝑝𝑟𝑒𝑓 − 𝑅 𝑙𝑎𝑟𝑔𝑒 𝑅 𝑝𝑟𝑒𝑓 ,
where 𝑅 𝑝𝑟𝑒𝑓 is the response to the grating of the optimal size, 𝑅 𝑙𝑎𝑟𝑔𝑒 is the response to the largest
grating. SSI = 1 suggests zero 𝑅 𝑙𝑎𝑟𝑔𝑒 (i.e., complete surround suppression), while SSI = 0 indicates
no surround suppression at all. SSIs before and after drug application or between LED-ON and
LED-OFF trials were compared.
Spatiotemporal receptive field (STRF). To map the fine structure of RF, the spike train to the
sparse noise stimuli was reversely correlated with the stimulus sequence to derive the spike-
triggered average response of the stimuli (Jones and Palmer, 1987). A 2D standard deviation was
calculated at each time lag from 0-200 ms as the strength of RFs. The RF showing strongest
response was used to determine the size of RF. The background response level was determined by
averaging the spike numbers at the 4 borders of RF and subtracted from RF. The RF map was
further divided by its 2D standard deviation to obtain Z scores. Pixel value with Z score < 4 was
set to be 0. The area of connected pixels was calculated and transformed to a circle with the same
area. The radius of the circle was defined as the RF size. To better visualize the boundary, the
thresholded RF map was fitted with a 2D elliptical Gaussian function Φ(𝑋 , 𝑌 ). 2𝜎 𝑋 and 2𝜎 𝑌 were
used to outline the boundary of RF.
Spike sorting. Spike sorting was performed following our previous study (Zhang et al., 2018a).
The raw signals from the 64-channel silicon probe were filtered through a 300-6000 Hz band-pass
filter. The spatially varying motion artifacts were removed by applying a local common average
referencing (L-CAR) scheme. The nearby four channels of the silicon probe were grouped as
tetrodes, and semi-automatic spike detection and sorting was performed using the Plexon offline
64
sorter (Dallas, Texas). Clusters with isolation distance > 20 was considered as separate clusters.
Spike clusters would be classified as single units only if the waveform SNR (Signal Noise Ratio)
exceeded 4 (12 dB) and the inter-spike interval was longer than 1.2 ms for more than 99.5% of the
spikes.
2.4.12 Statistics
The Shapiro-Wilk test was performed to test the normality of the data set. If the data were normally
distributed, parametric paired t-test or two-sample t-test was used. Otherwise, the nonparametric
Wilcoxon signed rank test or Mann-Whitney rank sum test was used. Two-sample Kolmogorov-
Smirnov test was used to determine whether the data from two groups were from the same
distribution. Bootstrapping approach was used to estimate the standard error of the mean (SEM)
of gOSI and gDSI, and permutation test is used to determine whether gOSI or gDSI values between
two conditions are significantly different. The significance level of the tests was set as 0.05. Unless
mentioned in the text, the data were reported as mean ± SEM.

65
Chapter 3: Contextual and Cross-Modality Modulation of
Auditory Cortical Processing through Pulvinar

3.1 Introduction
Thalamus is generally considered as a gate to the cerebral cortex. First-order thalamic nuclei, such
as the dorsal lateral geniculate nucleus (dLGN) and ventral medial geniculate body (MGBv), relay
bottom-up sensory information to primary sensory cortices, visual and auditory, respectively
(Kremkow and Alonso, 2018; Winer et al., 2005). They serve as the major driver of sensory
responses in the cortex for each respective sensory modality (Guillery and Sherman, 2002; Halassa
and Sherman, 2019). Compared to first-order thalamic nuclei, the functional roles of higher-order
thalamic nuclei in sensory processing are much less well understood, although many of them are
known to have broad connections with both primary and secondary cortices.
The lateral posterior nucleus (LP) of the thalamus is the rodent homologue of the primate pulvinar
nucleus (Harting et al., 1972, 1973). The latter is considered the largest thalamic complex (Harting
et al., 1973). Previous studies on LP/pulvinar have mostly been focused on its involvement in
visual-related functions, such as visual attention, visually guided behaviors and eye movements
(Dominguez-Vargas et al., 2017; Saalmann et al., 2012; Soares et al., 2017; Stitt et al., 2018; Zhou
et al., 2016), largely due to its extensive reciprocal connections with visual cortical areas (Beltramo
and Scanziani, 2019; Bennett et al., 2019; Juavinett et al., 2020; Kaas and Lyon, 2007; Nakamura
et al., 2015; Oh et al., 2014; Roth et al., 2016; Shipp, 2001; Stitt et al., 2018; Wong et al., 2009;
Zhou et al., 2018). Besides visual cortices, LP/pulvinar also has connections with other sensory
66
cortices including primary and secondary auditory cortices (Cappe et al., 2009; Hackett et al., 1998;
Nakamura et al., 2015; Oh et al., 2014; Shipp, 2007) and contains auditory responsive neurons
which exhibit short-latency responses (Chalupa and Fish, 1978; Gattass et al., 1978; Magariños-
Ascone et al., 1988; Woody et al., 1991; Yirmiya and Hocherman, 1987). However, the influence
of LP on auditory cortical responses has rarely been examined. The impact of LP activity on the
primary auditory cortex (A1) is of particular interest. It is known that the axonal projections from
LP to primary sensory cortices mainly terminates in layer 1 (L1), whereas those to secondary
sensory cortices mainly terminate in L4 (Fang et al., 2020; Roth et al., 2016; Zhou et al., 2018).
As L1 contains predominantly inhibitory neurons (Jiang et al., 2013; Mesik et al., 2019; Schuman
et al., 2019), it is likely that the LP input to the primary sensory cortex plays some modulatory
roles.
In the present study, we investigated the effect of LP activity on auditory responses of neurons in
superficial layers of A1 using bidirectional optogenetic manipulation approaches. We found that
LP activity improved auditory processing functions in A1 in that it helped to sharpen frequency
tuning of A1 L2/3 pyramidal neurons and to enhance the signal-to-noise ratio (SNR) of their
auditory responses, through subtractive suppression of their responses in analogous to a
thresholding effect. In addition, we found that such effect could play a role in modulating A1
responses under noise background. Furthermore, we found that LP, by receiving salient input from
SC, could also mediate the cross-modality modulation of A1 responses by visual looming stimuli.

67
3.2 Results
3.2.1 Bidirectional modulation of frequency tuning and SNR in A1 L2/3 by LP
To study whether LP could influence auditory cortical processing, we applied optogenetic
approaches to manipulate LP activity and examined changes in functional response properties of
A1 neurons using single-cell loose-patch recordings in awake mice (see 3.4). For reversible and
temporal silencing of neuronal spikes, we injected adeno-associated virus (AAV) encoding ArchT
into LP and implanted an optic fiber over LP to deliver green LED light (Figure 21A). Recordings
were made in the ipsilateral A1. To investigate auditory information processing functions, we
presented a set of tone pips of varying frequencies and intensities (50-ms duration, 2 – 45 kHz, 10
– 70 dB sound pressure level or SPL) and recorded tone-evoked spike responses from L2/3
pyramidal neurons (see 3.4). A typical L2/3 pyramidal neuron exhibited a V-shaped tonal
receptive field (TRF) with a distinct characteristic frequency (CF) (Figure 21B, left). We
interleaved LED-on and LED-off trials so that TRFs without and with LP silencing could be
compared in the same neuron. When LP was silenced, we observed an overall increase in response
level and TRF size as compared to the LED-off condition (Figure 21B, right; Figure 22A-B).
Nevertheless, the overall shape of frequency tuning (Figure 21C) and the CF of TRF (Figure 23A)
were largely preserved. Analyzing all the recorded L2/3 neurons, we found significant increases
in the spontaneous firing rate (FR), evoked FR and bandwidth of frequency tuning (as measured
at an intensity level of 60 dB SPL and at 20 dB above the intensity threshold) when silencing LP
(Figure 21D-F; Figure 24A). More importantly, the signal-to-noise ratio (SNR), as measured by
the ratio of evoked to spontaneous FR, was reduced (Figure 21G). These results indicated that
when LP activity was suppressed auditory information processing in A1 might be compromised
due to broadening of frequency tuning and a reduction in SNR.
68
Next, we optogenetically activated LP by injecting AAV encoding channelrhodopsin2 (ChR2) into
LP and delivering blue LED light pulses (Figure 21H). As shown by an example L2/3 pyramidal
neuron, activation of LP decreased the amplitude of tone-evoked responses and shrank the TRF
(Figure 21I-J; Figure 22C-D), without changing the CF (Figure 23B). Opposite to the effects of
silencing LP, activation of LP reduced the spontaneous and evoked FR, and narrowed the
frequency tuning bandwidth, while enhancing the SNR (Figure 21K-N; Figure 24B). Therefore,
the frequency tuning and SNR of A1 L2/3 pyramidal neurons could be bidirectionally modulated
by manipulating LP activity: increasing LP activity enhances frequency selectivity and SNR of
auditory responses and thus generally improves auditory cortical processing, and vice versa for
decreasing LP activity. As a control, we performed similar experiments in GFP-expressing
animals and did not observe any significant changes in auditory response level by either green or
blue LED light delivery (Figure 24B, 21D). In L4, we did not observe any changes of either
spontaneous or evoked FR induced by optogenetic manipulations of LP (Figure 25). Therefore,
the LP’s modulatory effect may be specific to superficial layers.
3.2.2 LP exerts a thresholding effect on A1 L2/3 responses
Comparing frequency tuning curves without and with LP manipulation (Figure 21C, 21J), it
appears that LP activity just shifted the A1 frequency tuning curve up or down, without changing
the tuning preference (i.e. the best frequency). To further elucidate the nature of LP modulation,
we plotted firing rates evoked by effective tones (see 3.4) with versus without LP manipulation
and then performed a linear regression analysis. As shown by the two example A1 neurons (the
same as shown in Figure 21B, 21I), data points were distributed along a line which had a positive
y-intercept for LP silencing (Figure 21O) but a negative y-intercept for LP activation (Figure
21P), supporting the notion that all the tone responses were elevated or reduced by a certain
69
amount when LP activity was manipulated. We did a similar analysis for all the recorded neurons
and found that the linearity was generally high (R
2
close to 1) (Figure 21Q). The slope of the best
fit line was close to 1 for both LP silencing and activation (Figure 21R), suggesting that there was
no change in the response gain (i.e., no evidence for a multiplicative effect). The y-intercept was
all positive for LP silencing but negative for LP activation (Figure 21S). These results strongly
suggest a subtractive-suppression-like modulation of A1 responses by LP activity, similar to a
thresholding effect produced by increasing the level of background noise (Liang et al., 2014).
Therefore, LP can bidirectionally modulate auditory responses in superficial layers of A1 through
a thresholding mechanism, i.e., by shifting A1 frequency tuning up or down.
70

Figure 21. Effects of bidirectional manipulation of LP activity on A1 response properties
(A) Left, schematic experimental condition. AAV-ArchT was injected into LP and green LED was
applied via an implanted optic fiber to silence LP neurons. Loose-patch recordings were made
from A1 neurons on the same side. Right, confocal image showing the expression of ArchT within
LP (marked by white dashed curve). Scale bar, 500 m. (B) Reconstructed tonal receptive field
(TRF) of an example A1 L2/3 pyramidal neuron without (left) and with (right) optogenetic
silencing of LP. Color scale indicates evoked firing rate. Red curve marks the TRF boundary. Red
double arrows depict the frequency tuning bandwidth at 60 dB SPL. Red arrowhead marks the
characteristic frequency (CF). (C) Normalized frequency tuning curves (at 60 dB SPL) without
(black) and with (green) LP silencing. Arrows mark the best frequency. (D –G) Normalized
spontaneous firing rate (D, p = 0.0078, Wilcoxon signed-rank test), evoked firing rate (E, p =
0.0079, paired t-test), tuning bandwidth at 60 dB SPL (F, p = 0.0069, paired t-test) and signal-to-
71
noise ratio (G, p = 0.022, paired t-test) of recorded A1 L2/3 neurons without (OFF) and with (ON)
LP silencing. **p < 0.01, *p < 0.05, n = 8 cells in 4 animals. Data points for the same cell are
connected with a line. (H) Left, experimental condition. AAV-ChR2 was injected into LP and blue
LED was delivered to activate LP neurons. Right, example image showing the expression of ChR2
within LP. Scale bar, 500 m. (I) TRF of an example A1 L2/3 pyramidal neuron without (left) and
with (right) optogenetic activation of LP neurons. (J) Normalized frequency tuning curves without
(black) and with (blue) LP activation. (K –N) Normalized spontaneous firing rate (K, ***p < 0.001,
paired t-test), evoked firing rate (L, p < 0.001, paired t-test), tuning bandwidth at 60 dB SPL (M,
p < 0.001, paired t-test), and SNR (N, p = 0.0059, paired t-test) of A1 L2/3 neurons without and
with LP activation (n = 13 cells in 5 animals). (O –P) Plot of normalized firing rates evoked by
effective tones (at 60 dB SPL) with vs. without LP silencing (O) or LP activation (P) for the
example cells shown above. Green and blue lines are the best fit linear regression line. Gray dashed
line is the identity line. (Q –S) Summary of parameters of linear fitting for all neurons in LP
silencing (ArchT) and activation (ChR2) groups, respectively. R
2
(Q): 0.84 ± 0.09 (mean ± SD, n
= 8 cells) vs. 0.88 ± 0.08 (n = 13 cells); slope (R): 1.05 ± 0.17 (not significantly different from 1,
p = 0.41, Z-test) vs. 0.94 ± 0.17 (not significantly different from 1, p = 0.26, Z-test); y-intercept
(S): 0.23 ± 0.11 (significantly > 0, p < 0.001, Z-test) vs. -0.19 ± 0.12 (significantly < 0, p < 0.001,
Z-test); Bar = SEM.
72

Figure 22. LED illumination had no effect on auditory response
(A) PSTH for responses of an example A1 L2/3 neuron to tone (best frequency) stimulation
without (top) and with (bottom) optical silencing of LP. Black and green horizontal lines mark the
duration of tone and LED stimulation, respectively. (B) Cumulative distribution of percentage
changes in tone-evoked responses in optogenetic inactivation and GFP-control (n = 12 cells)
groups. The difference is significant: p < 0.001, two-sample Kolmogorov–Smirnov test. (C) PSTH
for responses of an example A1 L2/3 neuron without (top) and with (bottom) optical activation of
LP (marked by blue line). (D) Cumulative distribution of percentage changes in tone-evoked
responses in optogenetic activation and GFP-control (n = 13 cells) groups. The difference is
significant: p < 0.001, two-sample Kolmogorov–Smirnov test.
73

Figure 23. LP activity manipulation did not change CF of TRF
(A –B) Summary of the CF values of A1 L2/3 neurons without (OFF) and with (ON) optical LP
silencing (A, p = 1.0, paired t-test, n = 8 cells) or activation (B, p = 1.0, paired t-test, n = 13 cells).
N.S., not significant.

Figure 24. Effects on BW20
(A –B) Normalized bandwidth of frequency tuning curves 20 dB above the intensity threshold
(BW20) without and with LP silencing (A, p = 0.0012, paired t-test, n = 8 cells) or activation (B,
p < 0.001, paired t-test, n = 13 cells). ***p < 0.001, **p < 0.01.

74

Figure 25. No effects in L4 of A1
(A –B) Normalized spontaneous (A, p = 1.0, Wilcoxon signed-rank test, n = 5 cells) and evoked
firing rates (B, p = 0.62, Wilcoxon signed-rank test, n = 5 cells) in A1 L4 neurons without (OFF)
and with (ON) LP silencing. (C –D) Normalized spontaneous (C, p = 0.23, paired t-test, n = 12
cells) and evoked firing rates (D, p = 0.70, paired t-test, n = 12 cells) in A1 L4 neurons without
and with LP activation.

3.2.3 LP modulation of A1 is mainly mediated by the LP-A1 projection
LP may modulate A1 responses through the direct LP to A1 projection, or indirectly through the
LP projections to secondary cortices (Arend et al., 2008b; Cappe et al., 2009; Hackett et al., 1998;
De La Mothe et al., 2006; Nakamura et al., 2015; Oh et al., 2014). We wondered whether the
direct LP-A1 projection contributed mainly to the observed modulatory effect of LP. By injecting
a retrograde dye, CTB-488, in A1, we examined the distribution of LP neurons projecting to A1
(Figure 26A). Numerous retrogradely labelled neurons were found in LP (Figure 26A, right),
with a bias towards its caudal part (Figure 27). Furthermore, injection of AAV-GFP into LP
revealed densely labeled axon terminals in L1 as well as deep layers of A1 (Figure 26B),
75
confirming the direct projection from LP to A1 (De La Mothe et al., 2006; Nakamura et al., 2015).
We also observed dense LP projections to L4 of secondary auditory cortices (data not shown). To
test whether the LP-A1 projection could account for the LP modulatory effect in A1, we
optogenetically silenced the LP-A1 axon terminals by injecting AAV-eNpHR3.0 in LP and placing
an optic fiber on the exposed A1 surface (Figure 26C). We observed effects on A1 L2/3 neurons
similar to silencing LP neurons per se: the spontaneous and evoked FR was increased, the
frequency tuning bandwidth was broadened, and the SNR was reduced (Figure 26D-G; Figure
28A). Conversely, we optogenetically activated the LP-A1 axon terminals by injecting AAV-
ChR2 in LP and shining blue light on the exposed A1 surface (Figure 26H). In line with the
effects caused by LP activation, the activation of LP-A1 axon terminals reduced the spontaneous
and evoked FR, sharpened frequency tuning, and enhanced the SNR in A1 L2/3 pyramidal neurons
(Figure 26I-L; Figure 28B).
We performed similar linear fitting on tone-evoked responses with versus without terminal
manipulations (Figure 26M, 26N). We observed a thresholding effect similar to manipulating LP
neurons per se: the linearity was high (Figure 26O), the slope of the best fit line was close to 1
(Figure 26P), and the y-intercept was a positive value for the terminal silencing while a negative
value for the terminal activation (Figure 26Q). These data suggest that the LP to A1 projection
largely mediates the LP modulatory effect on A1 responses.
76

Figure 26. Bidirectional activity manipulations of the LP-A1 projection
(A) Left, injection of CTB-488 into A1. Middle, image showing fluorescence at the injection site.
Scale bar, 500 m. Right, retrogradely labelled neurons in LP. Scale bars, 200 m. (B) Left,
injection of AAV-GFP into LP. Middle, expression of GFP at the injection site. Scale bar, 500 m.
Right, GFP-labeled LP axons in A1 and average normalized fluorescence intensity across cortical
depths (n = 6 brain sections, black line and green shade represent mean ± SD respectively). Scale
bar, 100 m. Cortical layers are marked. (C) Injection of AAV-eNpHR3.0 into LP and optical
silencing of LP-A1 axon terminals by placing the optic fiber over A1. (D –G) Normalized
spontaneous firing rate (D, p = 0.013, paired t-test, n = 9 cells in 4 animals), evoked firing rate (E,
p = 0.012, paired t-test), tuning bandwidth (F, p = 0.0078, Wilcoxon signed-rank test), and SNR
(G, p = 0.015, paired t-test) of A1 L2/3 neurons without and with LP-A1 axon terminal silencing.
**p < 0.01, *p < 0.05. (H) Injection of AAV-ChR2 into LP and optical activation of LP-A1 axon
terminals. (I-L) Normalized spontaneous firing rate (I, p = 0.0038, paired t-test, n = 10 cells in 4
animals), evoked firing rate (J, p = 0.013, paired t-test), tuning bandwidth (K, p = 0.009, paired t-
test), and SNR (L, p = 0.0081, paired t-test) of A1 L2/3 neurons without and with LP-A1 axons
terminal activation. (M-N) Plot of normalized firing rates evoked by effective tones (at 60 dB SPL)
with vs. without LP-A1 axon terminal silencing (M) or activation (N) for two example cells. (O –
77
Q) Summary of parameters of linear fitting in terminal silencing (eNpHR3.0) and activation (ChR2)
groups, respectively. R
2
(O): 0.87 ± 0.098 (n = 9 cells) vs. 0.84 ± 0.091 (n = 10 cells); slope (P):
1.05 ± 0.22 (not significantly different from 1, p = 0.48, Z-test) vs. 0.95 ± 0.18 (not significantly
different from 1, p = 0.42, Z-test); y-intercept (Q): 0.19 ± 0.077 (significantly > 0, p < 0.001, Z-
test) vs. -0.19 ± 0.083 (significantly < 0, p < 0.001, Z-test). Bar = SEM.

Figure 27. Projection from LP to A1
Images showing distribution of retrogradely labelled A1-prejecting LP neurons across rostral-
caudal axis. Scale bar, 200 m.

Figure 28. Effects of LP-A1 terminal manipulations on the TRF
(A –B) TRFs of two example A1 L2/3 neurons without (left) and with (right) LP-A1 terminal
silencing (A) or activation (B).

3.2.4 LP axons produce a disynaptic inhibitory effect on A1 L2/3 neurons
As the LP projection to the cortex per se is excitatory (Evangelio et al., 2018; Roth et al., 2016;
Zhou et al., 2018), an immediate question is how LP activity exerts a net inhibitory effect on A1
L2/3 pyramidal neurons. Since LP axons project to L1 (Figure 26B), which contains
predominantly inhibitory neurons (Jiang et al., 2013; Schuman et al., 2019), it is possible that LP
axons can indirectly suppress L2/3 pyramidal neurons via L1 inhibitory neurons (Ibrahim et al.,
78
2016; Jiang et al., 2013; Zhou et al., 2014). To test this idea, we injected AAV-ChR2 into LP and
made whole-cell voltage-clamp recordings from A1 neurons in slice preparations (Figure 29A).
Cuts were made in the tissue along boundaries between A1 and secondary auditory cortices to
prevent potential feedback input to A1 (see 3.4). TTX and 4AP were present in the bath solution
to ensure that only monosynaptic responses were recorded (Petreanu et al., 2009). We performed
whole-cell recordings from several types of neurons in A1: L1 inhibitory neurons labeled by
crossing GAD2-Cre with the Ai14 (Cre-dependent tdTomato) reporter, pyramidal neurons
identified as tdTomato-negative cells in GAD2-Cre::Ai14 animals, parvalbumin (PV) and
somatostatin (SOM) positive inhibitory neurons labeled by crossing PV-Cre or SOM-Cre with
Ai14, respectively. As shown by two example cells, blue light activation of LP-A1 axons resulted
in a robust excitatory postsynaptic current (EPSC) in the L1 inhibitory neuron, whereas no EPSC
was observed in the pyramidal (PYR) cell (Figure 29B). Overall, none of the PYR neurons we
recorded across L2-4 received direct input from LP, whereas more than 50% of L1 inhibitory
neurons received direct LP input (Figure 29C, 29E). A similar fraction of PV neurons in L2/3
received direct input from LP, whereas none of the recorded SOM neurons did so (Figure 29D,
29E). A much smaller fraction of PV neurons in L4 also received LP input (Figure 29E).
Together, these results indicate that LP-A1 axons preferentially innervate L1 inhibitory neurons
and superficial-layer PV inhibitory neurons, which may then provide disynaptic inhibition to L2/3
pyramidal neurons.
79

Figure 29. A1 cell types innervated by LP axons
(A) Slice recording paradigm. AAV-ChR2 was injected into LP. Blue LED light was applied to
A1 to active LP-A1 axon terminals. Whole-cell recordings were made from A1 neurons. (B) Light-
evoked monosynaptic EPSC (average trace) recorded (at -70 mV) in an example L2/3 pyramidal
(PYR, top) and L1 inhibitory neuron (bottom). TTX and 4AP were present in the bath solution.
Blue arrow marks the onset of 5-ms light pulse. Scale: 200 ms and 20 pA. (C) Plot of average
amplitude of light-evoked monosynaptic EPSCs against the cell’s cortical depth for all the
recorded L1 inhibitory neurons (red, GAD2+) and pyramidal cells (black). Dashed line marks the
boundary between L1 and L2/3. (D) Plot of average amplitude of light-evoked monosynaptic
EPSCs against the cell’s cortical depth for the recorded PV (magenta) and SOM (blue) neurons.
(E) Summary of connection probability between LP-A1 axons and L1 GAD2+ neurons, as well as
pyramidal, PV and SOM neurons in different layers. L1 vs. L2/3 PV, p = 0.12; L1 vs. L4 PV, ***p
< 0.001; L2/3 PV vs. L4 PV, ***p < 0.001, Fisher’s exact test. N.S., not significant.

3.2.5 LP plays a role in noise-related contextual modulation of A1 responses
Previously, it has been proposed that LP can provide contextual information to visual cortex (Fang
et al., 2020; Roth et al., 2016). Whether LP could play a similar role in auditory processing has
been unknown. In an acoustic environment, one common contextual factor is the background
noise. It has been shown that elevating the background noise level results in narrowing of
frequency tuning of A1 neurons without changing the tuning preference through a thresholding
effect (Liang et al., 2014). Since LP manipulations also produce a thresholding effect, we
80
wondered whether LP could contribute to the noise-related contextual modulation. We noticed
that LP neurons responded robustly to white-noise sound, with the response amplitude increasing
with increasing noise levels (Figure 30A). Previous anatomical and electrophysiological studies
have shown that the superior colliculus (SC) in the midbrain innervates LP (Beltramo and
Scanziani, 2019; Bennett et al., 2019; Fang et al., 2020; Gale and Murphy, 2014; Stepniewska et
al., 2000; Wei et al., 2015; Zingg et al., 2017). We thus wondered whether these noise responses
in LP could be driven by SC input. Comparing the onset latency of noise responses in SC, LP,
and A1 L4, we found that it was the shortest in SC, while similar between LP and A1 L4 (Figure
30B). This suggests that the responses in LP (at least the early part) are unlikely due to feedback
inputs from auditory cortices. Furthermore, silencing SC greatly reduced the amplitude of noise
responses in LP (Figure 30C). Our results thus indicate that the noise responses in LP are
primarily driven by bottom-up input likely from SC.
To test whether the noise-driven LP activity affects A1 frequency processing, we applied tones (of
varying frequencies) embedded in broadband noise of different levels (Figure 30D, 30E).
Frequency tuning of A1 neurons was compared before and after silencing LP with bupivacaine
(Fang et al., 2020; Lee et al., 2008; Moraga-Amaro et al., 2014). As shown by an example A1
L2/3 neuron (Figure 30D, 30E, gray), increasing the noise level reduced the amplitude of the
response evoked by the best-frequency tone. Silencing LP resulted in a general increase in the
tone-evoked response, regardless of the noise level, but to different degrees (Figure 30D, 30E,
red). Summarizing all the recorded L2/3 neurons, we found that silencing LP universally enhanced
tone-evoked responses in A1 across different noise levels (Figure 30F, upper panel). Notably, the
enhancement was larger at higher noise levels (Figure 30F, lower panel). This is consistent with
the notion that LP neuron responses increase with increasing noise levels (Figure 30A) and
81
therefore exert a larger suppressive effect in A1 under a higher noise background. Silencing LP
also elevated the spontaneous FR in a noise-level dependent manner so that the elevation was
larger at higher noise levels (Figure 30G). The SNR slightly decreased with increasing noise
levels in the control condition (Figure 30H, upper panel, black), indicating that high-level noise
has a detrimental effect on SNR, thus deteriorating auditory processing. Silencing LP not only
reduced SNR, but also accelerated the detrimental effect of background noise (Figure 30H, upper
panel, red). Again, the modulatory effect on SNR was larger at higher noise levels (Figure 30H,
lower panel). Finally, the frequency tuning bandwidth was reduced with increasing noise levels
(Figure 30I, upper panel), consistent with previous studies (Ehret and Schreiner, 2000; Liang et
al., 2014). Silencing LP not only broadened the tuning bandwidth, but also slowed down the
modulation of tuning bandwidth by increasing the noise level (Figure 30I). Together, these results
suggest that LP plays a role in contextual modulation of A1 frequency processing by noise
background.
82

Figure 30. LP plays a role in noise-related contextual modulation of A1 responses
(A) Left, PSTH for responses of an example neuron in the caudal LP to white noise sound (marked
by black line). Right, mean noise-evoked firing rate plotted against noise intensity (n = 10 LP
neurons). Bar = SD. (B) Left, onset latency of noise (at 60 dB SPL) evoked responses in SC (n =
17 cells), LP (n = 17) and A1 L4 (n = 8) neurons. ***p < 0.001, One-way ANOVA with
Bonferroni's multiple comparisons test. N.S., not significant. (C) Normalized noise-evoked firing
rate of LP neurons before (Ctrl) and after silencing SC (ΔSC). ***p < 0.001, paired t-test, n = 12
cells in 2 animals. (D) Left, spectrogram of the stimulus: a 50-ms CF tone (at 60 dB SPL)
embedded in low-level noise (at 0 dB SPL, 250 ms duration). Right, PSTHs for responses of an
example A1 L2/3 neuron to the tone embedded in noise before (black) and after (red) silencing LP
with bupivacaine. Black line marks the tone duration. (E) Response of the same cell to the same
tone (60 dB SPL) embedded in higher-level noise (45 dB SPL) before and after silencing LP. (F)
Upper, summary of evoked firing rates of A1 neurons at different noise levels before (black) and
after (red) silencing LP. Lower, change in evoked firing rate by LP silencing at different noise
levels. ***p < 0.001, **p < 0.01, *p < 0.05, paired t-test, compared to the values under 0 dB noise
condition, n = 9 cells from 4 animals. Bar = SD. (G) Summary of spontaneous firing rates before
and after silencing LP. (H) Summary of SNR before and after silencing LP. (I) Summary of tuning
bandwidths before and after silencing LP.

83
3.2.6 SC can drive the LP-mediated modulation of A1 responses
Since SC innervates LP (Beltramo and Scanziani, 2019; Bennett et al., 2019; Gale and Murphy,
2014; Stepniewska et al., 2000; Wei et al., 2015; Zingg et al., 2017) but does not project to auditory
cortices (Basso and May, 2017; Cang et al., 2018; Ito and Feldheim, 2018), we wondered whether
SC could provide direct input to drive the LP-mediated modulation of A1 responses. To confirm
the connectivity from SC to LP, we injected AAV-GFP into intermediate and deep layers, the
auditory related part of SC (Bednárová et al., 2018; Drager and Hubel, 1975; King and Palmer,
1985; Meredith and Stein, 1986; Wise and Irvine, 1983; Zingg et al., 2017). We found abundant
GFP-labeled axons in LP (Figure 31A), with a strong bias towards its caudal part (Figure 32).
Expressing ChR2 in intermediate and deep layers of SC and then performing whole-cell slice
recording from caudal LP neurons in the presence of TTX and 4AP further confirmed direct
innervations of LP neurons by SC axons (Figure 31B, 31C).
We next expressed ChR2 in intermediate and deep layers of SC (Figure 31D). Optogenetic
activation of these SC neurons induced effects on A1 L2/3 pyramidal neurons similar to activation
of LP: spontaneous and evoked firing rates were decreased, and SNR was increased (Figure 31E-
G). We also expressed ChR2 in SC-recipient LP neurons (Zingg et al., 2017) by first injection of
AAV-Cre in SC and second injection of AAV encoding Cre-dependent ChR2 in the caudal LP
(Figure 31H). In A1, optogenetic activation of SC-recipient LP neurons produced similar effects
to the activation of the general LP population: spontaneous and evoked firing rates were reduced,
and SNR was increased (Figure 31I-K). Conversely, we silenced SC by infusing bupivacaine
(Figure 31L). This resulted in increases of spontaneous and evoked firing rates, broadening of
frequency tuning and a reduction of SNR in A1 L2/3 pyramidal neurons (Figure 31M-P), similar
84
to silencing LP per se. Together, these results provide supporting evidence that SC can provide
input to drive LP-mediated modulation of A1 responses.

Figure 31. SC can drive LP-mediated modulation of A1 responses
(A) Injection of AAV-GFP into intermediate and deep layers of SC. Left, expression of GFP in
SC. Boundaries between superficial, intermediate and deep layers are marked by dashed curves.
Right, GFP labeled SC axons within LP. Scale bar, 500 m. (B) Left, slice recording paradigm:
expressing ChR2 in SC and whole-cell recording from LP neurons. Right, light-evoked
monosynaptic EPSC (average trace) in an example LP neuron before and after perfusing in CNQX.
Scale: 100 pA and 50 ms. (C) Average amplitudes of light-evoked monosynaptic EPSCs in LP
neurons. Neurons not showing a light-evoked EPSC were excluded. (D) Left, illustration of optic
activation of SC and recording in A1. Right, expression of ChR2 in SC. Scale bar, 500 m. (E –
G) Normalized spontaneous FR (E), evoked FR (F), and SNR (G) of A1 neurons without and with
optical SC activation. ***p < 0.001, **p = 0.0065, paired t-test, n = 9 cells in 4 animals. (H)
Transsynaptic labeling of SC-recipient LP neurons by first injection of AAV-Cre into intermediate
and deep layers of SC and second injection of Cre-dependent ChR2 virus in LP. Right, expression
of ChR2-EYFP in LP. Scale bar, 500 m. (I –K) Normalized spontaneous FR (I), evoked FR (J),
and SNR (K) of A1 L2/3 neurons without and with optical activation of SC-recipient LP neurons.
85
***p < 0.001, **p = 0.0022, paired t-test, n = 9 cells in 5 animals. (L) Experimental paradigm:
silencing SC by infusing bupivacaine and recording in A1. (M –P) Normalized spontaneous FR
(M), evoked FR (N), tuning bandwidth at 60 dB SPL (O) and SNR (P) of A1 L2/3 neurons before
and after SC silencing. ***p < 0.001, **p = 0.0030, *p = 0.017, paired t-test, n = 10 cells in 7
animals.

Figure 32. SC projection to LP
Images showing anterogradely labelled SC axons in LP across rostral-caudal axis. Scale bar, 500
m.

3.2.7 LP mediates visual stimuli induced modulation of auditory responses in A1
Both SC and LP process visual information and have been implicated in visual looming stimuli
induced defensive behaviors such as freezing (Wei et al., 2015; Yilmaz and Meister, 2013; Zingg
et al., 2017). Directly stimulating superficial layer SC neurons, which receive visual input (Zhao
et al., 2014; Zhou et al., 2017), elicited freezing responses in mice (Zingg et al., 2017). Notably,
LP neurons responded more strongly to visual looming stimuli compared to more commonly used
grating and noise-pattern stimuli (Figure 33). These results imply that visual looming stimuli
might be able to modulate A1 auditory responses via the SC-LP pathway. To test this idea, we
paired sound stimulus with visual looming stimulus (Figure 34A), which was an expanding dark
disk presented from the upper visual field (see 3.4), and compared auditory responses of A1 L2/3
pyramidal neurons with and without coupling the visual stimuli. In the presence of visual looming
stimuli, spontaneous and evoked firing rates were reduced (Figure 34B, 34C), while the SNR was
increased (Figure 34D). These results indicate that visual looming stimuli can indeed modulate
86
A1 responses. After silencing LP with muscimol (Figure 34E), the effects on A1 responses by
visual looming stimuli disappeared (Figure 34F-H), suggesting that LP primarily mediates the
visual looming stimuli induced modulation of auditory responses in A1.

Figure 33. LP neurons are activated strongly by visual looming stimuli
(A) PSTH for responses of an example LP neuron to visual looming stimuli. (B) Evoked firing
rates of LP neurons to visual moving gratings, flash white-noise patterns and looming stimuli,
assayed by single-unit recording (n = 14, 20, and 13 cells from 2 animals, respectively). *p =
0.049, **p = 0.0090, ***p < 0.001, unpaired t-test with Bonferroni’s correction. Bar represents
mean ± SEM.

Figure 34. Visual looming stimuli modulate A1 auditory responses via LP
87
(A) Pairing of visual looming (V) and sound (S) stimulation. (B –D) Normalized spontaneous FR
(B), evoked FR (C), and SNR (D) of A1 L2/3 neurons under sound only (S) and sound plus visual
looming (S+V) condition. ***p < 0.001, paired t-test, n = 12 cells from 2 animals. (E) Testing the
effect of visual looming on A1 responses under condition of silencing LP with muscimol. (F –H)
Normalized spontaneous FR (E, p = 0.25, paired t-test, n = 12 cells from 2 animals), evoked FR
(F, p = 0.42, paired t-test, n = 12 cells), and SNR (G, p = 0.79, paired t-test, n = 12 cells) of A1
neurons when LP was silenced. N.S., not significant.

3.3 Discussion
LP is considered a higher-order thalamic nucleus. In general, the influence of higher-order
thalamus on auditory cortical processing has remained obscure. In the present study, our results
demonstrate that LP activity can modulate auditory processing in superficial layers of A1. The
overall outcome of increasing LP activity is to improve auditory processing by sharpening
frequency tuning and enhancing SNR of auditory evoked responses. This is achieved by
subtractive suppression of auditory evoked responses together with suppression of spontaneous
firing activity. We also demonstrate that such modulatory effect is largely mediated by the direct
projection of LP to A1, where LP axons preferentially innervate L1 inhibitory neurons and
superficial-layer PV inhibitory neurons, leading to a net disynaptic inhibitory effect on L2/3
pyramidal neurons. The suppression of A1 L2/3 responses is consistent with our recent study
showing a similar suppressive effect in L2/3 of the primary visual cortex (V1), which results in an
enhancement of visual feature selectivity (Fang et al., 2020).
The subtractive effect on L2/3 responses suggests that L1 inhibitory neurons (and PV neurons in
superficial layers as well) may be broadly tuned for frequency. The frequency tuning property of
L1 inhibitory neurons in the auditory cortex has not been studied yet, but a previous study in the
visual cortex does suggest that L1 inhibitory neurons are broadly tuned for visual features such as
orientation and direction (Mesik et al., 2019), and PV neurons in superficial layers of A1 have
88
been shown to be broadly tuned for tone frequency (Cohen and Mizrahi, 2015; Li et al., 2015;
Liang et al., 2018; Maor et al., 2016). As for the local circuit involved, we suggest that both L1
inhibitory neurons and L2/3 PV neurons, which receive direct LP input, can directly inhibit L2/3
pyramidal cells (Jiang et al., 2013), generating an overall suppressive effect. For L1 inhibitory
neurons, their molecular identities have been reported to be diverse: 70% are NDNF+, 20% are
nAchR+, and 10% are VIP+ (Schuman et al., 2019). It will be of great interest to investigate which
of these cell types contributes to the LP-mediated modulation of A1 responses by using cell-type
specific mouse lines. In addition, whether VIP+ neurons in L2/3 can also be involved in the
suppressive effect remains to be tested.
The LP-mediated modulatory effect on A1 processing may be particularly pronounced and
beneficial when there is a high noise background. A previous study has demonstrated that
increasing background noise is equivalent to lowering the intensity of test (e.g. tone) stimuli, i.e.
shifting up the TRF by a certain Δthreshold value (Liang et al., 2014). In the current study, we
demonstrate that LP activity also exerts a similar thresholding effect on A1 frequency tuning. Our
results suggest that the noise effect on A1 frequency tuning may be achieved, at least partially,
through LP. The contribution of LP to noise-related contextual modulation of A1 frequency
processing is in a positive manner, in that it helps not only to prevent SNR of auditory responses
from being deteriorated quickly by high-level noise background (Figure 30H) but also to
accelerate the sharpening of frequency tuning with increasing noise levels (Figure 30I), which
may compensate somewhat for the detrimental effect on SNR. Because LP activity is modulated
by the noise level in that loud noise activates LP neurons more than soft noise (Figure 30A), LP
neurons have a stronger protective effect on A1 processing under loud than soft noise conditions.
89
It is known that LP/pulvinar has extensive reciprocal connectivity with cortical areas including
visual and auditory cortices (Hackett et al., 1998; De La Mothe et al., 2006; Nakamura et al., 2015;
Oh et al., 2014; Tohmi et al., 2014; Zhou et al., 2018; Bennett et al., 2019). Previously, it has been
suggested that LP/pulvinar serves in a cortico-thalamo-cortical (“transthalamic”) indirect route for
information transfer from one cortical area to another (Guillery and Sherman, 2002; Sherman,
2016). Besides cortical inputs, LP also receive strong inputs from SC (Beltramo and Scanziani,
2019; Bennett et al., 2019; Fang et al., 2020; Gale and Murphy, 2014; Stepniewska et al., 2000;
Wei et al., 2015; Zingg et al., 2017). In this study, we demonstrate that LP (mainly the caudal part)
receives direct input from the auditory related part of SC and projects to A1, and that silencing SC
produces changes in A1 functional response properties similar to silencing LP. These results
together suggest that SC can relay bottom-up input to LP to drive its modulation of A1 processing.
In addition, SC is a multisensory structure (Drager and Hubel, 1975; King and Palmer, 1985;
Meredith and Stein, 1986). The visual only part of SC, i.e. the superficial layer of SC, also projects
strongly to the caudal LP (Beltramo and Scanziani, 2019; Bennett et al., 2019; Fang et al., 2020).
Thus, visual input may be able to modulate A1 responses as well via LP. Here, we demonstrate
that visual looming stimuli modulate A1 L2/3 responses in a similar manner as increasing LP
activity and that silencing LP blocks this modulation. This highlights a cross-modality feature of
LP modulation. We should note however that the experiment with pharmacological silencing of
LP (Figure 34E-H) does not exclude possible involvements of LP-mediated pathways other than
the bottom-up SC-LP pathway. For example, visual looming signals may reach LP through
feedback pathways from visual and associative cortices and then modulate responses in A1. We
postulate that the multisensory nature of SC and its strong projections to LP potentially endow LP
with an ability to modulate A1 activity given any salient sensory stimuli.
90
In summary, our results suggest that a previously unrecognized pathway, the SC-LP-A1
pathway, can provide contextual and cross-modality modulation of A1 responses and auditory
processing to enhance the salience of acoustic information. How this pathway may interact with
the canonical colliculo-thalamo-cortical auditory pathway remains to be investigated in the
future.

3.4 Material and methods
3.4.1 Experimental animals
All experimental procedures used in the present study were approved by the Animal Care and Use
Committee at the University of Southern California. Male and female wild-type (C57BL/6J) and
transgenic (GAD2-Cre, PV-Cre, SOM-Cre, and Ai14) mice were obtained from the Jackson
Laboratory. Animals aged 8-12 weeks and weighed 18-28g were used in the experiments. Animal
sample sizes were determined by the estimated variances of the experiments and previous
experience from similar experiments and were sufficient for all the statistical testes. Mice were
housed under a 12 h light/dark cycle. Food and water were provided ad libitum. Randomization
methods were used to allocate experimental groups.
3.4.2 Viral and neural tracer injection
Viral injections were performed as previously described (Fang et al., 2020; Zingg et al., 2017). In
brief, stereotaxic coordinates were selected based on the Allen Mouse Brain Atlas (www.brain-
map.org). Mice were placed on a heating pad with homoeothermic control and anesthetized with
1.5% isoflurane throughout all surgical procedures. A small cut was made on the skin after shaving
91
to expose the skull. A craniotomy of 0.2 mm in diameter was made to expose the underlying cortex.
After removing the dura mater, a pulled glass pipette with a beveled tip of ~20 µm in diameter was
inserted to the target region. The viral solution or tracers were delivered by pressure injection. For
optogenetic silencing and activating LP and its terminals, AAV1-CAMKII-hChR2-eYFP (UPenn
Vector Core, 1.7×10
13
GC/ml), AAV1-CAG-ArchT-GFP (UPenn Vector Core, 1.7×10
13
GC/ml),
and AAV1-hSyn-eNpHR3.0-mCherry (UPenn Vector Core, 1.7×10
13
GC/ml) was injected into LP
(50 nl total volume, AP -2.4 mm, ML +1.6 mm, DV -2.4 mm) of wild-type animals, respectively.
For anterograde tracing of LP projections, AAV2/1-CB7-Cl-eGFP-WPRE-rBG (UPenn Vector
Core, 1.7×10
13
GC/ml) was injected into LP. For retrograde tracing of LP afferents, we injected
CTB488 into A1 (AP -2.6 mm, ML +4.4 mm, DV +0.6 mm). For optogenetically activating SC
axonal terminals in LP for slice recording, AAV1-CAMKII-hChR2-eYFP (UPenn Vector Core,
1.7×10
13
GC/ml) was injected into SC (AP -3.75 mm, ML +0.6 mm, DV -1.45 mm). For
transsynaptic labelling from SC to LP, AAV2/1-hSyn-Cre-WPRE-hGH (UPenn Vector Core,
2.5×10
13
GC/mL) was injected into SC, and AAV2/1-EF1a-DIO-hChR2-eYFP (UPenn Vector
Core, 1.6×10
13
GC/mL) was injected into LP. Animals were allowed to recover for at least 3 weeks
following the injections of viruses.
3.4.3 Histology and imaging
After experiments, animals were deeply anesthetized and transcardially perfused with phosphate
buffered saline (PBS) followed by 4% paraformaldehyde (PFA). The brain tissue was collected
and fixed in 4% PFA at 4 ˚C overnight and then sliced into 150 μm sections with a vibratome
(Leica, VT1000s). Nissl staining was used to visualize the cyto-architecture. Brain slices were first
rinsed with PBS for 10 min, and then incubated in PBS containing Neurotrace 620 (ThermoFisher,
N21483) and 0.1% Triton X-100 (Sigma-Aldrich) for 2 h. Images were taken with a confocal
92
microscope (Olympus FluoView FV1000). To identify and verify the injection site and spread of
virus expression and drug infusion, images were taken under a 4× objective. Regions with axonal
labeling were further imaged under a 10× objective for clearer visualization of innervation patterns.
3.4.4 Optogenetic and pharmacological manipulation
One week before recording sessions, the animals were prepared as previously described ( Chou et
al., 2018; Fang et al., 2020). In brief, mice were placed on the stereotaxic apparatus and
anesthetized with 1.5% isoflurane during the implantation. Optic fiber implantation was made at
least three weeks after injecting ChR2 or ArchT virus. For implantation, small holes of 500 µm in
diameter were drilled to allow the insertion of microinjection tubes (300 µm ID, RWD) or the fiber
optic cannula (200 µm ID, Thorlabs). The holes were drilled directly above LP (AP -2.5 mm, ML
+1.6 mm), or A1 (AP -2.6 mm, ML +4.4 mm). The injection tube or cannula were lowered to the
desired depth and fixed in place using dental cement. In the meantime, a screw for head fixation
was mounted on top of the skull with dental cement as well. For drug infusion during recording,
an injector (100 µm ID, RWD) was inserted into the microinjection tube, and 200 nl 0.5%
bupivacaine mixed with DiI (2 mg/ml) or fluorescent muscimol (1.5 mM, Life Technologies) were
slowly injected into SC (AP -3.75 mm, ML +0.6 mm, DV -1.45 mm) or LP (AP -2.5 mm, ML
+1.6 mm, DV -2.4 mm) through a micro-syringe. Since bupivacaine’s silencing effect lasts for
only 30-40 min (Fang et al., 2020), it was possible for us to sequentially record from multiple A1
neurons in the same animal and examine effects on them by silencing LP/SC. Optogenetic
manipulations were performed following our previous studies (Chou et al., 2018; Fang et al., 2020;
Ibrahim et al., 2016). To optogenetically activate LP, light from a blue LED source (470 nm, 10
mW, Thorlabs) was delivered at a rate of 20 Hz (20-ms pulse duration) via the implanted cannulas
using a patch cord (Ø200 µm, 0.22 NA SMA 905, Thorlabs) for ChR2 animals. The plastic sleeve
93
(Thorlabs) securing the patch cord and cannula was wrapped with black tape to prevent light
leakage. Similarly, for silencing LP, light from a green LED source (530 nm, 10 mW, Thorlabs)
was delivered continuously for stimulating ArchT-expressing neurons. For manipulation of
terminals from LP to A1, optic cannula connected with a blue and an amber LED (589 nm, 10 mW,
Thorlabs) source was placed on the surface of A1 to stimulate ChR2- and eNpHR3.0-expressing
LP axons, respectively. The optogenetic stimulation preceded sound stimulation by 50 ms.
Animals were allowed to recover for one week before recording session. During the recovery
period, they were habituated to the head fixation on the flat running plate. The head screw was
tightly fit into a metal post while the animal could run freely on the plate. Following recording
sessions, animals were euthanized, and the brain was imaged to verify the specificity of virus
expression and locations of implantations. Mice with mistargeted injections or misplacements of
drug infusion tubes or optic fibers were excluded from data analysis.
3.4.5 Slice preparation and recording
Three weeks following the injections, animals were decapitated following urethane anesthesia and
the brain was rapidly removed and immersed in an ice-cold dissection buffer (composition: 60 mM
NaCl, 3 mM KCl, 1.25 mM NaH2PO4, 25 mM NaHCO3, 115 mM sucrose, 10 mM glucose, 7 mM
MgCl2, 0.5 mM CaCl2; saturated with 95% O2 and 5% CO2; pH= 7.4). Coronal slices at 350 µm
thickness were sectioned by a vibrating microtome (Leica VT1000s), and recovered for 30 min in
a submersion chamber filled with warmed (35 °C) ACSF (composition:119 mM NaCl, 26.2 mM
NaHCO3, 11 mM glucose, 2.5 mM KCl, 2 mM CaCl2, 2 mM MgCl2, and 1.2 mM NaH2PO4, 2
mM Sodium Pyruvate, 0.5 mM Vitamin C). A1 or LP neurons were visualized under a fluorescence
microscope (Olympus BX51 WI). Patch pipettes (~4 -5 MΩ resistance) filled with a cesium-based
internal solution (composition: 125 mM cesium gluconate, 5 mM TEA-Cl, 2 mM NaCl, 2 mM
94
CsCl, 10 mM HEPES, 10 mM EGTA, 4 mM ATP, 0.3 mM GTP, and 10 mM phosphocreatine;
pH = 7.25; 290 mOsm) were used for whole-cell recordings. Signals were recorded with an
Axopatch 700B amplifier (Molecular Devices) under voltage clamp mode at a holding voltage of
–70 mV for excitatory currents or 0 mV for inhibitory currents, filtered at 2 kHz and sampled at
10 kHz. Tetrodotoxin (TTX, 1 μM) and 4-aminopyridine (4-AP, 1 mM) were added to the external
solution to isolate monosynaptic responses. Blue light stimulation (10 ms pulse, 3 mW power, 10-
30 trials) was delivered via a mercury Arc lamp gated with an electronic shutter. PV+, SOM+ or
L1 GAD2+ neuronal types were determined by the tdTomato expression. Laminar location of the
recorded cell was determined by its depth from the pial surface. CNQX (10 µM) was added to
verify glutamatergic transmission.
3.4.6 Sound and visual stimulation
White noise (50-ms, 70 dB sound pressure level or SPL) or tone pips (50-ms duration, 3-ms ramp)
of various frequencies (2–45.25 kHz, 0.1 octave interval) and intensities (10–70 dB SPL, at 10-dB
interval) were generated by custom-made software in LabView (National Instruments) through a
16-bit National Instruments interface, and delivered through a calibrated speaker (Tucker-Davis
Technologies) to the contralateral ear. The 322 testing stimuli were presented in a pseudorandom
sequence. The inter-stimulus interval between noise stimulation or tone pips was 1 s. For auditory
stimuli embedded in noise of different levels, 50-ms tone pips at different intensities and
frequencies as described above were embedded in white noise of different intensity levels (0, 15,
30, 45 dB SPL). The white noise was 250 ms in duration and preceded the tone pip by 100 ms.
Visual stimuli were generated in Matlab (MATLAB) with the Psychophysics Toolbox Version 2
(Brainard, 1997) and were presented on a ViewSonic VA705b monitor (1920 × 1440 pixels, 33.9
95
cm wide, 27.2 cm high, 60 Hz refresh rate, mean luminance 41 cd/m
2
) mounted on a flexible arm.
For visual noise stimulation, a set of dense white-noise patterns were presented. Each frame of the
stimuli consisted of a grid of 20 × 20 squares (each square 4° × 3°), intensities of which were
determined by a m-sequence. Each pattern was presented for 200 ms and 30-50 patterns were
presented according to the response level and fidelity. For grating stimulation, drifting sinusoidal
gratings (12 directions, 0°-330°, 30° per step, 6 orientations) were presented in a pseudorandom
order for 5-8 repetitions. The spatial frequency of the gratings was chosen to be 0.04 cycles per
second (cpd) and temporal frequency to be 2 cycles per second (Hz). Each grating drifted for 1.5
s and another grating appeared, which remained to be static for 3 s before drifting. The looming
stimulus was an expanding black disk presented from the upper visual field. It changed from 2° to
20° size (in diameter) within 250 ms. The largest black disk after the expansion stayed for another
250 ms, and the interstimulus interval was 250 ms. The auditory stimulus (tone pips as described
above) was paired with each cycle of the looming stimulation with an onset delay of 200 ms. We
usually only presented 30 cycles of looming stimulation to avoid a potential adaptation effect.
3.4.7 In vivo electrophysiology
One week after the preparation, animals were head-fixed on the running plate, and
electrophysiology recording with either optogenetic or pharmacological manipulation was carried
out in a sound-attenuation booth. Loose-patch recordings were performed as previously described
(Fang et al., 2020; Ibrahim et al., 2016; Liang et al., 2018; Zhou et al., 2014), with a patch pipette
filled with an artificial cerebral spinal fluid (ACSF; 126 mM NaCl, 2.5 mM KCl, 1.25 mM Na2PO4,
26 mM NaHCO3, 1 mM MgCl2, 2 mM CaCl2 and 10 mM glucose). A loose seal (0.1–0.5 GΩ) was
made on the cell body, allowing spikes only from the patched cell to be recorded. Signals were
recorded with an Axopatch 200B amplifier (Molecular Devices) under voltage-clamp mode, with
96
a command voltage applied to adjust the baseline current to near zero. Loose-patch recording
signals were filtered with a 100–5,000 Hz band-pass filter. The depths of the recorded neurons
were determined based on the micromanipulator reading. Spikes could be detected without
ambiguity because their amplitudes were normally higher than 100 pA, while the baseline
fluctuation was < 5 pA. Multichannel recordings were carried out by lowering a 64-channel
silicone probe (NeuroNexus) into the target region. Signals were recorded by an Open-Ephys
system. Multi-unit signals during sound stimulation were recorded and saved for offline analysis.
3.4.8 Data analysis
Noise-driven and tone-driven spike rates were analyzed within a 10-60 ms time window after the
onset of sounds. For quantifying evoked firing rates, average baseline firing rate calculated within
a 50 ms time window preceding the onset of sounds was subtracted. For quantifying changes in
spontaneous firing rate, spontaneous firing rates calculated within a 50 ms time window before
sound onsets were compared between LED-on and LED-off conditions. TRFs were reconstructed
according to the array sequence. The frequency–intensity space was up-sampled 5 times along the
frequency and intensity dimensions only for visualization purposes. The measurements of TRF
parameters (e.g. bandwidth) were made from the raw data. Boundaries of the spike TRF were
determined following previous studies (Liang et al., 2014; Schumacher et al., 2011; Sutter and
Schreiner, 1991; Xiong et al., 2013). A threshold at the value equal to the spontaneous spike rate
plus 20% of the peak evoked firing rate was then used to define significant evoked responses.
Responses to frequency–intensity combinations that met this criterion were considered to fall
within the TRF of the neuron, which generated the contour of the TRF (Xiong et al., 2013).
Characteristic frequency (CF) was defined as the frequency (in Hz) at which the lowest sound
pressure level was necessary to evoke a significant excitatory response. Bandwidth of TRF was
97
determined as the total frequency range for effective tones at 60 dB SPL or at 20 dB SPL above
the intensity threshold (i.e. BW20). SNR of auditory responses was calculated as the evoked firing
rate divided by the spontaneous firing rate. To quantify changes in evoked firing rate, average
tone-evoked firing rates across the TRF (under the control condition) were compared between
control and activity manipulation conditions. Onset latency of spike responses was determined
based on the generated peristimulus spike-time histogram (PSTH) as the interval between the
stimulus onset and the time point where spike rate exceeded the average baseline by 2 standard
deviations of baseline fluctuations.
To plot frequency tuning curve using the Envelope function (Sun et al., 2010), the tone-evoked
responses at 60 dB SPL with and without activity manipulation were normalized to the highest
evoked firing rate in the control condition. Linear regression fitting for the normalized evoked
firing rates with and without activity manipulation was performed for individual neurons in each
group, and R
2
, slope, and intercepts of the best fit line were determined.
For spike sorting of data obtained from multichannel recording, spike signals were filtered with a
300–3,000 Hz band-pass filter. The nearby four channels of the silicon probe were grouped as
tetrodes and semi-automatic spike sorting was performed using the offline sorter of Plexon (Dallas,
Texas) following our previous study (Zhang et al., 2018b). Clusters with isolation distance > 20
were considered as separate clusters. Spike clusters were classified as single units only if the
waveform SNR (signal-to-noise ratio) exceeded 4 (12 dB) and the inter-spike interval was longer
than1.2 ms for > 99.5% of the spikes.
3.4.9 Statistics
98
Shapiro–Wilk test was first applied to examine whether samples had a normal distribution, and a
p value < 0.05 indicated non-normality. In the case of a normal distribution, Z-test or two-sided
paired t-test was applied. Otherwise, the Wilcoxon signed-rank test was applied as a non-
parametric test. Two-sample Kolmogorov-Smirnov test was used to test whether data from two
groups were from the same distribution. Statistical analysis was conducted with Matlab. Data
were reported as mean ± SD unless otherwise mentioned.

99
References
Acuña, C., Gonzalez, F., and Dominguez, R. (1983). Sensorimotor unit activity related to
intention in the pulvinar of behaving Cebus Apella monkeys. Exp. Brain Res. 52, 411–422.
Ahmadlou, M., Zweifel, L.S., and Heimel, J.A. (2018). Functional modulation of primary visual
cortex by the superior colliculus in the mouse. Nat. Commun. 9, 3895.
Alitto, H.J., and Usrey, W.M. (2008). Origin and dynamics of extraclassical suppression in the
lateral geniculate nucleus of the macaque monkey. Neuron 57, 135–146.
Allen, A.E., Procyk, C.A., Howarth, M., Walmsley, L., and Brown, T.M. (2016). Visual input to
the mouse lateral posterior and posterior thalamic nuclei: photoreceptive origins and retinotopic
order. J. Physiol. 594, 1911–1929.
Almeida, I., Soares, S.C., and Castelo-Branco, M. (2015). The Distinct Role of the Amygdala,
Superior Colliculus and Pulvinar in Processing of Central and Peripheral Snakes. PLoS One 10,
e0129949.
Angelucci, A., Bijanzadeh, M., Nurminen, L., Federer, F., Merlin, S., and Bressloff, P.C. (2017).
Circuits and Mechanisms for Surround Modulation in Visual Cortex. Annu. Rev. Neurosci. 40,
425–451.
Arend, I., Rafal, R., and Ward, R. (2008a). Spatial and temporal deficits are regionally
dissociable in patients with pulvinar lesions. Brain 131, 2140–2152.
Arend, I., Machado, L., Ward, R., McGrath, M., Ro, T., and Rafal, R.D. (2008b). The role of the
human pulvinar in visual attention and action: evidence from temporal-order judgment, saccade
decision, and antisaccade tasks. Prog. Brain Res. 171, 475–483.
Audette, N.J., Bernhard, S.M., Ray, A., Stewart, L.T., and Barth, A.L. (2019). Rapid Plasticity of
Higher-Order Thalamocortical Inputs during Sensory Learning. Neuron 103, 277-291.e4.
Azeredo da Silveira, R., and Roska, B. (2011). Cell types, circuits, computation. Curr. Opin.
Neurobiol. 21, 664–671.
Baldwin, M.K.L., Wong, P., Reed, J.L., and Kaas, J.H. (2011). Superior colliculus connections
with visual thalamus in gray squirrels (Sciurus carolinensis): evidence for four subdivisions
within the pulvinar complex. J. Comp. Neurol. 519, 1071–1094.
Baldwin, M.K.L., Balaram, P., and Kaas, J.H. (2013). Projections of the superior colliculus to the
pulvinar in prosimian galagos (Otolemur garnettii) and VGLUT2 staining of the visual pulvinar.
J. Comp. Neurol. 521, 1664–1682.
Baldwin, M.K.L., Balaram, P., and Kaas, J.H. (2017). The evolution and functions of nuclei of
the visual pulvinar in primates. J. Comp. Neurol. 525, 3207–3226.
Basso, M.A., and May, P.J. (2017). Circuits for Action and Cognition: A View from the Superior
Colliculus. Annu. Rev. Vis. Sci. 3, 197–226.
Bednárová, V., Grothe, B., and Myoga, M.H. (2018). Complex and spatially segregated auditory
inputs of the mouse superior colliculus. J. Physiol. 596, 5281–5298.
Beltramo, R., and Scanziani, M. (2019). A collicular visual cortex: Neocortical space for an
ancient midbrain visual structure. Science 363, 64–69.
100
Bender, D.B. (1982). Receptive-field properties of neurons in the macaque inferior pulvinar. J.
Neurophysiol. 48, 1–17.
Bender, D.B. (1983). Visual activation of neurons in the primate pulvinar depends on cortex but
not colliculus. Brain Res. 279, 258–261.
Bennett, C., Gale, S.D., Garrett, M.E., Newton, M.L., Callaway, E.M., Murphy, G.J., and Olsen,
S.R. (2019). Higher-Order Thalamic Circuits Channel Parallel Streams of Visual Information in
Mice. Neuron 102, 477-492.e5.
Berman, R.A., and Wurtz, R.H. (2011). Signals conveyed in the pulvinar pathway from superior
colliculus to cortical area MT. J. Neurosci. 31, 373–384.
Bertini, C., Pietrelli, M., Braghittoni, D., and Làdavas, E. (2018). Pulvinar Lesions Disrupt Fear-
Related Implicit Visual Processing in Hemianopic Patients. Front. Psychol. 9, 2329.
Bonin, V., Mante, V., and Carandini, M. (2005). The suppressive field of neurons in lateral
geniculate nucleus. J. Neurosci. 25, 10844–10856.
Bourne, J.A., and Morrone, M.C. (2017). Plasticity of Visual Pathways and Function in the
Developing Brain: Is the Pulvinar a Crucial Player? Front. Syst. Neurosci. 11, 3.
Brainard, D.H. (1997). The Psychophysics Toolbox. Spat. Vis. 10, 433–436.
Cang, J., Savier, E., Barchini, J., and Liu, X. (2018). Visual Function, Organization, and
Development of the Mouse Superior Colliculus. Annu. Rev. Vis. Sci. 4, 239–262.
Cappe, C., Morel, A., Barone, P., and Rouiller, E.M. (2009). The Thalamocortical Projection
Systems in Primate: An Anatomical Support for Multisensory and Sensorimotor Interplay.
Cereb. Cortex 19, 2025–2037.
Casanova, C., and Molotchnikoff, S. (1990). Influence of the superior colliculus on visual
responses of cells in the rabbit's lateral posterior nucleus. Exp. Brain Res. 80, 387–396.
Chalupa, L.M., and Fish, S.E. (1978). Response characteristics of visual and extravisual neurons
in the pulvinar and lateral posterior nuclei of the cat. Exp. Neurol. 61, 96–120.
Chalupa, L.M., Williams, R.W., and Hughes, M.J. (1983). Visual response properties in the
tectorecipient zone of the cat’s lateral posterior-pulvinar complex: a comparison with the
superior colliculus. J. Neurosci. 3, 2587–2596.
Chou, X.-L., Wang, X., Zhang, Z.-G., Shen, L., Zingg, B., Huang, J., Zhong, W., Mesik, L.,
Zhang, L.I., and Tao, H.W. (2018). Inhibitory gain modulation of defense behaviors by zona
incerta. Nat. Commun. 9, 1151.
Cohen, L., and Mizrahi, A. (2015). Plasticity during motherhood: changes in excitatory and
inhibitory layer 2/3 neurons in auditory cortex. J. Neurosci. 35, 1806–1815.
D’Souza, R.D., Bista, P., Meier, A.M., Ji, W., and Burkhalter, A. (2019). Spatial Clustering of
Inhibition in Mouse Primary Visual Cortex. Neuron 104, 588-600.e5.
Desimone, R., and Duncan, J. (1995). Neural mechanisms of selective visual attention. Annu.
Rev. Neurosci. 18, 193–222.
Dhande, O.S., and Huberman, A.D. (2014). Retinal ganglion cell maps in the brain: Implications
for visual processing. Curr. Opin. Neurobiol. 24, 133–142.
101
Dhande, O.S., Stafford, B.K., Lim, J.-H.A., and Huberman, A.D. (2015). Contributions of
Retinal Ganglion Cells to Subcortical Visual Processing and Behaviors. Annu. Rev. Vis. Sci. 1,
291–328.
Dominguez-Vargas, A.-U., Schneider, L., Wilke, M., and Kagan, I. (2017). Electrical
Microstimulation of the Pulvinar Biases Saccade Choices and Reaction Times in a Time-
Dependent Manner. J. Neurosci. 37, 2234–2257.
Donnelly, J.F., Thompson, S.M., and Robertson, R.T. (1983). Organization of projections from
the superior colliculus to the thalamic lateral posterior nucleus in the rat. Brain Res. 288, 315–
319.
Drager, U.C. (1975). Receptive fields of single cells and topography in mouse visual cortex. J.
Comp. Neurol. 160, 269–290.
Drager, U.C., and Hubel, D.H. (1975). Responses to visual stimulation and relationship between
visual, auditory, and somatosensory inputs in mouse superior colliculus. J. Neurophysiol. 38,
690–713.
Durand, S., Iyer, R., Mizuseki, K., de Vries, S., Mihalas, S., and Reid, R.C. (2016). A
Comparison of Visual Response Properties in the Lateral Geniculate Nucleus and Primary Visual
Cortex of Awake and Anesthetized Mice. J. Neurosci. 36, 12144–12156.
Ellis, E.M., Gauvain, G., Sivyer, B., and Murphy, G.J. (2016). Shared and distinct retinal input to
the mouse superior colliculus and dorsal lateral geniculate nucleus. J. Neurophysiol. 116, 602–
610.
Evangelio, M., García-Amado, M., and Clascá, F. (2018). Thalamocortical Projection Neuron
and Interneuron Numbers in the Visual Thalamic Nuclei of the Adult C57BL/6 Mouse. Front.
Neuroanat. 12, 27.
Fang, Q., Chou, X.-L., Peng, B., Zhong, W., Zhang, L.I., and Tao, H.W. (2020). A Differential
Circuit via Retino-Colliculo-Pulvinar Pathway Enhances Feature Selectivity in Visual Cortex
through Surround Suppression. Neuron 105, 355-369.e6.
Fiebelkorn, I.C., and Kastner, S. (2020). Functional Specialization in the Attention Network.
Annu. Rev. Psychol. 71, 221–249.
Fiebelkorn, I.C., Pinsk, M.A., and Kastner, S. (2019). The mediodorsal pulvinar coordinates the
macaque fronto-parietal network during rhythmic spatial attention. Nat. Commun. 10, 215.
Fridman, M. (2019). Contextual modulation of visual thalamocortical circuits.
Gale, S.D., and Murphy, G.J. (2014). Distinct representation and distribution of visual
information by specific cell types in mouse superficial superior colliculus. J. Neurosci. 34,
13458–13471.
Ganz, J., Kaslin, J., Freudenreich, D., Machate, A., Geffarth, M., and Brand, M. (2012).
Subdivisions of the adult zebrafish subpallium by molecular marker analysis. J. Comp. Neurol.
520, 633–655.
Gattass, R., P.B. Sousa, A., and Oswaldo-Cruz, E. (1978). Single unit response types in the
pulvinar of the cebus monkey to multisensory stimulation. Brain Res. 158, 75–87.
Gharaei, S., Honnuraiah, S., Arabzadeh, E., and Stuart, G.J. (2020). Superior colliculus
modulates cortical coding of somatosensory information. Nat. Commun. 11, 1693.
102
Glickfeld, L.L., and Olsen, S.R. (2017). Higher-Order Areas of the Mouse Visual Cortex. Annu.
Rev. Vis. Sci. 3, annurev-vision-102016-061331.
Gomes, N., Soares, S.C., Silva, S., and Silva, C.F. (2018). Mind the snake: Fear detection relies
on low spatial frequencies. Emotion 18, 886–895.
Goodale, M.A., and Milner, A.D. (1992). Separate visual pathways for perception and action.
Trends Neurosci. 15, 20–25.
Grieve, K.L., Acuña, C., and Cudeiro, J. (2000). The primate pulvinar nuclei: vision and action.
Trends Neurosci. 23, 35–39.
Guillery, R.W., and Sherman, S.M. (2002). Thalamic relay functions and their role in
corticocortical communication: Generalizations from the visual system. Neuron 33, 163–175.
Hackett, T.A., Stepniewska, I., and Kaas, J.H. (1998). Thalamocortical connections of the
parabelt auditory cortex in macaque monkeys. J. Comp. Neurol. 400, 271–286.
Halassa, M.M., and Sherman, S.M. (2019). Thalamocortical Circuit Motifs: A General
Framework. Neuron 103, 762–770.
Harris, K.D., and Mrsic-Flogel, T.D. (2013). Cortical connectivity and sensory coding. Nature
503, 51–58.
Harris, K.D., and Shepherd, G.M.G. (2015). The neocortical circuit: themes and variations. Nat.
Neurosci. 18, 170–181.
Harting, J.K., Hall, W.C., and Diamond, I.T. (1972). Evolution of the Pulvinar (Part 1 of 2).
Brain. Behav. Evol. 6, 424–438.
Harting, J.K., Glendenning, K.K., Diamond, I.T., and Hall, W.C. (1973). Evolution of the
primate visual system: anterograde degeneration studies of the tecto-pulvinar system. Am. J.
Phys. Anthropol. 38, 383–392.
Herkenham, M. (1980). Laminar organization of thalamic projections to the rat neocortex.
Science 207, 532–535.
Hofbauer, A., and Dräger, U.C. (1985). Depth segregation of retinal ganglion cells projecting to
mouse superior colliculus. J. Comp. Neurol. 234, 465–474.
Hoy, J.L., Bishop, H.I., and Niell, C.M. (2019). Defined Cell Types in Superior Colliculus Make
Distinct Contributions to Prey Capture Behavior in the Mouse. Curr. Biol. 29, 4130-4138.e5.
Hugdahl, K., Wester, K., and Asbjørnsen, A. (1991). Auditory neglect after right frontal lobe and
right pulvinar thalamic lesions. Brain Lang. 41, 465–473.
Hupé, J.M., James, A.C., Payne, B.R., Lomber, S.G., Girard, P., and Bullier, J. (1998). Cortical
feedback improves discrimination between figure and background by V1, V2 and V3 neurons.
Nature 394, 784–787.
Ibrahim, L.A., Mesik, L., Ji, X. ying, Fang, Q., Li, H. fu, Li, Y. tang, Zingg, B., Zhang, L.I., and
Tao, H.W. (2016). Cross-Modality Sharpening of Visual Cortical Processing through Layer-1-
Mediated Inhibition and Disinhibition. Neuron 89, 1031–1045.
Ito, S., and Feldheim, D.A. (2018). The Mouse Superior Colliculus: An Emerging Model for
Studying Circuit Formation and Function. Front. Neural Circuits 12, 1–11.
103
Jaramillo, J., Mejias, J.F., and Wang, X.-J. (2018). Engagement of pulvino-cortical feedforward
and feedback pathways in cognitive computations.
Ji, W., Gămănuţ, R., Bista, P., D’Souza, R.D., Wang, Q., and Burkhalter, A. (2015). Modularity
in the Organization of Mouse Primary Visual Cortex. Neuron 87, 633–644.
Jiang, X., Wang, G., Lee, A.J., Stornetta, R.L., and Zhu, J.J. (2013). The organization of two new
cortical interneuronal circuits. Nat. Neurosci. 16, 210–218.
Jones, J.P., and Palmer, L.A. (1987). The two-dimensional spatial structure of simple receptive
fields in cat striate cortex. J. Neurophysiol. 58, 1187–1211.
Jouhanneau, J.S., Ferrarese, L., Estebanez, L., Audette, N.J., Brecht, M., Barth, A.L., and Poulet,
J.F.A. (2014). Cortical fos GFP expression reveals broad receptive field excitatory neurons
targeted by pom. Neuron 84, 1065–1078.
Juavinett, A.L., Kim, E.J., Collins, H.C., and Callaway, E.M. (2020). A systematic topographical
relationship between mouse lateral posterior thalamic neurons and their visual cortical projection
targets. J. Comp. Neurol. 528, 95–107.
Kaas, J.H., and Lyon, D.C. (2007). Pulvinar contributions to the dorsal and ventral streams of
visual processing in primates. Brain Res. Rev. 55, 285–296.
Kamishina, H., Conte, W.L., Patel, S.S., Tai, R.J., Corwin, J. V, and Reep, R.L. (2009). Cortical
connections of the rat lateral posterior thalamic nucleus. Brain Res. 1264, 39–56.
Karnath, H.O., Himmelbach, M., and Rorden, C. (2002). The subcortical anatomy of human
spatial neglect: putamen, caudate nucleus and pulvinar. Brain 125, 350–360.
Kastner, S., Fiebelkorn, I.C., and Eradath, M.K. (2020). Dynamic pulvino-cortical interactions in
the primate attention network. Curr. Opin. Neurobiol. 65, 10–19.
King, A.J., and Palmer, A.R. (1985). Integration of visual and auditory information in bimodal
neurones in the guinea-pig superior colliculus. Exp. Brain Res. 60, 492–500.
Kinoshita, M., Kato, R., Isa, K., Kobayashi, K., Kobayashi, K., Onoe, H., and Isa, T. (2019).
Dissecting the circuit for blindsight to reveal the critical role of pulvinar and superior colliculus.
Nat. Commun. 10, 135.
Kirchgessner, M.A., Franklin, A.D., and Callaway, E.M. (2020). Context-dependent and
dynamic functional influence of corticothalamic pathways to first- and higher-order visual
thalamus. Proc. Natl. Acad. Sci. U. S. A. 117, 13066–13077.
Komura, Y., Tamura, R., Uwano, T., Nishijo, H., Kaga, K., and Ono, T. (2001). Retrospective
and prospective coding for predicted reward in the sensory thalamus. Nature 412, 546–549.
Komura, Y., Nikkuni, A., Hirashima, N., Uetake, T., and Miyamoto, A. (2013). Responses of
pulvinar neurons reflect a subject’s confidence in visual categorization. Nat. Neurosci. 16, 749–
755.
Kremkow, J., and Alonso, J.-M. (2018). Thalamocortical Circuits and Functional Architecture.
Annu. Rev. Vis. Sci. 4, 263–285.
De La Mothe, L.A., Blumell, S., Kajikawa, Y., and Hackett, T.A. (2006). Thalamic connections
of the auditory cortex in marmoset monkeys: Core and medial belt regions. J. Comp. Neurol.
496, 72–96.
104
Lane, R.D., Allan, D.M., Bennett-Clarke, C.A., Howell, D.L., and Rhoades, R.W. (1997).
Projection status of calbindin- and parvalbumin-immunoreactive neurons in the superficial layers
of the rat’s superior colliculus. Vis. Neurosci. 14, 277–286.
Le, Q. Van, Isbell, L.A., Matsumoto, J., Le, V.Q., Nishimaru, H., Hori, E., Maior, R.S., Tomaz,
C., Ono, T., and Nishijo, H. (2016). Snakes elicit earlier, and monkey faces, later, gamma
oscillations in macaque pulvinar neurons. Sci. Rep. 6, 20595.
Van Le, Q., Isbell, L.A., Matsumoto, J., Nguyen, M., Hori, E., Maior, R.S., Tomaz, C., Tran,
A.H., Ono, T., and Nishijo, H. (2013). Pulvinar neurons reveal neurobiological evidence of past
selection for rapid detection of snakes. Proc. Natl. Acad. Sci. U. S. A. 110, 19000–19005.
Lee, C.C., and Sherman, S.M. (2008). Synaptic properties of thalamic and intracortical inputs to
layer 4 of the first- and higher-order cortical areas in the auditory and somatosensory systems. J.
Neurophysiol. 100, 317–326.
Lee, S., Carvell, G.E., and Simons, D.J. (2008). Motor modulation of afferent somatosensory
circuits. Nat. Neurosci. 11, 1430–1438.
Lee, S.H., Kwan, A.C., Zhang, S., Phoumthipphavong, V., Flannery, J.G., Masmanidis, S.C.,
Taniguchi, H., Huang, Z.J., Zhang, F., Boyden, E.S., et al. (2012). Activation of specific
interneurons improves V1 feature selectivity and visual perception. Nature 488, 379–383.
Li, L.-Y., Xiong, X.R., Ibrahim, L.A., Yuan, W., Tao, H.W., and Zhang, L.I. (2015). Differential
Receptive Field Properties of Parvalbumin and Somatostatin Inhibitory Neurons in Mouse
Auditory Cortex. Cereb. Cortex 25, 1782–1791.
Li, Y., Ma, W., Li, L., Ibrahim, L.A., Wang, S., and Tao, H.W. (2012). Broadening of inhibitory
tuning underlies contrast-dependent sharpening of orientation selectivity in mouse visual cortex.
J. Neurosci. 32, 16466–16477.
Liang, F., Bai, L., Tao, H.W., Zhang, L.I., and Xiao, Z. (2014). Thresholding of auditory cortical
representation by background noise. Front. Neural Circuits 8.
Liang, F., Xiong, X.R., Zingg, B., Ji, X., Zhang, L.I., and Tao, H.W. (2015). Sensory Cortical
Control of a Visually Induced Arrest Behavior via Corticotectal Projections. Neuron 86, 755–
767.
Liang, F., Li, H., Chou, X., Zhou, M., Zhang, N.K., Xiao, Z., Zhang, K.K., Tao, H.W., and
Zhang, L.I. (2018). Sparse Representation in Awake Auditory Cortex: Cell-type Dependence,
Synaptic Mechanisms, Developmental Emergence, and Modulation. Cereb. Cortex.
Linden, R., and Perry, V.H. (1983). Massive retinotectal projection in rats. Brain Res. 272, 145–
149.
Liu, B. hua, Li, Y. tang, Ma, W. pei, Pan, C. jie, Zhang, L.I., and Tao, H.W. (2011). Broad
inhibition sharpens orientation selectivity by expanding input dynamic range in mouse simple
cells. Neuron 71, 542–554.
Liu, B.H., Li, P., Sun, Y.J., Li, Y.T., Zhang, L.I., and Tao, H.W. (2010). Intervening inhibition
underlies simple-cell receptive field structure in visual cortex. Nat. Neurosci. 13, 89–96.
Macknik, S.L., and Livingstone, M.S. (1998). Neuronal correlates of visibility and invisibility in
the primate visual system. Nat. Neurosci. 1, 144–149.
105
Maeshima, S., and Osawa, A. (2018). Thalamic Lesions and Aphasia or Neglect. Curr. Neurol.
Neurosci. Rep. 18, 39.
Magariños-Ascone, C., Buño, W., and García-Austt, E. (1988). Monkey pulvinar units related to
motor activity and sensory response. Brain Res. 445, 30–38.
Mahn, M., Prigge, M., Ron, S., Levy, R., and Yizhar, O. (2016). Biophysical constraints of
optogenetic inhibition at presynaptic terminals. Nat. Neurosci. 19, 554–556.
Maor, I., Shalev, A., and Mizrahi, A. (2016). Distinct Spatiotemporal Response Properties of
Excitatory Versus Inhibitory Neurons in the Mouse Auditory Cortex. Cereb. Cortex 26, 4242–
4252.
Marques, T., Nguyen, J., Fioreze, G., and Petreanu, L. (2018). The functional organization of
cortical feedback inputs to primary visual cortex. Nat. Neurosci. 21.
Masterson, S.P., Li, J., and Bickford, M.E. (2009). Synaptic organization of the tectorecipient
zone of the rat lateral posterior nucleus. J. Comp. Neurol. 515, 647–663.
Masterson, S.P., Zhou, N., Akers, B.K., Dang, W., and Bickford, M.E. (2019). Ultrastructural
and optogenetic dissection of V1 corticotectal terminal synaptic properties. J. Comp. Neurol.
527, 833–842.
McFadyen, J., Mattingley, J.B., and Garrido, M.I. (2019). An afferent white matter pathway from
the pulvinar to the amygdala facilitates fear recognition. Elife 8.
Meredith, M.A., and Stein, B.E. (1986). Visual, auditory, and somatosensory convergence on
cells in superior colliculus results in multisensory integration. J. Neurophysiol. 56, 640–662.
Mesik, L., Huang, J.J., Zhang, L.I., and Tao, H.W. (2019). Sensory and motor-related responses
of layer 1 neurons in the mouse visual cortex. J. Neurosci. 1719–1722.
Mishkin, M., and Ungerleider, L.G. (1982). Contribution of striate inputs to the visuospatial
functions of parieto-preoccipital cortex in monkeys. Behav. Brain Res. 6, 57–77.
Moraga-Amaro, R., Cortés-Rojas, A., Simon, F., and Stehberg, J. (2014). Role of the insular
cortex in taste familiarity. Neurobiol. Learn. Mem. 109, 37–45.
Nakamura, H., Hioki, H., Furuta, T., and Kaneko, T. (2015). Different cortical projections from
three subdivisions of the rat lateral posterior thalamic nucleus: a single-neuron tracing study with
viral vectors. Eur. J. Neurosci. 41, 1294–1310.
Nassi, J.J., and Callaway, E.M. (2009). Parallel processing strategies of the primate visual
system. Nat. Rev. Neurosci. 10, 360–372.
Nassi, J.J., Lomber, S.G., and Born, R.T. (2013). Corticocortical feedback contributes to
surround suppression in V1 of the alert primate. J. Neurosci. 33, 8504–8517.
Niell, C.M., and Stryker, M.P. (2008). Highly selective receptive fields in mouse visual cortex. J.
Neurosci. 28, 7520–7536.
Nienborg, H., Hasenstaub, A., Nauhaus, I., Taniguchi, H., Huang, Z.J., and Callaway, E.M.
(2013). Contrast Dependence and Differential Contributions from Somatostatin- and
Parvalbumin-Expressing Neurons to Spatial Integration in Mouse V1. J. Neurosci. 33, 11145–
11154.
106
Nurminen, L., Merlin, S., Bijanzadeh, M., Federer, F., and Angelucci, A. (2018). Top-down
feedback controls spatial summation and response amplitude in primate visual cortex. Nat.
Commun. 9, 2281.
Ogino, T., and Ohtsuka, K. (2000). Effects of superior colliculus inhibition on visual motion
processing in the lateral suprasylvian visual area of the cat. Invest. Ophthalmol. Vis. Sci. 41,
955–960.
Oh, S.W., Harris, J.A., Ng, L., Winslow, B., Cain, N., Mihalas, S., Wang, Q., Lau, C., Kuan, L.,
Henry, A.M., et al. (2014). A mesoscale connectome of the mouse brain. Nature 508, 207–214.
Olavarria, J. (1979). A horseradish peroxidase study of the projections from the latero-posterior
nucleus to three lateral peristriate areas in the rat. Brain Res. 173, 137–141.
Padmala, S., Lim, S.-L., and Pessoa, L. (2010). Pulvinar and Affective Significance: Responses
Track Moment-to-Moment Stimulus Visibility. Front. Hum. Neurosci. 4.
Pardi, M.B., Vogenstahl, J., Dalmay, T., Spanò, T., Pu, D.-L., Naumann, L.B., Kretschmer, F.,
Sprekeler, H., and Letzkus, J.J. (2020). A thalamocortical top-down circuit for associative
memory. Science 370, 844–848.
Perry, V.H., and Cowey, A. (1984). Retinal ganglion cells that project to the superior colliculus
and pretectum in the macaque monkey. Neuroscience 12, 1125–1137.
Pessoa, L., and Adolphs, R. (2010). Emotion processing and the amygdala: from a “low road” to
“many roads” of evaluating biological significance. Nat. Rev. Neurosci. 11, 773–783.
Petersen, S.E., Robinson, D.L., and Morris, J.D. (1987). Contributions of the pulvinar to visual
spatial attention. Neuropsychologia 25, 97–105.
Petreanu, L., Mao, T., Sternson, S.M., and Svoboda, K. (2009). The subcellular organization of
neocortical excitatory connections. Nature 457, 1142–1145.
Pfeffer, C.K., Xue, M., He, M., Huang, Z.J., and Scanziani, M. (2013). Inhibition of inhibition in
visual cortex: The logic of connections between molecularly distinct interneurons. Nat. Neurosci.
16, 1068–1076.
Priebe, N.J., and Ferster, D. (2008). Inhibition, Spike Threshold, and Stimulus Selectivity in
Primary Visual Cortex. Neuron 57, 482–497.
Purushothaman, G., Marion, R., Li, K., and Casagrande, V.A. (2012). Gating and control of
primary visual cortex by pulvinar. Nat. Neurosci. 15, 905–912.
Redgrave, P., Westby, G.W., and Dean, P. (1993). Functional architecture of rodent superior
colliculus: relevance of multiple output channels. Prog. Brain Res. 95, 69–77.
Reinhold, K., Lien, A.D., and Scanziani, M. (2015). Distinct recurrent versus afferent dynamics
in cortical visual processing. Nat. Neurosci. 18, 1789–1797.
Rockland, K.S. (2019). Distinctive Spatial and Laminar Organization of Single Axons from
Lateral Pulvinar in the Macaque. Vis. (Basel, Switzerland) 4.
Roth, M.M., Dahmen, J.C., Muir, D.R., Imhof, F., Martini, F.J., and Hofer, S.B. (2016).
Thalamic nuclei convey diverse contextual information to layer 1 of visual cortex. Nat. Neurosci.
19, 299–307.
107
Rubio-Garrido, P., Pérez-de-Manzo, F., Porrero, C., Galazo, M.J., and Clascá, F. (2009).
Thalamic input to distal apical dendrites in neocortical layer 1 is massive and highly convergent.
Cereb. Cortex 19, 2380–2395.
Saalmann, Y.B., Pinsk, M.A., Wang, L., Li, X., and Kastner, S. (2012). The pulvinar regulates
information transmission between cortical areas based on attention demands. Science 337, 753–
756.
Scholl, L.R., Foik, A.T., and Lyon, D.C. (2020). Projections between visual cortex and pulvinar
in the rat. J. Comp. Neurol.
Schumacher, J.W., Schneider, D.M., and Woolley, S.M.N. (2011). Anesthetic state modulates
excitability but not spectral tuning or neural discrimination in single auditory midbrain neurons.
J. Neurophysiol. 106, 500–514.
Schuman, B., Machold, R.P., Hashikawa, Y., Fuzik, J., Fishell, G.J., and Rudy, B. (2019). Four
Unique Interneuron Populations Reside in Neocortical Layer 1. J. Neurosci. 39, 125–139.
Shang, C., Liu, Z., Chen, Z., Shi, Y., Wang, Q., Liu, S., Li, D., and Cao, P. (2015). BRAIN
CIRCUITS. A parvalbumin-positive excitatory visual pathway to trigger fear responses in mice.
Science 348, 1472–1477.
Sherman, S.M. (2016). Thalamus plays a central role in ongoing cortical functioning. Nat.
Neurosci. 19, 533–541.
Sherman, S.M., and Guillery, R.W. (1998). On the actions that one nerve cell can have on
another: distinguishing “drivers” from “modulators”. Proc. Natl. Acad. Sci. U. S. A. 95, 7121–
7126.
Sherman, S.M., and Guillery, R.W. (2002). The role of the thalamus in the flow of information to
the cortex. Philos. Trans. R. Soc. Lond. B. Biol. Sci. 357, 1695–1708.
Shipp, S. (2001). Corticopulvinar connections of areas V5, V4, and V3 in the macaque monkey:
a dual model of retinal and cortical topographies. J. Comp. Neurol. 439, 469–490.
Shipp, S. (2007). Structure and function of the cerebral cortex. Curr. Biol. 17, R443–R449.
Soares, S.C., Maior, R.S., Isbell, L.A., Tomaz, C., and Nishijo, H. (2017). Fast Detector/First
Responder: Interactions between the Superior Colliculus-Pulvinar Pathway and Stimuli Relevant
to Primates. Front. Neurosci. 11, 67.
de Souza, B.O.F., Frigon, É.-M., Tremblay-Laliberté, R., Casanova, C., and Boire, D. (2020).
Laminar distribution of cortical projection neurons to the pulvinar: A comparative study in cats
and mice. J. Comp. Neurol.
Stepniewska, I., Ql, H.X., and Kaas, J.H. (2000). Projections of the superior colliculus to
subdivisions of the inferior pulvinar in New World and Old World monkeys. Vis. Neurosci. 17,
529–549.
Stitt, I., Zhou, Z.C., Radtke-Schuller, S., and Fröhlich, F. (2018). Arousal dependent modulation
of thalamo-cortical functional interaction. Nat. Commun. 9, 2455.
Sugawara, E., and Nikaido, H. (2014). Properties of AdeABC and AdeIJK efflux systems of
Acinetobacter baumannii compared with those of the AcrAB-TolC system of Escherichia coli.
Antimicrob. Agents Chemother. 58, 7250–7257.
108
Sutter, M.L., and Schreiner, C.E. (1991). Physiology and topography of neurons with
multipeaked tuning curves in cat primary auditory cortex. J. Neurophysiol. 65, 1207–1226.
Takahashi, T. (1985). The organization of the lateral thalamus of the hooded rat. J. Comp.
Neurol. 231, 281–309.
Tohmi, M., Meguro, R., Tsukano, H., Hishida, R., and Shibuki, K. (2014). The extrageniculate
visual pathway generates distinct response properties in the higher visual areas of mice. Curr.
Biol. 24, 587–597.
Trojanowski, J.O., and Jacobson, S. (1975). Peroxidase labeled subcortical afferents to pulvinar
in rhesus monkey. Brain Res. 97, 144–150.
Ungerleider, L.G., Galkin, T.W., Desimone, R., and Gattass, R. (2014). Subcortical Projections
of Area V2 in the Macaque. J. Cogn. Neurosci. 26, 1220–1233.
Vaiceliunaite, A., Erisken, S., Franzen, F., Katzner, S., and Busse, L. (2013). Spatial integration
in mouse primary visual cortex. J. Neurophysiol. 110, 964–972.
Vertes, R.P., Linley, S.B., and Hoover, W.B. (2015). Limbic circuitry of the midline thalamus.
Neurosci. Biobehav. Rev. 54, 89–107.
Wang, B.A., and Pleger, B. (2020). Confidence in Decision-Making during Probabilistic Tactile
Learning Related to Distinct Thalamo-Prefrontal Pathways. Cereb. Cortex 30, 4677–4688.
Wang, Q., Gao, E., and Burkhalter, A. (2011). Gateways of ventral and dorsal streams in mouse
visual cortex. J. Neurosci. 31, 1905–1918.
Ward, R., Danziger, S., and Bamford, S. (2005). Response to visual threat following damage to
the pulvinar. Curr. Biol. 15, 571–573.
Wei, P., Liu, N., Zhang, Z., Liu, X., Tang, Y., He, X., Wu, B., Zhou, Z., Liu, Y., Li, J., et al.
(2015). Processing of visually evoked innate fear by a non-canonical thalamic pathway. Nat.
Commun. 6, 6756.
Wilke, M., Mueller, K.-M., and Leopold, D.A. (2009). Neural activity in the visual thalamus
reflects perceptual suppression. Proc. Natl. Acad. Sci. U. S. A. 106, 9465–9470.
Wilke, M., Turchi, J., Smith, K., Mishkin, M., and Leopold, D.A. (2010). Pulvinar inactivation
disrupts selection of movement plans. J. Neurosci. 30, 8650–8659.
Wilke, M., Kagan, I., and Andersen, R.A. (2013). Effects of pulvinar inactivation on spatial
decision-making between equal and asymmetric reward options. J. Cogn. Neurosci. 25, 1270–
1283.
Wilke, M., Schneider, L., Dominguez-Vargas, A.-U., Schmidt-Samoa, C., Miloserdov, K.,
Nazzal, A., Dechent, P., Cabral-Calderin, Y., Scherberger, H., Kagan, I., et al. (2018). Reach and
grasp deficits following damage to the dorsal pulvinar. Cortex. 99, 135–149.
Winer, J.A., Miller, L.M., Lee, C.C., and Schreiner, C.E. (2005). Auditory thalamocortical
transformation: structure and function. Trends Neurosci. 28, 255–263.
Wise, L.Z., and Irvine, D.R. (1983). Auditory response properties of neurons in deep layers of
cat superior colliculus. J. Neurophysiol. 49, 674–685.
109
Wong, P., Collins, C.E., Baldwin, M.K.L., and Kaas, J.H. (2009). Cortical connections of the
visual pulvinar complex in prosimian galagos ( Otolemur garnetti ). J. Comp. Neurol. 517, 493–
511.
Woody, C., Melamed, O., and Chizhevsky, V. (1991). Patterns of activity coding discrimination
of auditory stimuli differ between mid- and posterolateral thalamus of cats. J. Neurosci. 11,
3379–3387.
Xiong, X.R., Liang, F., Li, H., Mesik, L., Zhang, K.K., Polley, D.B., Tao, H.W., Xiao, Z., and
Zhang, L.I. (2013). Interaural level difference-dependent gain control and synaptic scaling
underlying binaural computation. Neuron 79, 738–753.
Xiong, X.R., Liang, F., Zingg, B., Ji, X.Y., Ibrahim, L.A., Tao, H.W., and Zhang, L.I. (2015).
Auditory cortex controls sound-driven innate defense behaviour through corticofugal projections
to inferior colliculus. Nat. Commun. 6, 1–12.
Yang, F.-C., and Burwell, R.D. (2020). Neuronal Activity in the Rat Pulvinar Correlates with
Multiple Higher-Order Cognitive Functions. Vis. (Basel, Switzerland) 4.
Yilmaz, M., and Meister, M. (2013). Rapid innate defensive responses of mice to looming visual
stimuli. Curr. Biol. 23, 2011–2015.
Yirmiya, R., and Hocherman, S. (1987). Auditory- and movement-related neural activity interact
in the pulvinar of the behaving rhesus monkey. Brain Res. 402, 93–102.
Zhang, W., and Bruno, R.M. (2019). High-order thalamic inputs to primary somatosensory
cortex are stronger and longer lasting than cortical inputs. Elife 8.
Zhang, G.-W., Sun, W.-J., Zingg, B., Shen, L., He, J., Xiong, Y., Tao, H.W., and Zhang, L.I.
(2018a). A Non-canonical Reticular-Limbic Central Auditory Pathway via Medial Septum
Contributes to Fear Conditioning. Neuron 97, 406-417.e4.
Zhang, G.-W., Shen, L., Zhong, W., Xiong, Y., Zhang, L.I., and Tao, H.W. (2018b).
Transforming Sensory Cues into Aversive Emotion via Septal-Habenular Pathway. Neuron 99,
1016-1028.e5.
Zhang, S., Xu, M., Kamigaki, T., Hoang Do, J.P., Chang, W.-C., Jenvay, S., Miyamichi, K., Luo,
L., and Dan, Y. (2014). Selective attention. Long-range and local circuits for top-down
modulation of visual cortex processing. Science 345, 660–665.
Zhao, X., Liu, M., and Cang, J. (2014). Visual cortex modulates the magnitude but not the
selectivity of looming-evoked responses in the superior colliculus of awake mice. Neuron 84,
202–213.
Zhou, H., Schafer, R.J., and Desimone, R. (2016). Pulvinar-Cortex Interactions in Vision and
Attention. Neuron 89, 209–220.
Zhou, M., Liang, F., Xiong, X.R., Li, L., Li, H., Xiao, Z., Tao, H.W., and Zhang, L.I. (2014).
Scaling down of balanced excitation and inhibition by active behavioral states in auditory cortex.
Nat. Neurosci. 17, 841–850.
Zhou, N., Masterson, S.P., Damron, J.K., Guido, W., and Bickford, M.E. (2018). The Mouse
Pulvinar Nucleus Links the Lateral Extrastriate Cortex, Striatum, and Amygdala. J. Neurosci. 38,
347–362.
110
Zhou, N.A., Maire, P.S., Masterson, S.P., and Bickford, M.E. (2017). The mouse pulvinar
nucleus: Organization of the tectorecipient zones. Vis. Neurosci. 34, E011.
Zihl, J., and von Cramon, D. (1979). The contribution of the “second” visual system to directed
visual attention in man. Brain 102, 835–856.
Zingg, B., Chou, X.-L., Zhang, Z.-G., Mesik, L., Liang, F., Tao, H.W., and Zhang, L.I. (2017).
AAV-Mediated Anterograde Transsynaptic Tagging: Mapping Corticocollicular Input-Defined
Neural Pathways for Defense Behaviors. Neuron 93, 33–47.

Abstract (if available)

Abstract We interact with the external world using senses. Sensory processing is essential for our survival. Encoding and decoding of sensory information allows us to understand the external world and take actions to it when necessary. External sensory stimuli are collected by our sensory organs and sent to the brain. These ever-changing signals are transduced and encoded to establish an internal, dynamic and optimized representation of the external world, which is further shaped by internal brain state, previous experience, and learning and memory, and sometimes distorted under illusory or pathological conditions. This internal modelling of external world is then decoded to serve as the basis to execute higher cognitive function, such as conceptualization and decision making, and to guide our behaviors and actions. After decades of inter-disciplinary investigations and tremendous efforts, we begin to elucidate the strategies of encoding and decoding sensory stimuli in context or noisy background. Most of the sensory information sent to the cortex is routed via the thalamus first. While a small portion of the thalamus (first-order thalamic nuclei) are dedicated to faithfully convey sensory the information to the primary sensory cortex, the vast majority of the thalamus are higher-order nuclei whose functionality is poorly understood. Here I will present two studies on the most prominent higher-order thalamic nucleus, the pulvinar nucleus (or the lateral posterior nucleus in rodents), trying to uncover its function in contextual and cross-modal sensory processing. ❧ In the first study, we investigated the modulation of visual processing via the pulvinar nucleus. In the mammalian visual system, information from the retina streams into parallel bottom-up pathways. It remains unclear how these pathways interact to contribute to contextual modulation of visual cortical processing. By optogenetic inactivation and activation of mouse lateral posterior nucleus (LP) of thalamus, a homolog of pulvinar, or its projection to primary visual cortex (V1), we found that LP contributes to surround suppression of layer (L) 2/3 responses in V1 by driving L1 inhibitory neurons. This results in subtractive suppression of visual responses and an overall enhancement of orientation, direction, spatial, and size selectivity. Neurons in V1-projecting LP regions receive bottom-up input from the superior colliculus (SC) and respond preferably to non-patterned visual noise. The noise-dependent LP activity allows V1 to “cancel” noise effects and maintain its orientation selectivity under varying noise background. Thus, the retina-SC-LP-V1 pathway forms a differential circuit with the canonical retino-geniculate pathway to achieve context-dependent sharpening of visual representations. ❧ In the second study, we explored whether pulvinar can modulate auditory processing in a similar manner and examined its role in the cross-modal integration. Indeed, bidirectional activity modulations of LP or its projection to the primary auditory cortex (A1) in awake mice reveal that LP improves auditory processing in A1 supragranular-layer neurons by sharpening their receptive fields and frequency tuning, as well as increasing the signal-to-noise ratio (SNR). This is achieved through a subtractive-suppression mechanism, mediated largely by LP-to-A1 axons preferentially innervating specific inhibitory neurons in layer 1 and superficial layers. LP is strongly activated by specific sensory signals relayed from the SC, contributing to the maintenance and enhancement of A1 processing in the presence of auditory background noise and threatening visual looming stimuli, respectively. Thus, a multisensory bottom-up SC-pulvinar-A1 pathway plays a role in contextual and cross-modality modulation of auditory cortical processing. ❧ These two studies together suggest that the pulvinar nucleus modulates cortical sensory processing of various modalities and enhances the saliency of sensory stimuli in a noisy background or a multisensory confounding context. They will deepen our understanding of higher-order thalamic nuclei in sensory processing and shed light on future studies to explore how the pulvinar nucleus corroborate with other higher-order thalamic nuclei and cortical regions to optimize our sensory perception and decision making.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Neural circuits underlying the modulation and impact of defensive behaviors

PDF

Neural circuits control and modulate innate defensive behaviors

PDF

Functional properties of the superficial cortical interneurons

PDF

From sensory processing to behavior control: functions of the inferior colliculus

PDF

Synaptic mechanism underlying development and function of neural circuits in rat primary auditory cortex

PDF

Exploring sensory responses in the different subdivisions of the visual thalamus

PDF

Functional synaptic circuits in primary visual cortex

PDF

Synaptic circuits for information processing along the central auditory pathway

PDF

Spatial and temporal precision of inhibitory and excitatory neurons in the murine dorsal lateral geniculate nucleus

PDF

Subnetwork organization of the superior colliculus and visual system in the mouse brain

Asset Metadata

Creator Fang, Qi (author)

Core Title Contextual modulation of sensory processing via the pulvinar nucleus

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Neuroscience

Publication Date 02/01/2021

Defense Date 12/18/2020

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

Tag auditory cortex,bottom-up visual pathway,contextual modulation,cross-modality modulation,higher-order thalamus,in vivo electrophysiology,inhibitory neuron,mouse,OAI-PMH Harvest,optogenetics,orientation/direction tuning,primary visual cortex,pulvinar,receptive field,size tuning,superior colliculus,surround suppression

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Chen, Jeannie (committee chair), Hires, Samule Andrew (committee member), Tao, Huizhong (committee member), Zhang, Li (committee member)

Creator Email qfang@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-419104

Unique identifier UC11666689

Identifier etd-FangQi-9257.pdf (filename),usctheses-c89-419104 (legacy record id)

Legacy Identifier etd-FangQi-9257.pdf

Dmrecord 419104

Document Type Dissertation

Rights Fang, Qi

Type texts

Source 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 a...

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