University of Southern California Dissertations and Theses

Habituation Of The Multiple Unit Discharge Response To White Noise Stimulation In The Unanesthetized Rabbit

(USC Thesis Other)

Habituation Of The Multiple Unit Discharge Response To White Noise Stimulation In The Unanesthetized Rabbit

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content This dissertation has been
microfilmed exactly as received
70-5233
WILLISTON, John Stoddard, 1935-
HABITUATION OF THE MULTIPLE UNIT DISCHARGE
RESPONSE TO WHITE NOISE STIMULATION IN THE
UNANESTHETIZED RABBIT.
University of Southern California, Ph.D., 1969
Psychology, experimental
University Microfilms, Inc., Ann Arbor, Michigan
HABITUATION OP THE MULTIPLE UNIT DISCHARGE
RESPONSE TO WHITE NOISE STIMULATION
IN THE UNANESTHETIZED RABBIT
by
John Stoddard Williston
A Dissertation Presented to the
FACULTY OP THE GRADUATE SCHOOL
UNIVERSITY OP SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OP PHILOSOPHY
(Psychology)
June 1969
UNIVERSITY O F SOUTHERN CALIFORNIA
THE GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES, CALIFORNIA 9 0 0 0 7
This dissertation, written by
John Stoddard WLlliston
under the direction of his.... Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by The Gradu
ate School, in partial fulfillment of require
ments for the degree of
D O C T O R OF P H I L O S O P H Y
Dean
Date.
JUNE 1969
DISSERTATION COMMITTEE
Chairman
t .
ACKNOWLEDGMENTS
Many people have played a part in the writing of
this dissertation, many more than can reasonably be in
cluded in a short section. It is a pleasure, however, to
express my gratitude to those who have had an especially
important role.
Many of the skills and insights leading to this
dissertation were learned at San Francisco State College
under the tutelage of, among others, Drs. Joseph J.
Fortier, Walter J. Coppock, Jr., and Lewis F. Petrinovich.
These psychologist-scientists provided a frame of
reference and sense of direction which was to persist to
the present.
At the University of Southern California, Dr.
Everett J. Wyers became immediate supervisor, guidance
committee chairman, collaborator, close friend, and source
of inspiration. It is difficult, indeed, to adequately
express the feeling of gratitude felt toward him for the
time, patience and understanding which he devoted to his
students. Another very strong influence was Dr. William
W. Grings who was always available when needed with calm,
incisive intelligence and a highly developed sense of
empathy. It has been a privilege to be his student, both
ti
in the classroom and outside of it. His service on my
guidance and dissertation committees has been greatly
valued. Dr. John P. Meehan, Jr. of the Physiology Depart
ment was the outside member of the dissertation committee
and made a number of helpful suggestions which strengthened
and sharpened this study.
The chairman of the dissertation committee, Dr.
Gary 0. Galbraith, was involved in this work in many ways.
Both multiple unit recording technique and its application
to the question of the influence of the nonspecific systems
on classical sensory pathways were learned from him, as
was much of the methodology necessary for the hybrid
computer analysis. In addition, he was in large part
responsible for whatever coherence emerged in the final
draft of this study. My debt of gratitude to Dr. Gal
braith, of course, goes far beyond these few examples.
His influence may be judged by the extent of the change
and direction of my research interests and activities.
Three organizations participated in the support of
this study. Part of the work was completed while the
author was a National Institute of Mental Health Pre-
doctoral Fellow. This award was made under Grant #MH
10554 and included materials used in the study. The work
was finished while the writer was a Research Assistant
under the supervision of Dr. Gary 0. Galbraith. Support
lii
for this position was received from National Aeronautical
and Space Administration Grant #NGR 05-018-044. Computa
tion time on the hybrid computer and much additional
assistance was received from the Systems Simulation
Laboratory of the Engineering School at the University of
Southern California.
The author would also like to thank Mrs. June
Brown for a very fine job of final typing of the manu
script. Her careful attention to detail and willingness
to make the extra effort has been appreciated.
Of the many who made this dissertation possible,
my wife Henni must receive the deepest appreciation. It
is no easy matter to live with the anxiety and frustra
tions associated with another's academic career, but she
did so for many years. Her daily activities included
much more than most wives devote to their marriages, and
included many hours of typing, editing, and related work.
This debt can never be repaid but can at least be ac
knowledged.
TABLE OP CONTENTS
Page
ACKNOWLEDGMENTS ............................... ii
LIST OP TABLES............................... vii
LIST OP FIGURES............................... viii
INTRODUCTION ................................. 1
Chapter
I. HABITUATION ............................. 3
Definitions.............. 3
Phylogenetic considerations .......... 6
Higher-order reflexes: the orienting
response ........................... 7
Neurological models of habituation . . 9
Neuronal mechanisms of habituation . . 12
II. ELECTRICAL RESPONSES IN THE BRAIN .... 17
D. C. potentials..................... 17
EEG...................... 18
Single unit recording................ 19
Multiple unit recording.............. 21
Problems in interpretation .......... 21
Multiple unit research .............. 24
Effects of stimulation.............. 27
III. HABITUATION OP EVOKED NEURAL ACTIVITY . . 40
Habituation of evoked potentials . . . 41
EEG arousal response ................ 44
The effects of ablation.............. 46
Unit activity and habituation........ 47
IV. METHODS................................. 53
Subjects.......................... 53
Surgei'y ............................. 53
Electrophysiological data acquisition . 54
Experimental procedure ........ . . . 57
: ■ v
i Chapter
Data analysis ........................
V. RESULTS .................................
Evoked multiple unit activity ........
Habituation of the evoked multiple
unit response ......................
Intertrial interval effects ..........
Stimulus intensity manipulations . . .
Dishabituation ......................
VI. DISCUSSION .............................
VII. SUMMARY.................................
LIST OP REFERENCES ...........................
LIST OF TABLES
Table
1. Mean root-mean-square voltage levels
recorded from histologically verified
electrode placements used in study . . .
Page
68
LIST OF FIGURES
Data Acquisition System ..............
The Multiple Unit Response to White
Noise Stimulation .................. «
Data Processing Hybrid System ........
The Multiple Unit Evoked Response as a
Function of Stimulus Intensity . . . .
Averaged Transient Analysis of the
Integrated Multiple Unit Response to
White Noise Stimulation ..............
Group Evoked Unit Response Means Over
Trials During the First and Second Days
of Stimulation: Major Groups ........
Group Evoked Unit Response Means Over
Trials During the First and Second Days
of Stimulation: Small Groups .....
Effects of Intertrial Interval on
Evoked Unit Activity ................
Effects of Stimulus Intensity on Evoked
Unit Activity ........................
Changes in Group Evoked Unit Activity
Following Dishabituation ............
1
INTRODUCTION
The search for the neural basis of learning has
attracted a large share of the research energies of a
number of different areas of specialization. Approaches to
the problem have taken many forms from direct mechanical
manipulation of the brain to the most minute analysis of
the cell membrane. One of the most popular techniques has
been the observation of neuroelectric activity during the j
i
process of learning. In spite of its extensive use and j
potential promise, however, this approach has had limited !
success due to the basic uncertainties regarding the
nature of the electrical response itself.
Several recent developments have simplified matters
somewhat by providing a new kind of neuroelectric response
J
to study: the massed activity of populations of neurons. j
In the present study, this technique has been used to i
I !
J i
examine the simplest form of learning, habituation. 1
j ;
Recently, Thompson and Spencer (1966) have proposed that
habituation can be defined by nine characteristic para-
! I
meters which occur in both the global behavior of the
intact organism and the monosynaptic reflex arc in the
spinal cat. The experiments described here were designed
to examine these parameters in the neuronal spike discharge
response of intact rabbit brain during habituation to the
presentation of white noise stimulation. Because of
2
certain theoretical models advanced hy Thompson and
Spencer and others, it was expected that the neuronal
spike discharge response of a number of areas in the brain
would exhibit patterns of response during habituation
which conform to the nine parameters of response habitua
tion.
3
CHAPTER I
I
HABITUATION
This chapter will he concerned with the development
of the definition and characteristics of habituation as a
j . i
i general introduction to the concept and its scope of ap- j
I i
|plication. The primary emphasis is upon the habituation |
!of peripheral responses. The relationship between these
!
i :
|peripheral events and those occurring within the central ;
i i
|nervous system will be touched upon in this chapter and
|developed at greater length in Chapter III. |
S I
] ,
Definitions !
i ' 1 1
t |
| Habituation has been operationally defined by
; i
:Harris (19^3) as “response decrement as a result of re-
i ;
jpeated stimulation" (p. 385)* This definition bears some
resemblance to the definition of learning offered by
Kimble (1961, p. 2): “learning refers to a more or less
| permanent change in behavior which occurs as a result of
! practice." Both authors attempt to distinguish their sub-
i :
i
iject from processes such as maturation, aging, trauma,
i 1
sensory adaptation, or effector fatigue. Another j
characteristic common to both may be reversibility of
response change. A learned response is usually subject
to weakening due to extinction, and a habituated response j
jto strengthening by spontaneous recovery. These and other j
similarities prompted Thorpe (1956) to suggest that
habituation is a form of learning. It differs from most
other learning phenomena in that it involves response de- j
Icrement rather than response increment, but it may resemble
i
!some types of learning, such as passive avoidance and dis
crimination. Thompson and Spencer (1966) have included j
i |
■under the general definition of habituation a variety of j
i
'other terms, including adaptation, accommodation, fatigue j
I(but not at the peripheral effector), inhibition, negative i
i ;
learning, extinction, stimulus satiation, etc., which ap- j
|pear generally to be operationally identified with each
other. One utility of this identification could be an
emphasis on the diverseness of the influence of what may
be a unitary process.
j ;
j Of the various definitions of habituation, the most
i
jcomplete operational definition was proposed in the Thomp- j
json and Spencer article (1966). After a review of the
|habituation literature, they concluded that nine opera-
jtions appear to characterize the habituation of virtually
i :
all response systems. Most studies have not used all nine
procedures, but each of them have been demonstrated with
a wide variety of responses. These defining operations,
as enumerated by Thompson and Spencer, are:
(1) Given that a particular stimulus elicits a i
response, repeated applications of the stimulus ;
result in decreased response (habituation). j
The decrease is usually a negative exponential j
function of the number of stimulus
5
presentations.
(2) If a stimulus is withheld, the response tends
to recover over time (spontaneous recovery).
(3) If repeated series of habituation training and
spontaneous recovery are given, habituation
i becomes successively more rapid (this might be
called potentiation of habituation).
(4) Other things being equal, the more rapid the
frequency of stimulation, the more rapid and/orj
| more pronounced is habituation.
i
(5) The weaker the stimulus, the more rapid and/or !
j pronounced is habituation. Strong stimuli may
yield no significant habituation.
(6) The effects of habituation training may pro
ceed beyond the zero or asymptotic level.
Thus, habituation may proceed beyond the point
of response cessation.
j (7) Habituation of response to a given stimulus
| exhibits stimulus generalization to other
j stimuli.
(8) Presentation of another (usually strong)
stimulus results in recovery of the habituated
response (dishabituation).
j
(9) Upon repeated application of the dishabitua-
tory stimulus, the amount of diBhabituation
| produced habituates (this Mght be called
habituation of dishabituation).
! Thompson and Spencer (1966) proposed that any
I '
'response change which exhibited these parameters be
i :
classified as habituation. Thus, even if these parameters
were found in the behavior of only part of a nervous j
i :
jsystem, it could still be claimed that habituation had,
! |
j in fact, occurred. In their work they have examined the j
habituation of simple reflexes in the acute spinal cat. |
|Because this preparation does conform to the nine basic |
parameters, It was concluded that habituation was demon
strated and, further, that the neural mechanism Involved
might be essentially the same one at work in intact,
ihigher-order animals. The mechanism proposed involves the
!
|action of interneurons, which are present at all levels of
jthe central nervous system, upon effector cells. It can,
!therefore, be persuasively argued that at least the basic
jmechanism of habituation might be the same for both spinal
|reflex and the more sophisticated mammalian cortex. The
jimplications of this and other neural theories of habitua-
jtion will be considered later.
i
Phylogenetic Considerations
j One outstanding characteristic of habituation is
|its occurrence at all phylogenetic levels (see Thorpe,
j
11956, for review). Even at the level of the single cell
i
jhabituation has been demonstrated which conforms in at
I least several respects to habituation seen at higher levels
! Jennings (1905)» for example, obtained a decrement in the
| contraction response of the Stentor and Vortlcella to
itactual stimulation which could not be explained by ef-
jfector fatigue because of the demonstration of dis-
ihabituation. Since then, further work (see Danisch, 1921)
has indicated that the rate of habituation is a function
of stimulus intensity and that spontaneous recovery occurs
at intervals comparable to some response systems in more
advanced phyla. It would be difficult to ascribe this
i
kind of habituation to the action of Interneurons because
|the entire animal is composed of only one cell. At the
j j
same time, it might be possible that some basic membrane j
;process is common to both Stentor and mammals which is in- j
jvolved in the production of response decrement. |
As one ascends the phylogenetic scale, longer j
i t
|retention and greater discriminative ability is seen, j
|along with larger response repertoires, but many of the
|basic parameters of habituation remain the same. The fact |
I J
|that animals with radically differing nervous system
I organization can demonstrate the same parameters of
; l
; habituation behavior supports an explanatory mechanism
I I
‘based upon a common neuronal mechanism. This is true be-
|cause vertebrate and invertebrate nervous systems are
!quite different in gross organization but are built with
cells which are quite similar (Herrick, 1962).
Hlgher-order Reflexes: The Orienting Response
I Nearly all reflexes habituate. (See Harris, 194-3,
j :
!for review.) Two notable exceptions are those reflexes
jassociated with antigravity function and the prevention of |
I injury. The complexity of these reactions can range from
ia simple monosynaptic reflex to the quite complex patterned:
reflex-like response to stimulus novelty termed the orient
ing response (OR). This response was described by Pavlov
1 :
(1927) as the "investigatory” or "what is it" reaction.____ j
Its components, according to Lynn (1966), include a number
of autonomic and skeletal responses. In response to a
'novel stimulus, for example, an animal might turn his head
!in the direction of the stimulus, approach the source of
|this disturbance, and sniff It. Autonomic responses might
j
include heart rate, respiratory, vasomotor, and galvanic
I skin response changes. In addition, an electroencephalo-
j graphic arousal pattern might be seen along with decreases
jin sensory thresholds. The OR contains many components
|of reflex-like activity which are subject to habituation.
j
;Not all of these responses habituate at the same rate.
: For example, skeletal movements may disappear in one or
|two presentations of the stimulus, while other responses,
like vasomotor reactions, may continue to be evoked for
imany more. A number of ingenious experiments with the OR
i
jhave demonstrated the sensitivity of this response to
changes in the stimulus. Sokolov (1963a) has shown that
;the OR can be evoked by a decrease in the intensity of the
'habituated stimulus. In addition, if a stimulus of a
|certain duration is repeatedly presented until habituation
I occurs and is then lengthened In presentation time, an OR
!occurs at the point where the stimulus would normally end.
jEven a number presented out of sequence can trigger an OR
when habituation has occurred to numbers presented one at
a time in sequence (Unger, 1964). The ability to respond
to this kind of subtlety must surely involve cortical
9
mechanisms, particularly when semantic discrimination is
used.
i
i
I Neurological Models of Habituation
j
| One school of thought holds that stimulus mediation
iis accomplished at cortical levels entirely (Sokolov,
1963a). According to this theory, an incoming sensory
I impression leaves traces of itself which persist at the j
! cortex for some time. If the same stimulus is repeated
!again, the trace accumulates and the incoming signal is
!also compared with previously established traces or
j i
I"models." If the stimulus does not resemble any model in
j the cortex, the cortex does not respond. This leaves the
|midbrain reticular formation free to respond to the
I stimulus reaching it via the collateral pathways and
j initiate an OR. If the stimulus does resemble a cortical
j ;
!model, then impulses are generated in the cortex which ' •
1 travel to the reticular area and block the collateral
i sensory system. Since the transmission speed of this
|
j collateral system is relatively slow, it is possible for
1 the fibers from the cortex to block the collaterals before
they can transmit information to reticular cells. The
process of OR habituation, according to this view, con
sists of an active presynaptlc inhibitory process exerted
by one neurological structure upon another. A two-stage,
active, inhibitory process appears demanded by the results
from tbe OR research which posed too many problems for the
10
original, one-process theory proposed by Pavlov (1927)*
,
He believed that habituation was an expression of cortical
fatigue or internal inhibition generated by stimulus
i !
1 !
•repetition. This theory cannot explain a number of find- !
| |
lings. It does not, for example, appear to be able to ex-
|
I plain why an OR is produced by a reduction in the in
tensity of a habituated stimulus. If cortical fatigue
i j
(were responsible for response deficit, then a decrease in j
!
|stimulus intensity would not be expected to produce an
|increased reaction in the fatigued cells. The same ob-
i
j Section is involved with the increase in stimulus duration
i . i
iexperiments. In this case no additional energy is supplied
by increasing the presentation time of the stimulus, but
!
|an OR is still obtained. Other problems include the fact
i ,
jthat an evoked potential continues to be seen in the
!presumably fatigued cortex after completion of habituation
land the fact that continued presentation of a habituated
I stimulus will often lead to the reappearance of the OR.
jIf the cortical cells are, in fact, fatigued, it is dif-
|ficult to understand why they still can respond to a
I stimulus with an evoked response. Finally, even greater
fatigue produced by continued stimulus presentation beyond
the point of OR evocation should not lead to the recovery
of the OR. This last observation has been reported by many j
j I
(investigators (Lynn, 1966). These problems would appear j
! .
I to rule out simple cortical fatigue as the mechanism of OR I
habituation.
A number of other theories have been advanced to
j
explain the neural mechanism of habituation of the OR.
iLynn (1966) has critically reviewed these proposals and
! concluded that, while all suffer from certain problems,
i j
;the cortical model theory proposed by Sokolov and presented !
! above is the most promising. Even this model, however, is
; embarrassed by the demonstration of habituation of orienta- j
jtion in the partially decorticate cat by Sharpless and |
|Jasper (1956) and Thompson and Welker (1963). The fact
| that OR decrement occurs despite the absence of cortex
!
which is thought to be involved in the mechanism producing
jthe decrement implies that the mechanism can also function
I in other cortical or subcortical areas. Sharpless and
j :
jJasper (1956) suggested that habituation of the OR, or
’ ’arousal response," might occur at a number of levels and
i
that the level of complexity which could be selectively
habituated was dependent upon the neurological mechanisms
available for mediation of the stimulus. If an animal is
to be habituated to one pattern of auditory stimulation
land still give an OR to a second, dissimilar pattern, then
! !
|the presence of the auditory cortex appears to be neces- j
! I
sary. In the absence of this cortical tissue, the re-
i
mainder of the brain can still discriminate between tones
I
of different frequencies. If the auditory input was |
j
Interrupted at the midbrain level by section at the |
12
brachium of the inferior colliculus, only modality-specific
habituation was possible in this study. Following
i
|habituation to an auditory stimulus, an OR was only seen j
i ;
!to stimuli in another sense modality.
| One resolution of this conflict between theories
! might lie in the careful evaluation of the receptive and
iefferent capacities of the subject under consideration.
Jasper (1964) has emphasized the importance of tonic
j
control of lower centers by higher ones. The structural
mechanism of habituation in the intact animal may not be
the same as that in the lesioned one. Removal of cortex,
I for example, may eliminate some form of tonic regulation
and permit other processes to occur which might otherwise
!be occluded by cortical function. !
: i
j ;
I Neuronal Mechanisms of Habituation I
| -
| Habituation which conforms to certain of the para
meters described by Thompson and Spencer (1966) has been
j i
|observed with responses varying in complexity from the
|isometric twitch of the tibialis anterior muscle of the
! !
|spinal cat to the stimulus-seeking behavior of intact
i :
higher mammals. This fact has suggested that the process j
involves a decremental neuronal mechanism common to all j
i
animal preparations. However, as the foregoing die- |
i
cussion indicated, data gathered from intact higher j
mammals does not appear to allow a simple, passive,
13
fatigue-like mechanism to account for the phenomena ob
served at this level. One entry into the resolution of
I this problem might be a re-evaluation of the presumption
i j
|that a decrease in stimulus intensity leads to a decrease i
! i
in the neural activity which drives the orienting response, i
i • :
I i
! The coding of stimulus intensity in neural net- |
works is complex and the relationship between stimulus
jintensity and neural activity is not necessarily direct.
|It will be seen in the next section that decreases in !
I light, for example, lead to increased unit activity in j
|visual pathways (Arduini and Pinneo, 1962; Brooks, 1966).
i j
!In the auditory system cells have been described which
i '
j ;
!fire to weak intensities but not to strong ones (Rose,
j i
iet al., 1963). Thus, a change in the parameters of
|stimulation might be expected to produce changes in the
i
;activity of cells not previously triggered by the
|habituated stimulus. If habituation, as a process, in
volved a decrease in synaptic efficiency, an increase in
|neural firing would be expected from a decrease in the
| :
|intensity of a stimulus because different neurons are
activated. Given this assumption, Thompson and Spencer
(1966) believe that the model of the stimulus proposed by
Sokolov is unnecessary because incoming stimulation, when
repeated, would tend to decrease neural response along ;
those pathways normally associated with that stimulus. i
The pathways of decreased efficiency might constitute the j
14
"trace" of the habituated stimulus. When the habituated j
stimulus is weakened, the resulting activity might be j
composed of the activation of some of the previously
|weakened pathways and new elements not previously used.
i
! Stimulus generalization would then be a function of the j
|proportion of previously activated elements to new ones. j
This proposal has the merit of simplicity and j
I adaptability to the paradox outlined above, where habitua- j
|tion is obtained despite the absence of the cortical |
!material presumed necessary to obtain it. This new pro- !
i :
|posal would involve a decremental process occurring at all s
I levels but with greater intensity in some structures and
less in others. Thompson and Spencer (1966) have pro
posed that several decremental processes observed in ;
i j
!spinal preparations underlie the process of habituation
jin any preparation in which it is observed. One of these
|two processes, variously termed "low frequency depression,"!
j"homosynaptlc depression," "defacilitation," or "sub-
synaptic depression," involves apparent decreases in !
synaptic efficiency and has been studied in simple reflex |
systems in the spinal cord (Kandel and Spencer, 1967»
|
review this material and its potential application to
i !
habituation theory). It seems likely that this kind of
inhibition is present in more central structures and is j
elicited by low frequency synaptic bombardment as well as J
electrical stimulation. Low frequency in this case refers j
15
to iterative activation at the rate of less than 10/
second, which is in the range of afterdischarge to stimuli
observed in many areas of the thalamus (Anderson and
Eccles, 1964). If this low frequency depression builds up
in structures more susceptible to afterdischarge, these
would be expected to exhibit more habituation than
structures less susceptible to afterdischarge. Those
structures exhibiting less iterative discharge, like the
classical sensory system, would be expected to show less
decrement in response to a stimulus. The prediction from
this hypothesis, therefore, is that more habituation will
be seen in those neurological structures which generate
afterdischarge to the stimulus than those which do not.
The test of this hypothesis is to measure the amount of
neural activity in neurological structures during
habituation to a stimulus. If a response to a stimulus
is present, those structures which show the greatest inter-
stimulus activity change, i.e., afterdischarge, to the
presentation of that stimulus will also exhibit the great
est decrement in the response to that stimulus.
The multiple unit response provides a measure of
overall neural spike activity and might provide the means
to test this hypothesis. To evaluate this and other
questions, an examination of neuroelectrical activity
measures will be presented in the next chapter, and the
16
third chapter will be concerned with the changes in these
activity measures seen during habituation.
17
CHAPTER II
ELECTRICAL RESPONSES IN THE BRAIN
It was seen in the last section that habituation is
a response-defined process in which many of the ongoing
activities in an animal are subject to alteration. It has
been observed that nearly all peripheral responses evoked
i
! by a stimulus can be subject to habituation (Harris, 194-3),
I
i but some electrical responses in the brain can be very
!resistant to habituation and may never extinguish. This
i section examines the electrical activity itself, particu
larly multiple unit activity, as an introduction to the
I next section on habituation of these responses.
iD. Q» Potentials
If one electrode Is placed on the brain and a second
electrode Is placed either close by or at some distance on
the body, an electrical potential can be detected with a
frequency spectrum ranging from much less than 1 to over
10,000 Hz. At the lower end of the spectrum, a "steady
state" D. C. potential may reach amplitudes exceeding 25
millivolts. This potential may show major variations over
hours and may be an indirect result of slow processes
associated with feeding, sleep, etc. (Oowen, 1967). More
rapid oscillations may correlate with activation, atten
tion, expectancy, sensory stimulation, and similar shorter-
18
term processes (Walter, 1964). These infra-slow potentials
can range from .1 to 1 Hz and are observed in many areas j
i
of the brain. Much research has been devoted to it re- j
i i
j
cently, particularly in its relationship to the orienting j
I response and motivation. Whether these shifts are j
i I
I important in the control of central processes or are an I
i !
epiphenomenon is still a matter of conjecture and some
|debate (Livingston, 1966). !
i i
:
I I
I EEG
I !
The electroencephalogram (EEG) is generally des- j
icribed as lying in the frequency range of 1 to 50 Hz.
i .
i This frequency domain has been intensively studied in many !
Scontexts and is known to be subject to many influences
I ;
!arising both internally and externally. Despite this
! ;
jmassive research effort, its fundamental nature still re
gains something of a mystery, although recent work shows
that at least part of it must arise from summated post-
! ;
I synaptic potentials (Creutzfeldt, et al., 1966). A major
j i
!problem in working with this signal is its complexity.
I
|It appears to be the result of the activity of large
Inumbers of neurons, many of which may be affecting the
! j
!total signal in unknown ways for a variety of reasons. One!
I
of these unknowns concerns the geometric relationship be
tween cells, and between these cells and the recording j
electrode. A second problem lies in the contribution to j
j
the EEG made by axon spike discharges, slow repolarizing___;
19
processes, dendritic potentials, etc. Any of these pro
cesses way influence the BEG and yet their relative
contributions may be partly dependent upon the exact
placement of the electrode and its electrical pickup
characteristics. A third complexity arises from in
fluences extraneous to the area of study. A signal may be
!picked up from areas relatively remote to the electrode.
! !
I Heath and Galbraith (1966) and Buchwald, et al., (1966b)
j !
Shave published results showing this kind of artifactual
icontamination of evoked potentials due to the spread of i
|signal evoked in other areas. Potentials arising from ■
I i
Inon-neural sources, e.g., eyeball rotation, muscle poten
tials, may also be recorded.
| !
| The extraction of meaningful information from the !
EEG is difficult at best. Current studies on frequency
analysis of the EEG and the relationship between the power |
! :
|spectrum and other physiological and behavioral variables
|show that these relationships exist, but work in the area
!has progressed slowly. (Por review, see Adey, 1965).
!
| :
[Single Unit Recording
| Very high frequency activity in the brain, ap- j
jproximately 1000 Hz, appears to result only from the all- '
or-nothing spike discharge of cells and their axons, thus
the origin of this phenomenon seems well established. ;
When the study of a single cell is desired, the recording
surface of the electrode can he made extremely small (1 f1 )
and a record taken of activity inside the neuron itself or
i
l
jat the cell membrane. These single-unit techniques have
yielded much information about the function of the neuron
|and its electrical reactions to outside influences. Un
fortunately, technical limitations have prevented any
!
: large-scale applications of this technique to the study
!
|of the relationship between gross behavior and unit
|neural processes. Recent innovations in technique have
|
made it possible to study single cells in the behaving
organism (see, for example, Evarts, 1966), but several
limitations are particularly severe in single unit work.
|One problem is the loss of cells over time. Micro-
!
|electrodes can seldom "hold" a cell for periods exceeding
i
]several hours; thus, manipulations exceeding this time
i
|limit are presently excluded. A second difficulty in
volves the nature of the single-cell recording itself.
j
jCells may be recorded which differ greatly from other
i
jcells in the immediate vicinity (Jasper, et al., i960).
|The significance of these differences is unknown but may
t
I
indicate the heterogeneity of many neural structures with
respect to functional organization. In view of this wide
degree of variance between cells within the same area, it
i
is difficult to estimate the significance of an extremely
small sample of a very large population with respect to
events occurring simultaneously in elements not sampled.
21
It is also difficult to discount possible bias arising
from the selective recording of cells which are more
easily detected by this technique. Finally, current micro-
I
jelectrode technique makes recording from more than one site
I
|difficult.
Multiple Unit Recording
I
The dividing line between single unit extracellular j
j i
'and multiple-unlt activity is a difficult one to draw as !
! ' i
the spike discharges of more than one unit are frequently
j
seen with microelectrodes. When the signal of interest is j
single-unit activity, the rest of the signal is usually
treated as noise. Increasing the size of the electrode
surface area decreases the influence of a single unit upon
[the total signal and increases the total number of units j
| I
jrecorded from (Schlag and Balvin, 1963). When slow EEG !
t !
: potentials are removed (by filtering) from the electrical
1 j
.activity being recorded from a relatively large macro-
jelectrode, e.g., 100// tip diameter or greater, the
I :
[resulting signal consists of a high frequency amplitude- j
[modulated signal, which appears to be the summated spike
activity of a very large number of cells (Buchwald, 1965). j
I
I
Problems in Interpretation j
J
The foregoing section has briefly considered three [
major frequency bands of the electrical response of the
brain. The basic difficulties associated with each make
it clear that they are hest used for specific applications
j
and that each may carry information that is unique to j
that frequency. An ensemble of unit activity recorded
from one electrode, the multiple-unit response, overcomes
certain difficulties associated with older types of
electrical recording and may be the most appropriate
type of technique to use in the analysis of certain
problems. Both B.C. and BEG share common problems which
do not plague multiple-unit analysis to the same degree.
I
The underlying mechanisms producing the slower frequency
responses are still unknown, and EEG and D.C. signals are
i i
I !
subject to many influences whose relative contributions
I are as yet in question. By contrast, very high frequency
| j
: activity is the result of a specifiable process, the unit i
spike discharge, which is the mode of distance trans-
| 1
j mission in the central nervous system.
i ;
i • !
! Slower frequency electrical activity is subject
! j
| to volume conductance, a fact which makes localization of j
the origin of a recorded signal difficult to specify. j
This is particularly true in D.C. infraslow potentials
and the evoked EEG potential, as noted earlier. This J
I ;
| problem is less acute in the case of high frequency sig-
!
j nals (Hendrix, 1965; Buchwald, 1966b), an advantage in j
studies involving evoked activity in specific structures,
as in the present series of experiments. j
i
The EEG can be more useful than D.C. recording in j
23
some circumstances because of its capacity to signal rapid
changes in the neuronal pool. The dichotomization of
those neuroelectric events which are relevant to the !
problem under study from those which are not is a dif
ficult matter. The question of signal versus noise
probably applies to all electrical responses in the
brain, but it may be most applicable to the EEG. Since
it is likely that no one single source is responsible for j
this activity, a variation in it might be due to many j
j
factors. Variation in very high frequency activity can
be traced to the spike potential and, thus, the problem
of the meaning of a signal and possible sources of noise
; is greatly simplified, though still complex. The use of
|
! multiple-unit technique confers certain distinct
1 i
| advantages over single-unit recording in that the popula-
| tion of cells sampled is much larger in relation to the
! total population and chronic implanted electrodes can I
! i
detect unit activity for months (Brooks, 1966). The j
i
choice of multiple-unit recording is an obvious one in j
the present study because of certain requirements:
!
distinct neural structures must be recorded from; overall
activity levels, as defined by quantity of action poten
tials over time, must be taken from these structures; and
this activity must be available over a number of days.
24
Multiple Unit Research
In 1962 and 1963, Arduini and his collaborators
published a number of papers on the visual system In
which the multiple unit response was used as a measure of
tonic neural activity. The first of these papers
(Arduini and Pinneo, 1962) summarized arguments in favor
of utilizing a mean square integration of high frequency
activity as a measure of the frequency or rate of spike
discharges in the unit population being measured. One of j
the premises of their interpretation is that, in a
situation where spikes summate with each other to produce
i ]
| the observed high frequency signal, the power of that
i
j signal is determined by the repetition rate of the spikes. :
; The mean-square integration of the high frequency signal
I can thus be a D.O. representation of the intensity of
1 . :
I unit activity at a given point in time. j
Several writers have questioned the need for mean- j
1
i square integration since publication of the original
i
Arduini and Pinneo paper. Schlag and Balvin (1963) have
pointed out that a number of integration techniques will
also produce an estimate of the power in the high fre- j
quency band. Buchwald, et al. (1966b) have agreed with
Schlag on this point and have shared his concern that the
assumptions made by Arduini and Pinneo might be unknow
ingly violated in actual laboratory recording situations, j
i
Both studies agree, however, that the multiple unit j
measure Is a useful one and that it can give information
unattainable by any other method.
In addition to unit activity, the noise produced
by the amplifier, electrode, and tissue resistance must
i
be considered. This artifactual noise level can be deter
mined by abolishing neural activity around the electrode
and observing the change in recorded signal strength.
Arduini and Pinneo (1962) did this by sacrificing their
animals and measuring the signal level before and after j
i
death. This noise figure can be deducted from the total
signal to obtain the actual physiological activity. The j
record taken from the living animal has several character- |
j istics which differentiate it from noise. The activity
i
i taken from an alert animal is higher in intensity and can
I ;
| demonstrate much greater variability. This variation may
| be associated with a stimulus change, movement, or be the
i I
effect of some unobserved cause. All of these character-
i
istics of the alert animal-is recording drop out at death j
(Arduini and Pinneo, 1962) or under deep anesthetic j
(Goodman and Mann, 1967).
Covariation of tonic multiple unit activity and
behavioral and BEG arousal levels of animals has been j
|
described by several writers. Goodman and Mann (1967)
studied the effects of general anesthetics upon tonic
activity in reticular and thalamic sites. Chronically j
implanted oats were observed during sleep and alertness
26
and while under the Influence of five kinds of anesthetics,
including pentobarbital, ether, and halothane. Depression
|
of unit activity was seen during the deeper levels of j
|
anesthesia up to the point of absence of any unit re
sponse to intense stimulation. A similar correlation be- |
j !
S tween behavioral states and multiple unit responsiveness
i
; was seen, except during paradoxical sleep, when unit
i j
| activity rose to levels a little higher than those seen j
during alert wakefulness. Podvoll and Goodman (1967) |
examined a number of areas in the brain during different |
| stages of arousal and noted that both thalamic and j
| reticular placements were much more responsive to arousal
i :
changes than any other area surveyed, including hippo-
: campus, caudate, both colliculi, cortex and amygdala.
| All structures showed an increase in unit activity during
| wakefulness and a decrease during slow wave sleep. Admini-^
! I
| stration of atropine led to slow wave activity in the EEG, j
I i
! but the unit activity in thalamus and reticular formation
| :
continued to vary with behavioral state. j
Schlag and Balvin (1963) observed the relationship
between multiple unit levels in the reticular formation
and the cortical EEG. Highest levels of unit discharge in :
the reticular formation were seen during cortical desyn-
I chronization and the unit activity was lower during
I |
cortical synchrony. Cortical spindles were also induced
by low frequency stimulation of the midline thalamus. The j
27
Intensity of stimulation required to produce spindling was
positively correlated with the level of reticular activity,
an effect also seen when reticular activity is artificially
raised with repetitive electrical stimulation. !
I i
These studies indicate that multiple unit activity
in thalamic and reticular sites is positively correlated
with arousal state and may be a better Indicant of these
states than any other electrical measure in the brain since
the measure provides a quantifiable level of electrical j
activity. The situation is less clear in the case of
other subcortical sites. The level of activity in these !
other areas is lower and appears less responsive to
i
!changes in level of arousal. Podvoll and Goodman believe
|that the high degree of relationship between thalamic
!sites and arousal may be due to the existence of a diffuse
|activating system throughout the thalamus. The influence
i :
|of this type of system may not extend to many extra-
thalamic subcortical areas to the same degree. j
i
Effects of Stimulation I
Multiple unit activity has been most frequently
studied in its relationship to external stimulation. j
Three kinds of stimuli have been most frequently used: j
1
visual, auditory, and chemical. One of the basic effects
of visual stimuli on tonic levels of unit activity was
first described by Arduini and Pinneo (1962) and was I
recently replicated by Brooks (1966) and Kasamatsu and_____ J
28
Iwana (1967). Recording from the optic chiasm and
lateral geniculate, Arduini and Pinneo observed that tonic
1
unit activity decreased when the animal was exposed to |
i
jsteady illumination and slowly increased following light
j
; offset and dark adaptation. This rather unexpected find
ing was ascribed to the operation of inhibitory inter
actions at the retina and might be similar to the lateral
! mechanism described by Hartline (194-9) for the llmulus
I eye. i
j ;
The multiple unit response to light in visual path- j
ways also includes a phasic increase in activity to onset !
and offset. Brooks (1966) measured incremental brightness
I
I thresholds at several background illumination intensities
in monkeys. She also observed multiple unit activity in
the lateral geniculate during the tests and found that the I
j
S test flashes produced no observable phasic unit response
iwhen the flash was below threshold for a behavioral |
i I
! response. When the flash was above threshold, the ’ 'on" !
| S
and "off" unit responses were directly proportional to the |
intensity of the flash at a given background illumination j
i
level. Another phasic response noted was an increase in
unit activity in the lateral geniculate to sudden, loud
auditory stimuli. This finding of a nonspecific response
to stimulus novelty was also reported by Podvoll and Good- |
man (1967). Both of these studies report decreases in j
this nonspecific response following stimulus repetition, ‘
j but neither specifically or systematically explored what
j appears to be a correlate of an orienting response.
Galbraith and Pinneo (1968) recorded the phasic
! evoked multiple unit response to light flash in optic
i
; chiasm, lateral geniculate, and visual cortex. Previous
! studies using multiple unit recording utilized relatively
i
i
long integrator time constants. As a result, rapid
changes in unit activity were not noted due to the
| relatively slow rise and fall times at the longer settings.
I
I Galbraith and Pinneo (1968) integrated at approximately 5
;msecs rather than the 100-7000 msec time constants pre
viously used. The signal obtained closely resembled the
EEG evoked response in most details, except for one
:intriguing difference— the late components of the multiple
unit evoked response were often missing. Since these late
components are usually identified with activity associated
Iwith nonspecific mechanisms like the reticular arousal
isystem, it is possible that these components might be
sensitive to states of excitation in the nonspecific
system. They also noted that, when flash pairs were
presented, the missing late components in the unit re-
i
Isponse were restored in the evoked response to the second
|flash. If the presence of the late component in the EEG
record and its absence in the unit response are due to
iessentially different forms of activity signalled by these
|two methods of recording, it may be that the evoked EEG
i 30
j
response represents in large part postsynaptic potentials
triggered by incoming action potentials. Multiple unit
activity may be composed of the spike activity which is
generated by the postsynaptic potentials. The difference
I
between the two signals, therefore, would represent the
difference between the input and output of the system
!
recorded from. When the late response was missing in the
unit record, the excitability of the cellular population
under observation may have been too low for these cells to
respond to the relatively diffuse excitation evoked by
nonspecific input. The effect of the first flash in a
pair might have been a rise in the excitability of the j
! cells to the point where they would respond to nonspecific
i :
I
j input with a spike action potential. Thus, the unit re-
i \
| sponse to the nonspecific input would be recovered.
i Habituation effects were not studied, so it is not known i
i i
j whether the late potentials changed during stimulus j
i * I
! repetition. j
In another visual stimulation study, Kaufman and j
|
Price (1967) recorded high pass activity from the scalps
of humans and were able to demonstrate a stimulus-locked
evoked response to sinusoidal light stimulation. A series ;
of elaborate control measures were necessary to show that
i
the response was due to the light, but their demonstration ■
did indicate that unit activity might have been detected
at a surprising distance from the cortex, though extremely j
31
attenuated.
Several multiple unit studies published were con
cerned with chemical stimulation. The Podvoll and Good
man (1967) and Goodman and Mann (1967) experiments have
already been discussed. Wayner (1967) observed changes in
tonic activity of hypothalamic and limbic sites following
hypertonic salt injection into the carotid artery. Des
pite the use of a rather crude technique, he found
reliable increases of activity in a number of areas im- j
j
plicated in the ‘ 'thirst circuit." Komisaruk, et al. j
(1967) reported reliable decreases in multiple and single j
j
j unit activity in the rat following progesterone injection, j
I A rise in activity was seen in the thalamus in response to ;
: j
i amyl nitrite vapor inhalation. One of the most interest-
l ing aspects of this paper was the perfect correlation be-
; 1
tween rises in cortical multiple unit activity and EEG
arousal. Single unit recordings were much less con-
i
sistent, indicating the sampling problem encountered with j
this technique. j
Another important area of multiple unit research
I
has been in audition. Starr and Livingston (1963) i
i
investigated the tonic activity in auditory pathways dur- j
1
jing prolonged exposure to white noise stimulation. Oats j
were chronically implanted in a number of auditory
stations, including the round window, cochlear nucleus, j
superior olive, trapezoid body, inferior colliculus, !
52
medial geniculate, and a variety of cortical sites. After
I
a recovery period, the cats were repeatedly given eight-
hour sessions which consisted of two hours of pre- j
!
stimulation control, two hours of 86 db. white noise
stimulation, and four hours poststimulation recovery. All
| electrical activity (bandpass = 0.2-10,000 Hz) was inte
grated and recorded on Esterline-Angus recorders for the
|
entire session. The fact that low frequency EEG activity j
i
was integrated along with high-frequency unit spike j
potentials makes interpretation of the results from higher
auditory stations very difficult. The situation at lower j
i
brain stem areas is better understood because less activity:
I in the EEG frequency range occurs there. The interpreta-
! tion of these data will focus primarily on the records
| taken from the lower centers.
j j
J Starr and Livingston found that white noise stimu- 1
|lation evoked a sustained increase in activity in the |
! ' I
j primary auditory pathway, which slowly Increased in
| !
amplitude for the first hour or so of stimulation. This !
increase in aotivity was greatest at the round window j
where it might increase over 50 percent. Much smaller
increases were seen at the level of the first synaptic j
relay station, the cochlear nucleus. Here activity in- j
creases averaged 35 percent. Increases were progressively j
i
smaller at higher auditory stations, but interpretation of j
these data is obscured, as noted earlier. A depression of !
i
: . . . . . . . . . . . . . . . . . . " i
I 33 ;
; spontaneous activity was seen in the cochlear nucleus,
superior olive, and inferior colliculus in the early post-
stimulus recovery period when compared with prestimulus
control period records. This poststimulus depression
decayed in less than 30 minutes in the inferior colliculus
and at somewhat faster rates at lower stations and medial
geniculate. The authors concluded that hoth the increment
in activity levels during stimulation and decrement fol-
i lowing stimulation were due to an active mechanism which
antagonized the action of the stimulus. This conclusion
; was largely based upon the loss of these changes during
! barbiturate anesthesia. While it is possible to attribute
| the slow rise in activity during the stimulation period
to the action of the middle ear (a conclusion supported by
; Carmel and Starr, 1963), the decrease in activity follow
ing stimulation is not so easily understood, as will be
seen later.
Carmel and Starr (1963) recorded from the round
window of cats with the same techniques used in the Starr
I and Livingston study. In addition, however, they section-
j
I ed one or both muscles of the middle ear. When both were
sectioned, a full response was obtained at the round
j window and no increase in activity was seen in the next
i
| two hours of stimulation. Nembutal anesthesia and de-
! cerebration gave approximately the same results. When the
j stapedius muscle was severed and the tensor tympani was
! left intact, the immediate response was higher than with
! normal animals and the rise to the final activity level
was faster, although not as fast as when both middle ear
muscles are out. This "dlsinhibitory" effect was also
seen to a lesser extent when the tensor tympani alone was
cut. Thus, the slow rise in round window activity ap
pears to be due to the action of both the stapedius and
tensor tympani muscles. Control of these muscles appears
i to be central, as their response to white noise drops out
when the eighth nerve is crushed. Carmel and Starr also
; observed the response to repeated presentations of 500
msec bursts of 90 db white noise. The size of the response
I
at the round window was attenuated in 10 responses and
returned after a two-minute rest. A reduction in the
! intensity of the white noise stimulation to 60 db either
abolished or markedly reduced this attenuation, which
indicates that the aversive effects of higher intensities
may have been responsible for the decrement. This middle
ear reflex will be discussed further in a later section on
: habituation. A conclusion to be drawn from these studies
is that the action of the muscles of the middle ear can be
! a potent factor in the evoked activity of more central
I receiving areas but also that the effect on these central
I structures is non-linear. Other mechanisms of activation
i and inhibition may also be Important in determining the
I :
| ultimate central effects of auditory stimulation.
35
Galin (1964) observed tonic activity in cats with
electrodes in the cochlear nucleus, inferior colliculus,
and medial geniculate. Each of these nuclei responded to
tones and white noise differently. In the cochlear nucleus
both tones below 3 KHz and white noise produced a sustained
increase in activity, and tones above 3 KHz produced a de
crease in response. In the inferior colliculus, the
inhibitory effects of tones extended to the frequency
range from 200 to 18,500 Hz, while white noise continued
to elicit sustained increases in activity. At the medial
geniculate, all tones produced a decrease in ongoing
activity, while white noise produced a small but steady
increase. Since habituatory effects were not considered,
it is not possible to fully evaluate these results, but
it may be that either the effects of some lateral
inhibitory mechanism similar to that observed in visual
systems is responsible or that repeated tone stimulations
have less of an activation effect than do white noise
presentations. In the former case, a decrease in activity
may have been seen because the receptive field of the
electrode may not have included a tonotopic projection
area in the structures sampled. Because white noise would
activate all tonotopic areas, an electrode anywhere in the
auditory system would not be outside a stimulated tono
topic area. One implication of this analysis is that
white noise stimulation may be the most appropriate audi-
36
| tory stimulus to use when investigating multiple unit
activity level changes, in response to either learning or
habituating procedures. In the present study, white
noise was chosen for the auditory stimulus on the basis of
these considerations.
It was noted earlier that activity in auditory
pathways may be influenced by both auditory input and
other activation systems. Galin noted that, even when the
■ response at the round window to steady tones or white
noise was constant, there was much variability in the
: activity level at inferior colliculus and cochlear nucleus.
These activity changes occurred in the absence of movement
' also. This observation suggests that variations in re-
; sponse at these levels can be due to central effects.
: Further evidence of central inhibition is seen in a
; second paper by Galin (1965)• Chronically implanted cats
were given 20 to 40 one-minute presentations of white
noise stimulation for 40 days. Following this, the noise
i was paired with intermittent shock for 7 to 21 days. Ig-
; norlng the rather unusual behavioral conditioning pro
cedures, it is still noteworthy that a rather unique
; phenomenon was seen— a depression in background activity
i developed which lasted for days after the noise-thock
j pairing was discontinued. Moreover, white noise presenta-
; tlons which had previously evoked sustained increases in
i activity level in the inferior colliculus now failed to
37
evoke any change from the reduced tonic level. The source
of this inhibition of activity is a mystery, but an
important point is that the consequences of reinforcement
are observed even at this level.
The last study covered in this section is by Buch-
wald, et al. (1966a). A 1.5 second i500 ops tone stimulus
was presented alone to implanted cats for 500 or more
I trials. Following this, tone was paired with shock to
; the hindpaw. After the conditioned association was
: established, it was extinguished, retrained, and finally
i trained under flaxedil paralysis. Multiple unit response
I was recorded from auditory stations, somatosensory
! stations, and a few nonspecific activation areas. In
general, habituation procedures led to a decrease in unit
| responsiveness at the inferior colliculus to tone and an
inhibition of activity in medial geniculate and auditory
cortex. An onset response was seen briefly in the medial
j lemniscal system, but no sustained discharge. Much the
j same picture was seen in the mesencephalic reticular
system. With the onset of reinforcement, unit accelera
tion developed in all auditory and reticular formation
i placements. Activity in somatosensory areas was later
shown to be correlated with movement-induced afferent in
put, and other activity did not develop. Thus, the
i
; primary effect of reinforcement was exerted in OS path-
i
| ways and nonspecific systems. It is interesting to note
38
I that the noise-shock pairing led to increases in the
; activity level of the inferior colliculus during auditory
stimulation, a result directly opposite to that found by
Galin (1965). Since Buchwald did not record absolute
; activity levels, it is not known whether tonic activity,
as opposed to evoked, might have changed since the Buch
wald technique did not make provision for this kind of
measurement. In the two studies, Galin and Buchwald used
i different integration techniques; this may have contri-
1 buted to the contradictory results. A second problem is
that Galin simply shocked the cats intermittently during
; white noise presentation, and this kind of procedure might
have rather unpredictable effects upon conditioning. A
1 third difference was the type of stimuli used— tone by
I Buchwald, and white noise by Galin. How these differences
might have affected the results is speculative.
A major problem with the Buchwald study is that only1
one recording site could be sampled at one time. Inter-
| actions between sites could not be seen and rapid changes,
i
: as during the first few trials of habituation and training,
| could not be detected. A second problem is common to all
| studies covered in this section: quantitative measures
I were not presented, making adequate evaluation of the data ;
j
i impossible. As a result, these studies must be considered
I as being merely suggestive of the notion that important
i
1 changes occur in primary and nonspecific pathways that may !
39
be sensitive indicators of the neurological changes under
lying behavioral learning (and unlearning). The technical
limitations of EEG and unit recording have prevented these
measures from yielding much information. Multiple unit
activity may provide a way to measure these changes
quantitatively and In relative isolation from other sig
nals which may also be important but which, in sum, ob
scure any single signal. The field is in its infancy and
the relative contributions made by different processes
which may interact at a single site is unknown. It has
been shown that at a single site both specific and non
specific influences are active and might be isolated by
the employment of averaged multiple unit evoked response.
Another method might be to study the interaction of both
activation system and modality-specific Input by simul
taneous recording of an area, its presynaptic specific
input area, and its presynaptic nonspecific input. The
analysis of these interactions, plus the influence of
other areas which might also exert an effect, is the
primary intent of this dissertation.
40
CHAPTER III
HABITUATION OP EVOKED NEURAL ACTIVITY
Prom the discussion concerning complications under
lying interpretation of the electrical activity of the
"brain, it is clear that conclusions drawn from this kind
of data must be approached with caution. The fact that a
strong correlation is found between a particular electrical
signal and some form of peripheral response does not neces
sarily prove a conclusion about the neural function under
lying that signal. An increase in the height of an
evoked response is not always an indication that excit
ability in a particular structure has gone up or even that
a greater amount of neural activity has been triggered.
Winters, at al. (1967), for example, have found an inverse
relationship between the size of the evoked response in
the reticular formation and behavioral arousal. A similar
observation was made earlier by Huttenlocher (i960, who
suggested that the effect might be due to occlusion: im
pulses arriving at an active nucleus might find fibers al
ready firing and thus be able to produce less of an evoked
response deviation from the baseline activity. Occlusion
like phenomena have also been observed at the level of the
cortex (Walter, 1964). Paintal (1966) has shown that
excitability changes in peripheral nerve can create a de-
I 41
: synchronization of fibers such that the apparent size of a
compound action potential appeared to decline when, in
fact, the total number of action potentials evoked by an
; electrical stimulation remained the same. This last
observation was made with a nerve subjected to cold block
which altered conduction time and resulted in a "smearing”
across time of what would ordinarily be a short synchroniz
ed deviation from a baseline. Prom these considerations
it is apparent that statements regarding the amount of
j activity signalled by an evoked response are somewhat sus-
j pect. With this in mind, it is now appropriate to briefly
review the literature on the habituation of the evoked EEG
' response.
Habituation of Evoked Potentials
The early literature on peripheral afferent
! inhibition relied in large measure on the demonstration of
i a decrement in the evoked response to sensory stimulation
at early synaptic relay stations following prolonged
i repetitive stimulation (Hernandez-Peon, I960). This kind
I of habituation has been reported for auditory, visual,
tactual, and olfactory stimulation (see Galeano, 1963,
for an early review). An examination of many of these
studies reveals, however, that a number of control pro
cedures were not observed and that data was frequently
not quantified or given adequate statistical treatment.
Worden and Marsh (1963) have criticized studies dealing
with auditory habituation and were unable to demonstrate
evoked response habituation in the cochlear nucleus when
the proper precautions were taken to insure uniform
stimulus presentation intensities. More recently, however,
Dunlop, et al. (1966) were able to demonstrate small
decrements in the evoked response to 20 msec tone bursts
in the inferior colliculus, medial geniculate, and audi
tory cortex. These changes were reliable and very quick
to occur reaching full development, typically, in the
first minute at brain stem sites. Very careful attention
was paid to the control procedures recommended by Worden
and Marsh so that no obvious criticisms seem indicated.
The authors suggest that this habituation was not ob-
: served by Worden and Marsh because the latter had taken
the second 50 responses as a sample for averaging and had
missed the fast developing change. In addition, very
little habituation was noted for repetition rates slower
than 1 tone per second. Interestingly, this habituation
was not abolished by anesthetizing the animal with
: nembutal. This last observation lends support to the
' possibility that this phenomenon is the result of the
reflex activation of the Bundle of Rasmussin, as this
I inhibitory influence is active in anesthetic states
| (Lebrandt, 1966).
i
Attempts to obtain habituation of evoked EEG
43
; responses in nonspecific structures have led to somewhat
i contradictory results also. Thompson and Shaw (1965)»
for example, were unable to obtain habituation to auditory,
visual, or somatic stimulation in association cortex. A
negative correlation was noted between degree of orient
ing to the stimulus and the size of the evoked response.
Their conclusion was that alerting led to evoked response
decrease. The opposite results have been found at the
cortical level by Teas (1964). Some of the variables
: noted above for evoked response studies in auditory path
ways may have created at least some of the confusion.
The evoked response taken from the reticular formation
and other nonspecific subcortical areas has been reported
to be very sensitive to arousal states and to exhibit
: great lability (Beyer and Sawyer, 1964). This sensitivity
to arousal might explain some contradictions in the
; literature on habituation of the evoked response in these
nuclei.
i This extremely brief review of some of the evoked
potential habituation literature suggests that no simple
• one-to-one transformation exists between changes in
I potentials evoked in different areas of the brain and
! waking arousal states and the size of the evoked response
I in at least some areas, i.e., decreases in arousal are
i associated with increases in the response. As a result,
; it may well be that the habituation of an arousal response
i has a potent facilitatory effect upon the evoked response
at the same time that other inhibitory processes are
building. In other words, the evoked response actually
observed could be the result of an interaction between an
inhibition built up during repetition and a facilitation
resulting from the decline of the arousal response. This
interaction could yield a result of no change or incon
sistent results and explain some of the confusion which
exists in the literature.
EEG Arousal Response
: The above hypothesis requires that an arousal
response be given to a stimulus. An EEG correlate of
arousal is the desynchronization response. The habitua
tion of the arousal response was studied by Sharpless and
Jasper (1956). Their report has become a classic and
their results have been replicated many times. In this
| study, they observed the electrical activity from the
brains of chronically Implanted cats while they habituated
; the arousal response to tones. The EEG arousal response
; was defined in terms of the presence of desynchronization
and its duration. In order to increase the clarity of the
j
I response, the animals were run during slow-wave sleep and
; tone presentation only given when the EEG was synchronized.
| Recordings were made from the cortex, reticular formation,
i medial thalamus, and posterior hypothalamus. In general,
I the pattern of activation observed at the cortex was seen. j
I .......... 45 ;
at the subcortical sites also.
The desynchronization, or arousal, response ex
hibited nearly all of the parameters seen in habituation
of peripheral reflexes. Two varieties of arousal were
found, a long-lasting (in terms of individual response
duration) response which was seen throughout the brain and
I habituated in five to ten trials on the average, and a
short duration response which was limited in distribution
I to the cortical projection area and thalamus and which
took an average of 30 trials to disappear. Both spon
taneous recovery and potentiation of habituation were
demonstrated, as were dishabituation, generalization, and
i below-zero training. As mentioned earlier, lesions made
at the auditory cortex, lateral lemniscus, and inferior
: colllculus led to changes in some of the specific
: features of habituation but not the general phenomenon.
The phenomena reported by Sharpless and Jasper
! have been replicated in many laboratories with both
: animals and man. Prom these studies, a number of
: theories have been proposed (Lynn, 1966) and demolished
1
| (Affani, et_ al., 1963). One problem with this kind of
1
j theorizing has been that, in a number of these theories,
j
I a single structure has been implicated in the proposed
inhibitory process and yet good habituation has been ob-
! tained in the absence of that structure. Another problem
I for these theories has been the demonstration of habitua-
46
tion to direct electrical stimulation of the reticular
formation and other areas believed inhibited during
habituation, Glickman and Feldman (1961) and Ursin, et_
al. (1967) have demonstrated habituation of the orienting
response to stimulation of the midbrain reticular forma
tion, septum, amygdala, and cortex. Since the inhibition
of the reticular formation was considered to be the
mechanism of habituation in some theoretical systems
! (Sokolov, 1963a; Gastaut, 1957; etc.), it would appear
ithat inhibition must at least also be generated at lower
levels, or perhaps all levels.
The Effects of Ablation
A number of studies have observed the change in
ihabituation following ablation of one or more structures.
The results are mainly concerned with peripheral rather
; than neuroelectric responses because of the technical
problems involved, but these experiments have implicated
: a number of structures involved in habituation. The
specific contribution made by each structure may be more
: related to the response affected than to the general pro-
j cess alone. Bagshaw, et, al. (I965) showed that amygdaloid
i ~ ■ “
lesions were more associated with GSR habituation than
i other responses in monkeys. Jarrard (1968) observed a
; decrement in activity following hippocampal lesions in
! rats (also see Douglas, 1967, for review). Orientation
47
to sound was depressed by lesions in the auditory cortex
in a study reported by Thompson and Welker (1965).
Glaser and Griffin (1962) and Griffin (1963) have shown a
decrement in heart rate acceleration following frontal
cortex ablation. Interestingly, an acceleration of the
Inhibition of heart rate was found following stimulation
of the same area which persisted for days (Glaser, 1965).
Caudate lesions apparently interfere with locomotor
responses to unfamiliar surroundings (Buchwald, et al.,
1961b). These results and others like them (see Diamond,
et al., 1963) suggest that a number of structures parti
cipate in response inhibition associated with habituation
and that no one central control center may exist. A
great deal of systematic work in the area of ablation
effects on habituation remains to be done in the future,
as the literature to date is still quite limited.
Unit Activity and Habituation
Other evidence on the neurology of habituation
comes from unit activity recording. The problems involved
in unit work were outlined in the previous chapter and
have prevented wldescale application of the technique
until quite recently. The list of studies involving
habituation is still quite short. Early studies were
reported in the Moscow Symposium. Jasper, et al. (i960)
recorded the unit response to 5/second light flash from
; 48
• a number of cortical sites. Various complex discharge
patterns were noted but, after a number of trials, 15-20
percent of the cells sampled had shown decreases in over
all response rate. Sokolov (1963b, 1965) has found
habituation in cortical unit populations. He also be
lieves that he has found a cell type which increases its
frequency of firing during the presentation of a habitual
stimulus and loses this response following dishabituation.
Morrell (I960) did not observe any consistent changes in
|the visual cortex to the repeated presentations of tone
;but did observe a decline in responsiveness in the
j
mesencephalic reticular formation. Scheibel and Scheibel
;(1965) reported a systematic investigation of the response
|of units of the reticular formation of paralyzed cats to
repeated sciatic stimulation. A slow decline in unit
response to this stimulation was observed to undergo
spontaneous recovery, and then demonstrate the potentiation
effect. This potentiation of habituation could become
I very well established. The amount of dishabituation to a
change in the stimulus was proportionate to the degree of
change in the stimulus. Electrical stimulation of the
|sigmoid gyrus of the cortex also gave a burst response in
the reticular area and this response also habituated. The
last report is particularly important to this study be
cause it suggests that unit activity in at least the non-
|specific areas would conform to the nine parameters of
I 49 1
habituation listed earlier. This possibility is further
indicated by the observations made with the EEG arousal
response. It would seem likely that a pattern of unit
acceleration might accompany the desynchronization re
sponse observed in the thalamus and reticular formation by
Sharpless and Jasper (1956).
It was noted earlier that Podvoll and Goodman
; (1967) reported multiple unit activity in specific sensory
pathways in the thalamus which was very sensitive to
: arousal state of the animal. This last observation sug-
; gests that activity in the specific relay stations of the
i thalamus might also demonstrate habituation decrement.
; If changes in arousal level of the animal were reflected
i ;
in the intensity of the multiple unit signal in the nuclei,
i then examples of all of the parameters of habituation
could conceivably be found there also. An important dif-
I ference between the evoked multiple unit response in non-
: specific nuclei and evoked potentials recorded from the
same areas is that increases in arousal lead to an in-
; crease in the multiple unit response of these areas while
j they might lead to a decrease in the evoked potential.
| If the previous interpretation regarding the interaction
[ !
j between arousal level and inhibitory decrement is cor-
i :
! rect, then it might be possible to see a decrease of
i :
evoked unit activity in the presence of an unchanged
evoked potential.
Previous work with evoked potentials has been
characterized by difficulties in control procedures and
inconsistent results in habituation paradigms. Work with
the arousal response has tended to give consistent results
which were more congruent with behavior. Unfortunately,
this EEG arousal response is not always easily identified
or manipulated. Based upon previous work with the
multiple unit response, it seems likely that the basic
: habituation phenomena seen with the EEG arousal response
could also be seen using the multiple unit activity
| measure. It is hypothesized, therefore, that the basic
j parameters of habituation proposed by Thompson and Spencer
1 and enumerated in the first chapter here will be found to
hold for the multiple unit activity recorded from the
i
reticular formation. It is expected that similar findings
j will be made in other areas which show the EEG arousal
i response. The situation regarding a primary sensory
! thalamic relay station like the medial geniculate is not
I quite so clear. Based upon the sensitivity to arousal
; levels reported for this site by Podvoll and Goodman
| (1967)* it is predicted that the same general pattern of
! parameters will be found here.
| Dunlop, at al. (1966) reported a decrement in the
I
j evoked response at the inferior colliculus level which
; was smaller than that seen at the medial geniculate.
| Starr and Livingston (1963) also noted a greater readiness
to habituate in the medial geniculate than inferior
colliculus with practically no tendency at all in the
second structure. It was predicted, based upon the re
sults of these two studies, that less decrement in re
sponse would be seen at this level than in the thalamus
and, second, that habituatory deficit would be relatively
small if it existed at all.
Some of the older neurological theories postulate
the action of one structure upon a second to explain
habituation of the orienting response. What proportion
of cells of a nucleus might be involved in this activity
is unknown, as is the effect on the multiple unit signal.
If the percentage is considerable, then this activity
might subtract from the deficit occurring because of the
decline in arousal state. The net result might appear as
a slow or nonexistent decrement when compared to other
! structures. This conclusion is obviously speculative,
! but it suggests a possible response which might be seen.
One puzzling finding regarding the evoked poten-
i tials has been that they fail to habituate despite the
I fact that the structure they are recorded from no longer
j gives a desynchronization response (Sharpless and Jasper,
I
! 1956). This has been widely interpreted as meaning that
i
i the habituation process occurred at some point beyond the
I first synaptic relay at the cortex, for example. It was
1
I also noted that Galbraith and Pinneo (1968) have found
i that the evoked multiple unit response shows a very strong
I correlation with the evoked potential. If an Interaction
effect between arousal and baseline could explain this
observation, then some kind of systematic variation might
be observed in the presence of unit activity under the
EEG curve. A decline, for example, might be seen in the
unit activity which was not reflected in the evoked poten
tial.
CHAPTER IV
METHODS
Subjects
Young adult male New Zealand white rabbits weigh
ing approximately 7 lbs. were obtained from a local supply
house. These animals were inspected for ear infections
at time of delivery and all animals used in the study
were free of external infection at the time of surgery
and habituation sessions. The data presented is based
! upon fifteen animals.
I
! Surgery
Shortly after arrival subjects were chronically
implanted with bipolar electrodes stereotaxically directed
i to various auditory, nonspecific, limbic, and other sites,
i Sodium pentobarbital (Nembutal) was initially used as the
! only anesthetic during surgery but a high subject attri-
i tion rate resulted due to rather narrow tolerance limits
| for this drug in the rabbit. A more satisfactory pro-
j cedure involved the slow intravenous injection of Nembutal
! until a moderately deep anesthetic level was reached. This
i i
| level was characterized by the loss of reflex responses to
i painful stimuli everywhere but at the head. Xylocaine
| :
| infiltration of scalp incision and stereotaxic pressure
i i
i :
I points was then given. This combined use of local and
54
general anesthesias was quite adequate for the surgical
procedures used here and Improved the recovery rate to
100 percent. The Ss were implanted with the Trent Wells
stereotaxic instrument and rabbit headholder according to
coordinates obtained from the atlas published by Sawyer,
et al. (1954). All electrodes were bipolar, made of 36
gauge stainless steel wire, and were coated with epoxalite
They had a vertical tip separation of 1 mm. The bottom
electrode was bared only at the cut end; the second wire
was also scraped | mm up from the end. The result was an
asymmetric bipolar arrangement in which the larger sur
faced wire could serve as the indifferent electrode of
the pair (Arduini and Pinneo, 1962). All animals received
six bipolar electrodes implanted in the left hemisphere
and brainstem. Electrodes were secured to the skull with
dental acrylic cement and leads were soldered to an
amphenol subminiature connector. The entire assembly was
embedded in an acrylic skull cap, which was also anchored
to the skull with stainless steel screws.
Electrophyslologlcal Data Acquisition
The data acquisition system is illustrated in
Pigure 1. Electrical activity was led from the animal
via shielded cables to a Grass IV EEG machine. Four
channels of the amplified signal were tapped off at the
J-9 stage, and passed through a multiple stage R-C filter
J9
Figure 1. Data Acquisition System
Filter
Filter
Heath Audio
Amplifier .
Tektronix
Model 122
Wollensak
Tape
Recorder
Grass
Model IV
EEG
Machine
BRS
Program
Logic
Monitoring
Equipment
Honeywell
7600
FM Tape
Recorder
j with a low frequency cut-off at 1 KHz and an attenuation
: of 18 db per octave below that frequency. The resulting
high frequency signals were permanently stored on a Honey
well 7600 FM tape recorder with a tape speed of 3-3/4
in./second in the extended mode. At this tape speed, the
frequency response was flat from D.C. to approximately
1250 Hz. Data on these four channels consists of high
frequency activity with amplitude down 18 db at 500 and
approximately 1400 Hz. As noted in the Introduction, it
was assumed that physiological activity in this frequency
;band is due to unit spike activity and very little else.
In addition, a fifth channel of activity was
I recorded from some animals with a Tektronix model 122
preamplifier set at frequency cut-offs of 80 Hz and 10 KHz.
;A second-stage Tektronix amplifier increased the signal to
iusable levels and this signal was also recorded along with
;the other channels. At times, the EEG signal from one of
:the bipolar electrodes was simultaneously recorded on tape
|
I along with the multiple unit record taken from the same
electrode. Ink tracings of EEG activity were also
recorded with the Grass'machine.
i
Multiple unit activity was monitored during experi
ments by amplifying the playback record from the tape
!recorder, integrating it, and using the D.O. level to
i
drive a microammeter. The multiple unit signal was also
imonitored with an earphone.
White noise stimuli were produced by a Grason-
I Stadler generator, Model 901B, taped on a Wollensak Model
T1500 tape recorder and replayed at 7i ips through a 4"
Quam speaker. Stimulus durations were two seconds and
noise levels were 60, 80, and 100 db (re 0.0002 dyne/cm2).
Sound level determinations were made with a General Radio
Type 1555A Sound Survey Decibelmeter placed in the
vicinity of a restrained animal's head. A determination
I of the voltage necessary to drive the speaker at a parti-
i cular sound intensity was made and this value was used to
; monitor the stimuli presented to the animals. These
j
I voltages were checked with a decibel meter at several
points during and after .completion of the experiment and
I
I found to be stable. A 25-watt electric light bulb located
t
| 16 inches to the left of the restrained animal was used as
I a dishabituatory stimulus. Stimulus presentations were
I controlled by BRS solid state programming equipment.
I i
| i
; Experimental Procedure
I Approximately two weeks after surgery, Ss were
! given three daily adaptation exposures to the experimental
; apparatus of 40 minutes each. The animals were placed in
i
| a 4.5 x 8.25 x 18 inch plexiglas restraint box with
! removable stock and placed in a 23 x 21 x 35 inch dark,
I sound attenuated chamber. No noise was introduced during
these sessions except for a constant and low (40-45 db) hum
from fans, equipment, etc. The head cable was attaohed_J
; and evaluation of the recording characteristics of the
electrodes was made at this time.
Stimulus presentation began on the next day follow
ing the last adaptation session. After 20 additional
minutes of adaptation, white noise was presented through
the speaker located 18 inches behind and 18 inches above
the rabbit's head. One half of the animals were given a
two-second burst of white noise at random intervals with
j a mean interstimulus interval of 30 seconds. The second
group was presented the same stimulus at a mean inter-
; stimulus interval of 10 seconds. Both groups received a
: total of 56 trials per day for two consecutive days at
| one of the three white noise intensity levels, given a
day of rest, and this sequence repeated at the remaining
intensity levels. In this way, each of the three white
i noise intensity levels was presented for two days to
; every animal. The sequence of intensity levels presented
! to any one animal was randomly determined with the re
striction that all three intensity levels would be
presented to each animal. A dishabituation trial was given
: on trials 16, 24, 32, 40, and 48. On these trials the
1 light was switched on for two seconds immediately prior to
| white noise presentation and light offset was stimultaneous
i
with noise onset.
At the conclusion of the experiment, Ss were placed
I in the experimental situation again and multiple unit
! 59
; activity was recorded. Following this the rabbits were
given a surgical dose of nembutal and activity levels
again recorded. This dose was the same as that used dur
ing electrode implantation procedures. Finally, a lethal
dose of nembutal was injected and a third recording made.
This procedure was employed in order to determine the
noise level of the electrode and recording system with all
conditions equivalent to the experimental procedure ex-
I cept the presence of a physiological signal. The ratio of
signal to noise determined by this procedure was utilized
; as a correction figure in the analysis of experimental
data.
All brains were fixed in 10 percent formalin with
the electrodes in place, then frozen and sectioned at 75X|
■ with a sliding microtome. Electrode placements were
verified with the photographic technique described by Guz-
I man (1958).
I Data Analysis
Figure 2A illustrates the basic multiple unit
' measure recorded on FM tape during an experimental session,
i Essentially, it is an amplitude modulated high frequency
I
; signal. As noted in the Introduction, it is believed that
| the amplitude, or power, of this signal is determined by
I the rate of unit spike discharge in the immediate vicinity
j
I of the electrode. Thus, when the average rate of firing
White Noise White Noise
Figure 2. The Multiple Unit Response to White
Noise Stimulation
!“ " ■ 61;
: in the population of cells under observation increases,
the amplitude of the high frequency signal also increases.
This change in the power of the high frequency signal is
proportional to the change in the frequency of the rate
of discharge of the cellular population sampled. This
basic data was processed using an integration technique
basically similar to that utilized by Arduini and Pinneo
(1962) and others. The high frequency multiple unit
• signal is converted into a D.O. level which varies in
: direct proportion to variations in the power of the high
i frequency signal. Figure 2B illustrates this relationship,
| and the analog circuit used to derive this conversion is
; illustrated in Figure 3» The time constant on the inte-
: grator was always set for the equivalent of 100 msec,
which was the level usually used in previous multiple unit :
I studies. At this setting the instantaneous power of the
; integrated D.C. level represents the average power of the
; high frequency signal over the period of the time con-
■ stant.
The basic measure used in this study was the mean
j multiple unit activity level (MMTJL) recorded during a
i
I period of analysis. This measure was obtained by sampl-
| ing the integrated D.O. signal for a period of two seconds ;
I
! at the rate of 50 per second and deriving an average level
I !
| of activity for that period. Four of these averages were
| calculated for each trial: one immediately before pre- ;
r .......................... 62!
i i
; sentation of the white noise stimulus, one during the
two-second noise period * a third immediately following
noise, and the fourth immediately following the third. In
this manner the level of activity of a neural structure
before, during, and after stimulus presentation was ob
tained which could be compared with other levels taken at
other times and at other sites. This level could be ex
pressed in microvolts by comparing it with a calibrated
1000 Hz signal injected into the data acquisition system.
In addition to the MMUL, a percentage figure was
calculated which provided a means of direct comparison
between different electrodes and animals. The first two-
i
|second MMUL taken Just before stimulus presentation served
|as a control period against which activity changes evoked
:by the noise could be compared. This calculation was made
!by subtracting the first MMUL from the second and dividing
jthe difference by the first MMUL:
MMUL II - MMUL I = EVOKED UNIT ACTIVITY (EUA)
MMUL I
I The result represents the change in MMUL from the pre-
i stimulus control period to the period of noise stimula-
t :
! tion, expressed as a percentage. Two other percentage
| figures were calculated between the prestimulus control
| period and the two periods following stimulus offset by
I
! using the above formula. In all cases these measures had
the advantage of controls against the effects of variables ;
i '
| which might remain constant within one placement in one i
r ”...... 63 !
I animal but vary in unknown ways between placements and
I animals. An example of this kind of variation is the
recording characteristics of the electrodes which partial
ly determine the intensity of the signals received at the
amplifiers.
The multiple unit data, recorded on FM tape, was
processed at the Systems Simulation Laboratory at the
University of Southern California. Figure 3A illustrates
i
the data processing configuration. The FM tape was played
I back with a Model 8100 Honeywell tape recorder and led
| into a Beckman 2132 EASE Analog computer. At this point,
; the signal was refiltered, amplified, full-wave rectified,
and integrated to give an average voltage measure with a
i time constant equal to 100 msec. The analog program used
: to get this integrated signal is illustrated in Figure 3B}
The output of the analog computer, the integrated
; multiple unit signal, was recorded on paper with an Offner
j Model D.O. polygraph and also digitized by an Adage Analog-
to-Digital converter. The digital output was sampled by an
IBM System 360-44 digital computer at the rate of 50
| samples per second. These samples were added and averaged
: in two-second periods to give the MMUL and further proces-
! sed to give the percentage measures described above.
| The data from this study filled over 600 pages of
I single spaced computer-printed output with each animal
i
j contributing over 13*000 separate raw data points and
Honeywell
8100
Tape
Recorder
Beckman
2132A EASE
Analog
Computer
Figure 3A. Hybrid System
Figure 3B. Analog Computer Program
— w ---
Adage
A-D
D-A
Converter
IBM
System 360
Model 44
Digital
Computer
Polygraph
Offner
10
Figure 3. Data Processing
Hybrid System
cr\
!
6 5 !
several hundred or more derived statistics. In order to
reduce this bulk of information to manageable proportions
for analysis, the data from individual trials were grouped
into blocks of seven consecutive trials separated by an
eighth, dishabituation trial. A mean for each block of
seven trials was automatically taken by the digital com
puter for each channel of unit activity and printed in a
summary section. Thus, each data channel was summarized
! by means of seven blocks of seven trials each. Dis-
i habituation trials were analyzed separately and not in-
; eluded in the trial blocks.
' Correlations between different placements within
i the same animal were also computed for blocks of seven
I ; trials in order to detect changes which occurred during
| the course of habituation. However, correlations were
i :
: not obtained for all animals or measures.
! The MMUL was used to detect excitability changes
j occurring during repeated stimulation. It was expected
: that the increases in activity levels seen in early stages
| of the presentation of a novel stimulus would tend to
i habituate out over trials and that this change would be
I seen in the MMUL measures. More rapid, phasic changes
i i
j might not be detected by this method if the net fluctuation
I over the two-second period equalledzzero. This kind of
! response was sought through averaged transient analysis
|
| of the evoked multiple unit response measure for a number
of animals. These averaged evoked responses were plotted
in analog form using the Adage Digital-to-Analog converters
to reconvert data from the digital computer into analog
potentials which were recorded with the Offner Dynagraph.
Some of the other measures were also plotted with the
Dynagraph employing similar procedures.
In summary, multiple unit activity recorded from
rabbits exposed to repeated white noise stimulation was
processed on a hybrid computer system in order to derive
a number of descriptive measures of that activity. These
measures included average multiple unit activity levels
for two-second periods of analysis, changes between these
average measures associated with experimental manipula
tions, correlations between these measures taken from dif
ferent structures in the same animal, and the averaged
evoked multiple unit response.
67
CHAPTER V
RESULTS
I The data from this study will be presented in the
following order: (1) characteristics of the multiple
unit signal recorded from different areas of the rabbit
brain; (2) evoked multiple unit responses to white noise
stimulation of different intensities; and (3) changes
i occurring to the evoked responses following repeated
! presentation of the stimulus.
: Table 1 presents the mean intensity of the
; multiple unit activity recorded from histologically
I verified areas and the number of placements in each group.
| The recordings were made during the sacrifice session and
l
i the values are corrected for system noise by recording
I before and after death as noted in the methods chapter.
These intensity figures agree with those obtained by
i Podvoll and Goodman (1967)* who found that thalamic and
: reticular areas in the cat were more active, in terms of
i multiple unit activity, than other sites like limbic,
1
cortical, and extrapyramidal areas. Although sample
| sizes are small in this study, the same general relatlon-
i
| ship between brain areas appears to apply despite some
| differences in recording technique. Another observation
| made by a number of investigators was that the unit
TABLE 1.— Mean root-mean-square voltage levels recorded
from histologically verified electrode placements in
animals used in this study.
Structure N = Mean RMS Voltage
(in microvolts)
Inferior Colliculus 12 2.8
Medial Geniculate 12 3.0
Mesencephalic Reticular
Formation 10 3.4
Subthalamus 4
4.3
Caudate, Head
3 3.2 ;
Thalamic Central Grey 1 12.0
Frontal Cortex-Dorso-Medial 3
112
Lateral Geniculate 1 6.6 1
Hippocampus 1 1.0
Amygdala 2 1.6
Hypothalamus 2 1.2
i
: activity level measured at a particular electrode tended
to remain stable for long periods of time (Buchwald, et,
alo, 1966; Brooks, 1966; Galln, 1965)* This was also true
of the activity studied in the present experiments.
Variations were noted in response to different behavioral
states of the animals, e.g., startle, but the overall
average level tended to remain fairly constant.
Several points should be made about the values in
Table l.before the absolute levels, expressed in micro
volts, are compared with figures derived in other studies,
i A major difference between these data and those obtained
; elsewhere is in the narrowness of the frequency bandpass
of the signal reported here. The figures are derived from
a signal consisting almost entirely of the activity in the
frequencies very near 1 KHz, while those reported else
where tend to include activity considerably lower and
I higher in frequency. Both Brooks (1966) and Podvoll and
; Goodman (1967), for example, used frequency cutoffs from
less than 100 Hz to 10 KHz and got higher mean square
; voltages.
| Evoked Multiple Unit Activity
Figure 4 illustrates the nature of the multiple
| unit response recorded at each of the three different
i
| stimulus intensities. These scores represent the mean
I value of the evoked unit activity (expressed as percentage
Per Cent Change Prom Baseline Activity
50
40
0Inferior Colliculus
/ •Reticular Formation
30
/ ^Medial Geniculate
20
10
60 80 100
Stimulus Intensity
Figure 4. The Multiple Unit Evoked Response
as a Function of Stimulus Intensity.
deviations from the control pre-stimulus baseline) during
i
the first seven stimuli presented to an animal at a
! particular intensity. An increase in the intensity of
the white noise stimulus is seen to be associated with an
increase in the multiple unit response.
Some differences in the pattern of response to
these stimulus intensity levels are also noted. The in
ferior colliculus shows a higher evoked response at all
three intensities than either of the other two structures.
: This finding is consistent with data reported earlier by
! Starr and Livingston (1963) and Galin (1964). Both audi-
i
■ tory structures show increases in response going from 60
! to 80 db levels while the mesencephalic reticular forma
tion appears to demonstrate little or no change. All
three areas give an increase in response to the 100 db
i white noise, however, with the largest increase from the
: 80 db level occurring at the reticular formation.
Figure 5 shows four averaged evoked integrated unit
! responses to an 80 db stimulus recorded from one of the
; animals in this study. Bach curve represents the average
I of the first fifteen presentations in a session and were
| reconstructed from the original computer printout. Both
i
| an overall baseline shift towards increased activity and
| superimposed transients are seen at all four points.
| These rapid components appear to resemble the evoked
i
! responses taken from the same area, a finding consistent
72
Subthalamus
Reticular Formation
Medial Geniculate
Inferior Colliculus
White Noise
(2 Seconds)
Figure 5. Averaged-Translent Analysis of the
Integrated Multiple Unit Response to
White Noise Stimulation
73
with the work done by Galbraith and Pinneo (1968) in the
visual system. Some locations showed little or no con
sistent response. Included in this category were several
placements in the hypothalamus and basolateral amygdala.
Virtually all of the other locations observed and listed
in Table 1 gave some response to at least early presenta
tions of the white noise.
Habituation of the Evoked Multiple Unit Response
A number of parameters of habituation were examined
in this study. All but one of these comparisons were
within subjects and were analyzed with the Wilcoxon
signed-ranks test, Sign Test, or Freidman nonparametric
analysis of variance. The between subjects comparison,
intertrial interval, was examined with the Mann-Whitney
U Test.#
Figure 6 shows the group response curves during
the first and second habituation sessions at each of
three separate anatomical locations. Each point on the
curve represents the group mean for that block of seven
trials with the eighth, dishabituation, trial omitted.
(Thus, trial block 1 refers to trials 1-7, block 2 to
8-15, and so on.) Each curve consists of data from all
animals with electrodes in the structure noted above and
* All nonparametric statistical procedures are des
cribed in Siegel, S., Nonparametric statistics for the
behavioral sciences. New York: McGraw-Hill Book Co.,
Inc., 1956.
30
Inferior
Colliculus
Medial
Geniculate
Reticular
Formation
2?
24
21
18
15
12
9
6
3
6 6 4 1 2 1 2 2
Blocks of Seven Trials
Figure 6. Group Evoked Unit Response Means Over Trials During the First and
Second Days of Stimulations Major Groups
' includes measures at both interstimulus intervals and all
| three Intensities of white noise. In every case a decline
of responsiveness is seen on both the first and second
sessions of exposure to a white noise stimulus. This is
I
true for the three groups shown in Figure 6 and the three
additional groups illustrated in Figure 7.
Within groups comparisons using the Wilcoxon signed
ranks test for the difference between the first and last
seven trial block scores showed that the largest signal,
the response from the Inferior colliculus, significantly
declined over the first session (T = 0; P < .005 for n =
10). This loss of response, perhaps the most common
; defining characteristic of habituation, was also seen in
the other five nuclei and was one of the most reliable
: findings in this study. In the case of the medial
geniculate the decline is also significant (T = 2;
P < .01 for n = 10) although the relative size of the
| response itself is smaller. A similar situation is seen
i
| at the mesencephalic reticular formation (T = 35;
I P < .025 for n = 9). The data plotted in Figure 7 comes
| from groups too small to test for statistical signifi-
' cance (n = 4 for frontal contex and subthalamus, 3 for
' caudate). In spite of the sample size, however, trends
; towards the basic effects can be seen.
Spontaneous recovery over the 23 hours between
I
first and second day sessions Is also seen In all areas
Mean Evoked Unit Activity (EUA)
30
Frontal
Cortex
Caudate Subthalamus
21
12
o— o.
I i 1 1 1 1 l
1 2 3 ^ 5 6 ? 1 2 3 4 5 6 7 1 2 3 4 5 6 7
Blocks of Seven Trials
Figure 7. Group Evoked Unit Response Means Over Trials During the First and
Second Days of Stimulation: Small Groups
r 77
; except the frontal cortex. In no case, however, does the
mean response for the first block of seven trials on day
2 equal the first block score on day 1. It seems likely
that this was due to a greater decrement in response
developing on the second day during the presentation of
the first seven stimuli in trial block 1. The tendency
for response magnitudes to be smaller on day 2 of habitua
tion is seen in all six structures.
The tendency for habituation to be more rapid and
profound with repeated training has been termed "potentia
tion of habituation" (Thompson and Spencer, 1966). In
this study the effect was significant by the second day
i of habituation in at least the auditory structures and
reticular formation and appears as a definite tendency in
the others. Differences between the two days were tested
I by comparing the overall mean for each day in each animal.
These differences were tested with the Wilcoxon signed-
; ranks test and showed that the inferior colliculus
responded significantly less on the second day (T = 2;
P<»05). This was also true for the medial geniculate
(T= 1; P^.01) and the reticular formation (T = 3;
^*05)- Another effect which was very typical in this
study was the late-session "rebound" increase in response
i seen on the day 2 records in Figure 6. The effect is not
| so clear in Figure 7 except in the caudate record. It
i
I should also be noted that the first structure to show this
78
phenomenon on the second day is the reticular formation.
The inferior colliculus begins to recover a trial block
later and the medial geniculate shows a clear recovery on
trial block 7. One effect of this recovery was to reduce
differences between trial blocks 1 and 7. In spite of the
narrowed margin, however, the inferior colliculus showed
a decline in response over trials on the second day which
was still significant (T = 4; P = .05). A significant
decline was also seen in the medial geniculate (T = 3;
P ^ .025) but not in the reticular formation (T = 4.5;
P > .05).
Intertrial Interval Effects
The major between-subjects comparison was between
animals presented the auditory stimulus with a mean
intertrial interval of 10 seconds and a group with an
interval of 30 seconds. Figure 8 illustrates this data
which was derived from all sessions in each animal and
summarized into the two group curves. It will be noted
that the abscissa is now calibrated in units of time
rather than trials. As a result, the curve of the 10-
second group begins after eight trials at that interval
(80 seconds) and ends after 56 trials or 9 minutes and 20
seconds. The slower, 30-second group takes 28 minutes to
complete the 56-trial session.
Several trends are evident in all three structures
Mean Evoked Unit Activity (EUA)
30
27
24
21
18
15
12
9
6
3
Inferior
• ^^Colli cuius
■ V
Medial
Geniculate
. M
•
0
•
&
o
\
0
1
/
f t ■■■«!•
Reticular
Formation
\
\
y
0-o.-0
V
i i f
■< I
4 8 12 16 20 24 28 4 8 12 16 20 24 28 4 8 12 16 20 24 28
Minutes of Stimulus Presentation
Figure 8. Effects of Intertrial Interval on Evoked Unit Activity
VO
80
in the figure. A simple temporal effect is noted in
which the 10-second group reaches its greatest decrement
in response earlier than the longer interval animals. In
the reticular formation, for example, the lowest level of
responsiveness is reached in eight minutes (following 48
trials) in the 10-second group compared to 12 minutes and
24 trials. A second effect suggests that more may he
involved than simply the number of trials given per unit
time since the greatest decrement in response occurs in
the shorter interval group in all three structures.
When the group scores are compared at the point of great
est response depression for each group, the comparisons
are significant in all three structures as measured by the
Mann-Whitney U Test. This difference was most reliable
in the inferior colliculus (U = 465.5; .01; df = 52)
and somewhat less so in the reticular formation (U = 322;
p ^ .05) and medial geniculate (U = 360; p < .05).
Stimulus Intensity Manipulations
Figure 9 illustrates stimulus intensity effects on
the evoked unit response in the reticular formation and
the two auditory stations. Each curve was obtained by
pooling data from both intertrial interval groups and all
six daily sessions.
In general, an approximately additive effect of
intensity upon the average size of the evoked activity is
Mean Evoked Unit Activity (EUA)
50
45
40
35
30
25
20
15
10
5
Inferior Medial Reticular
• Colliculus
. \
Geniculate Formation
- \ / "
#
\
\ . / N • \
- v V
\ \
-
\ " * X
- # N
\ ^
_ Q V ^ C" 0 —°
\
0 \ ^ \
N 0 \ -__0— 0
0
0— •— o'
0 0—
\ 9
lOOdb
80db
1 2 3 4 5 6 ? 1 2 3 4 5 6 7
Blocks of Seven Trials
1 2 3 4 5 6 7
Figure 9. Effects of Stimulus Intensity on Evoked Unit Activity
82
seen which is significant (p-< .01) in all three record
ing areas using the Friedman two-way analysis of variance.
A decline in response over trials is also noted in all
structures at all three stimulus intensities. A closer
look reveals a difference in the rate of response
decrement recorded in the inferior colliculus in the 100
db noise condition when compared with the other two
recording areas. The response to 100 db noise starts at
a higher level in the colliculus and declines at a
relatively slower rate than in the other two structures.
Some tendency towards a comparatively slower rate of
response decrement in the colliculus is also seen at the
two less intense noise levels.
In all three structures very little difference in
the rate of habituation to the three intensities of
stimulation is evident over the entire session when rate
is defined as percentage decline from the level of
response of the first block of trials. Unfortunately, the
use of the first block of trials as a baseline of re
sponse is not very satisfactory because it is confounded
with the rapidly developing decremental processes under
way at the beginning of the habituation session. If an
inhibitory process builds up at different rates during
these first seven trials, then no equivalent baselines can
be derived between stimulus intensity conditions because
these baselines would have already been influenced by the
83
process which they are intended to measure. One test of
this possibility is to calculate the amount of decrement
developed between trial blocks #1 and #2 in all three
stimulus intensity conditions to see if the rates of
change differ between intensities.
In the inferior colliculus, the percentage change
was 25 percent between the first two trial blocks under
the 100 db condition and 44 percent with the 60 db
stimulus. This difference of 19 percent suggests that
habituation is occurring at a faster rate in the less
intense stimulus condition in the early trials and that
the level of response in the first block of trials is
differentially affected. A similar situation is also seen
in the medial geniculate and reticular formation, although
differences were not quite as large as in the colliculus
(9.5 and 4 percent).
Dlshabituation
Figures 10A and 10B present the effects of dis-
habituation obtained from the first two days of recording
and includes data from all intensity and interstimulus
interval conditions. In this analysis, however, the re
sponse Itself is calculated somewhat differently as the
effect deals with a change in the amount of activity evoked
following the dlshabituation.
In Figure 10A dlshabituation is defined as the
Per Cent Increase I n EUA Following Dlshabituation j
100
90
80
70
60
50
40
30
20
10
0
-10
10A 10B
•I
«
Inferior
Colliculus
Reticular
Formation
4. X
1 5 1 5
Dlshabituation Trial
Figure 10. Changes in Group Evoked Unit Activity Following Dlshabituation
oo
85
change In the evoked unit activity (EUA) between the dis-
habituation trial and the trial immediately preceding it
which served as a baseline. The remainder was then ex
pressed as a percentage of the overall mean EUA of the
immediately preceding trial block. Thus, the response
given in Figure 10A is the percentage increase in the
mean EUA following dlshabituation.
Sign tests (Siegel, 1956) were used to evaluate the
data in Figure 10. In the medial geniculate, a signifi
cant increase in activity (x = 4; p ^ .02) was observed
on the first dlshabituation trial which declined signifi
cantly (x = 3; p < .01) from the first to the fifth
repetition to a point not significantly different from
chance (x=9; p = .59). Similar results were obtained
from the inferior colliculus. Here the initial increase
and decline were very nearly significant (x = 4; p = .059,
for both comparisons), although the amount of change is
somewhat less than the thalamic placement. In the
reticular formation none of the comparisons were signifi
cantly different from baseline.
One problem with the foregoing analysis is that
the method of response calculation uses the two-second
pre-noise control period as a baseline from which the EUA
is derived. During a dlshabituation trial, however, an
experimental treatment Is given during this control period
which can change the level of activity considerably.
86
Figure 10B shows how much the change in baseline may have
affected the response in the reticular formation. In this
calculation, the control period from the immediately pre
ceding trial was used to derive a modified EUA. The
percentage response was then calculated according to the
formula used to derive the data in Figure 10A.
The results of this baseline correction procedure
are seen in Figure 10B. Here all three structures show a
significant increase in activity following dlshabituation.
The rise in reticular formation activity was very signi
ficant (x = 2; p < .01), as it was also in the medial
geniculate (x = 5; p < .01) and inferior colliculus
(x = 6; p = .032). The decline in the effectiveness of
the dlshabituation procedure was also significant in the
reticular formation (x = 2; p< .01). Thus, the reticular
formation, when compared to the auditory stations, can
show either a distinctively different or essentially
similar pattern of response, depending upon the type of
measure used. The major element of difference appears to
be the amount of activity evoked by the light during the
pre-noise period, the reticular formation being much more
responsive to the light stimulus presented alone than the
two auditory structures.
87
CHAPTER VI
DISCUSSION
The primary objective of this study was to test the
proposition that the level of massed unit discharge of a
number of brain locations can be a meaningful measure of
the neural substrates of learning and arousal. Other
studies have attempted to correlate this "multiple unit"
activity with overt behavioral arousal level and have
demonstrated that a strong relationship exists between
unit activity level and both the depth of sleep (Podvoll
and Goodman, 1966) and level of anesthesia (Goodman and
Mann, 1967; Winters, et al., 1967). Although decreases
in unit activity were also noted following anesthesia in
this study, the question of changes in unit activity
levels with learning was the primary one.
One implication of this proposition is that the
multiple unit discharge response to a novel stimulus
would show progressive changes with stimulus repetition
and manipulation similar to those found in behavioral and
other neuroelectric measures of arousal. Prior to the
undertaking of this study, only two research groups had
examined the multiple unit response during learning. Both
investigated the activity changes associated with tone-
shock pairing and both were subject to criticism regarding
procedure and data collection (see Chapter II for critical
review of Buchwald, et al., 1966a; and Galin, 1964). It
was felt that a more quantitative effort was required to
demonstrate valid changes in activity resulting from
experience. Moreover, it appeared desirable to carefully
examine the baseline from which associative learning
proceeds. Finally, the process of habituation is funda
mental to the organization of behavior (Hernandez-Peon,
I960) and of intrinsic interest.
In the Introduction it was concluded that the
criteria for habituation proposed by Thompson and Spencer
(1966) might provide the best measures of the habituation
of auditory evoked unit activity. As expected the cor
respondence was rather good in those areas intensively
studied, and information ovtained from other areas sup
port the notion that the same basic processes were opera
tive there also. In order to facilitate an orderly
development of the discussion of the data, the Thompson
and Spencer criteria will be repeated and conclusions
regarding each parameter presented.
The first criterion is the most commonly used
qualifying characteristic of habituation:
1. Given that a particular stimulus elicits a
response, repeated applications of the stimulus
result in decreased responsive (habituation). The
decrease is usually a negative exponential function
of the number of stimulus presentations.
In this study, decrements in the unit response to
89
the stimulus were found in virtually all areas observed.
Two notable exceptions were limbic placements, hypo
thalamus and amygdala, where no consistent response to
noise was observed. In the three largest groups, inferior
colliculus, medial geniculate, and reticular formation, a
highly significant decrement was observed which roughly
resembled an exponential curve. A similar pattern was
seen in other cortical and subcortical sites, although the
exact form of the response curve varied with the structure
involved. This last point may be important in evaluating
the source of the decrement since a mechanism of habitua
tion based on a single action of one anatomical structure
upon another might be expected to produce a fairly uniform
action. A closer look at the group and individual data
reveals a great deal of response variability between
structures and suggests that a simple single-factor model
may be inadequate.
2. If a stimulus is withheld, the response tends
to recover over time (spontaneous recovery).
In this study, an apparent recovery to the original
response strength was only seen in the caudate nucleus.
In the rest of the structures examined, the recovery seen
on the second day was only partially complete and was not
seen at all in the frontal cortex. At least part of this
apparent lack of recovery may be a product of the trial
block method of data summarization. It is clear from the
90
evidence on second-day facilitation of habituation that
the decremental process is proceeding at a faster rate
during the second session. It seems likely, therefore,
that the influence of this faster rate of habituation
would be seen in a seven trial average and that a lower
starting point might be explained by this mechanism alone.
A somewhat similar observation was made by Sharpless and
Jasper (1956) in their work with the BEG arousal response
and they were forced to "rest" their animals for days in
order to dissipate the accumulated effects of habituation.
3. If a repeated series of habituation training
and spontaneous recovery are given, habitua
tion becomes successively more rapid (this
might be called potentiation of habituation).
Several types of "potentiation" were seen in this
study by the second day of habituation training. The rate
of response decrement was faster and the extent of the
attenuation was substantially larger in every placement
group studied. Somewhat unexpected was the finding of
little relationship between degree of potentiation at
several sites and their level in the central nervous
system. As noted in the Introduction, it would be ex
pected from a theory of habituation based upon an inter
neuron mechanism that potentiation would tend to be
stronger in regions of greater interneuronal influence.
Thus, the medial geniculate should show a greater effect
than the inferior colliculus. This difference was not
91
observed. Unfortunately, the possibility exists that a
peripheral influence, e.g., the middle ear muscle reflex,
might also be potentiated on the second day. This
mechanism could influence the total activity level of the
inferior colliculus to a greater extent than in the
medial geniculate (see Starr and Livingston, 1963).
Thus, the two factors— interneurons and peripheral
action— might act antagonistically to produce the results
observed.
4. Other things being equal, the more rapid the
frequency of stimulation, the more rapid
and/or more pronounced is habituation.
This effect is commonly seen to occur as a positive
function between number of trials per unit time and ex
tent (or accumulation) of response reduction. In a
number of cases, however (e.g., Thompson and Spencer,
1966), greater response depression will occur in the
shorter of two intertial interval groups following an
equal number of trials. These two effects might be due
to different mechanisms. For example, the phenomenon of
habituation might be visualized as resulting from the
interaction of two mutually antagonistic processes. One
of these may involve an increment of inhibitory tendency
following each stimulus presentation. An opposing force
could be the action of the dissipation of this inhibitory
tendency and be partially dependent upon time, as indi
cated by the phenomenon of spontaneous recovery. The
92
finding that unequal response declines result when two
groups are habituated at different stimulus presentation
rates for an equal amount of time apparently results from
a greater degree of inhibitory increment in the group
having the most trials. On the other hand, an unequal
level of response inhibition following an equal number of
trials would suggest a difference in the amount of dis
sipation of inhibition over the dissimilar lengths of
training time.
Both effects were apparent in this study, suggest
ing that some form of dissipation mechanism with a fairly
rapid action is involved. The adjective "rapid" is used
in a relative sense as the difference between groups in
session time is less than 20 minutes. An even faster
dissipation rate was seen by Webster, et al. (1965)* who
found differences in asymptotic level of habituation be
tween groups which differed in intertrial interval rates
of less than one second. These end-point differences were
obtained in a matter of seconds, suggesting that this
process may be nonlinear with respect to speed of action,
may involve more than one mechanism, or both.
5. The weaker the stimulus, the more rapid
and/or more pronounced is habituation. Strong
stimuli may yield no significant habituation.
Like the intertrial Interval criteria, this para
meter of habituation actually contains two potential
effects. One assertion is made about the rate of decrease
93
and another about the endpoint or asymptote of that de
crease. In this study, a statistically significant dif
ference in response strength following habituation was
found between stimulus intensity conditions, but the
evidence for much difference in the rate of habituation
was not very convincing in any structure examined. Thus,
the possibility exists that, at the level of the central
nervous system studied here, the degree of response
decrement is positively correlated with the absolute size
of the response. The result would be that the total
percentage of the response which is lost through habitua
tion remains approximately proportional between intensity
conditions. The neural substrates of this mechanism are
speculative but might be consistent with the interneuron
proposal if the numer, or discharge frequency, of inter
neurons involved were proportional to the stimulus
intensity.
6. The effects of habituation training may pro
ceed beyond the zero or asymptotic response
level.
7. Habituation of response to a given stimulus
exhibits stimulus generalization to other
stimuli.
No specific test for either of these criteria were
included in this experiment, but several observations are
relevant. The zero response level was not attained in the
structures studied here as they tended to continue to
respond throughout training. On the first day of train
ing, however, what appeared to be an asymptote of
response decline was seen in all but the cortical place
ment, and response strength was fairly constant for the
last three trials or so. Habituation on the second day
rapidly fell to levels considerably below the steady level
reached on the day before, indicating that acquisition on
the first day may have continued beyond the point of
response decline.
8. Presentation of another (usually strong)
stimulus results in recovery of the habituated
response (dishabituation).
9. Upon repeated application of the dis-
habituatory stimulus, the amount of dishabitua
tion produced habituates (this might be called
habituation of dishabituation).
Both of these effects were elicited reliably in
the two auditory structures. In the reticular formation,
the situation was somewhat less clear, due partly to the
method of calculating the response. When the pre
stimulus baseline was derived from a fairly neutral condi
tion, both the rise in response strength over the pre-
dishabituation response level and "habituation of dis
habituation" effects were significant. As noted in the
Results section, this was not true when the baseline
control period was taken during presentation of the light
stimulus. The difference apparently stems from the
greater sensitivity of the reticular formation to activa
tion from the light stimulus. When the anatomical con-
95
nectlons of the reticular formation are contrasted with
the inferior colliculus or medial geniculate, this find
ing is not particularly surprising since the two auditory j
structures are less richly endowed with multimodal sensory !
projection fibers.
In summary, the multiple unit activity of a number
of anatomical areas (specific afferent, nonspecific
arousal, extrapyramldal, and association) has been shown
to be responsive to diffuse arousal (and, in some cases,
specific afferent) influences evoked by an auditory
stimulus. These responses were found to rapidly attenuate j
over repeated stimulus presentations and to display both j
j
recovery from this decrement following a 24-hour rest
period and considerable savings in rehabituation.
t
The reticular formation and two auditory structures,I
l
in addition, showed enhanced habituation with faster inter-'
trial intervals and weaker stimulation. Dishabituation
and habituation of dishabituation were also demonstrated.
|
It would appear, therefore, that a habituation process has ■
occurred which strongly resembles that observed with more
peripheral responses. Because this process appears to
adhere to the same parameters found to hold with the BEG
i
arousal response, it is further suggested that the two j
may be separate aspects of the same neural arousal i
mechanism. j
It was noted earlier that responses associated with !
96
the orienting response tend to drop out at different times
and that this tendency has been attributed to different j
thresholds of excitation for different response systems j
I
(Sokolov, 1963a). If some quantitative measure of central j
excitability state were available, then the relationships \
i
I
between this measure of activation and the occurrence of a j
response might give a better estimation of response
probability than exists at the present time.
The results of this study indicate that a pervasive !
influence on unit activity is exerted by arousal processes j
j
associated with orientation and that a measure of central |
excitatory state can be made with this technique. This |
i
conclusion is based upon the basic similarity in the j
records of activity obtained from a variety of non
specific subcortical and two auditory placements during |
i
the development of habituation and associated phenomena. ■
Thus, the basic results of Podvoll and Goodman (1967) are
supported and extended to a systematic habituation para
digm. It is suggested, therefore, that the multiple unit i
response is a relatively reliable index of arousal which
may provide a baseline against which hypothesized response !
system exdltability might be determined.
Some theoretical systems postulate that arousal
thresholds may play a deterministic role in the establish
ment of the probability of the elicitation of a conditioned
or learned response (Spence, 1966). It is possible that j
97
the absolute value of the multiple unit response taken
j
from more than one electrode in the reticular formation j
and thalamic placements will be positively correlated I
with the occurrence of a conditioned response. No direct
test of this hypothesis is available in the literature,
but it should be noted that Buchwald, et al. (1966a) found j
a definite rise in evoked reticular formation activity
during the development of a leg flexion response which j
apparently did not diminish late in training. j
The pervasive influence of this arousal response
appeared to extend to the inferior colliculus, which dls-
!
played the basic parameters of response habituation in |
i
much the same way as the reticular formation. This last
result is at variance with the data from one animal
reported by Podvoll and Goodman (1966). In their report, \
i
the inferior colliculus showed a fairly uniform response
to white noise stimulation which did not tend to
habituate although a decrement was observed in the medial
geniculate. In the present study, some variance was noted j
in the response at both the medial geniculate and in
ferior colliculus so that, by simple sampling alone, it
would be possible to see a decrease in the medial geni
culate response in the absence of a major decrement at
|
the level of the inferior colliculus. On other trials, j
I
however, the opposite conclusion might be reached— a de- j
cline in response at the inferior colliculus and not at j
98
the medial genciulate. Over trials the averaged responses
showed reliable decreases with the inferior colliculus
changing less than the medial geniculate but still showing !
a decrement.
This degree of independence of the two auditory
structures also implies that the decrement observed in the
auditory system is not due simply to the attenuating action;
of the middle ear muscle, although some effect might be
expected from this mechanism (Carmel and Starr, 1963)»
particularly at the level of the inferior colliculus. It
should be noted that a second possibility exists, namely j
t
that the activation level of the rest of the brain might '
indirectly depend upon the level of tension in the middle |
ear muscle in this experimental situation. The work by
Starr and Livingston (1963) on long-lasting, white noise
stimulation does not support this interpretation because
only very limited areas of the brain were responsive to
prolonged auditory stimulation per se. The involvement
of the rest of the brain appears to be a result of the
stimulus change in the environment. This question can ;
only be properly answered by neutralizing the action of ;
i
the middle ear muscles and noting the change in the j
arousal response at a number of levels beyond the round
window.
i
One of the more puzzling phenomena observed in
i
this study was the “rebound" recovery of the potentiated j
99
inhibition seen on the second day of habituation. A
number of authors have reported this effect in both man
and animals (Galeano, 1963) and the effect has been at- i
tributed to the spread of some sort of inhibition to the
cortex or reticular formation (Sokolov, 1963). Accord
ing to this formulation, the spread of the inhibition to
i
those structures engaged in active inhibition of the j
arousal response disinhibits parts of the response lead
ing to the return of the habituated responses.
Several observations Indicate that the effect may, j
in fact, result from the attainment of some minimum !
arousal threshold level: Figure 6 shows that an appreci-
i
able rebound was only seen on the second day of habitua- j
tion and that the degree of recovery was less in the
inferior colliculus than in the two other major
structures. On the first day of habituation, the develop- !
ment of a greater response decrement may have been checked j
by the introduction of the dishabituatory stimulus on
I
trial No. 16. As a result, the general finding was a
cessation of response decrement after the second trial I
I
block. On the second day responses reached lower levels
than attained on the first day. Recovery began in the
reticular formation on trial block 4 and was joined by the ;
l
medial geniculate, inferior colliculus, and caudate
i
shortly thereafter. The fact that recovery began in the j
reticular formation after it had reached a minimal level
100
of activity suggests that this recovery might represent
a release from inhibition arising from another source. A
second possibility might be that the potentiation of
habituation effect seen on the second day opposed by a
greater relaxation of the middle ear muscle and that this
recovery of activity represents a greater amount of ex
citation getting through to the cochlear nucleus. This
alternative is contradicted by the finding that the in
ferior colliculus, a structure less sensitive to arousal
level, is least affected by the rebound and that the
reticular formation apparently shows the onset of the re
bound earlier than did the inferior colliculus.
CHAPTER VII
SUMMARY
The basic concern of this study was the spike dis
charge behavior of massed populations of neurons in dif-
j
ferent functional systems in the brain and its relation
ship with those parameters which appear to characterize
habituation in many other response systems, including
I
such diverse behaviors as monosynaptic motor reflexes and j
i
exploration. On the basis of previous work with the !
multiple unit response in various arousal states and |
j
certain theoretical formulations, it was concluded that j
habituation of the evoked unit activity to a stimulus
should proceed according to basic parameters observed in j
i
other responses, although the rate and exact form of this
habituation might vary depending upon the location of !
the recording electrodes.
Fifteen rabbits were chronically implanted with
I
asymmetrical bipolar electrodes stereotaxically directed j
I
to a number of sites in the brain, particularly the mesen- j
cephalic reticular formation, brachium of the inferior I
colliculus, and medial geniculate. Following surgery and |
a recovery period, the animals were restrained inside a j
sound-attenuated box for three sessions of adaptation to J
the experimental environment and then give six additional j
102
days of habituation training to white noise.
The animals were randomly assigned to either a ten
or thirty-second intertrial interval group and also to a
sequence of auditory intensity conditions. Two days of
habituation at each of three noise intensity levels were
given at the assigned intertrial interval with a day of
rest between intensity conditions. Each day's session
consisted of 56 presentations of a two-second burst of
white noise and a two-second dishabituatory bright-light
stimulus immediately preceding the noise onset on the
16th, 24th, 32nd, 40th, and 48th trials. Electrical
activity was amplified, filtered, and stored on magnetic
tape for future analysis. Further signal processing was
done on a large analog-digital hybrid computer which was
programmed to extract a number of response measures. The
major response reported was defined as the net change in
unit activity between a two-second control period preced
ing noise onset and the two-second period immediately
following.
In general, significant decrements in the multiple
unit response to the novel white noise stimulus developed
in auditory, nonspecific, association, and extrapyramidal
sites. These decrements occurred faster and more pro
foundly on the second day of stimulation, indicating that
the effects of training persisted overnight. Some
spontaneous recovery did occur, but the response may not
103
have returned to prestlmulus levels. Other parameters
were examined in the auditory and reticular formation
placements. A shorter intertrial condition led to faster
and more profound habituation, as did weaker stimulus
intensities. Dishabituation was seen in the two auditory
structures, but was more ambiguous in the reticular
formation. Here, an interaction with the excitation
evoked by the light stimulus led to questions regarding
possible occlusional mechanisms. When dishabituation
trials were repeatedly presented, this arousal response
also diminished in Intensity in all structures observed.
These results support the view that a net decrease
in the firing of neurons in a number of neural areas
parallels the loss of the arousal or EEG desynchronization
response in the central nervous system and a variety of
peripheral responses. The development of this stimulus-
specific decrement may reflect the loss of an excit
ability necessary for evocation of these responses. The
variability observed between structures does not support
the view that the decrement appears at early synaptic
relays due to inhibition exerted there from higher
structures but does suggest the existence of an Inhibitory
mechanism operating at all levels of stimulus input and
processing.
Increased excitability following the introduction
&
of a dlshabituatory stimulus supports the view that this
104
process is one of sensitization. It also suggests that an
increase in unit activity may accompany the decrease in
evoked EEG potential repotted by a number of investigators
to be associated with increased arousal states.
LIST OF REFERENCES
105
LIST OF REFERENCES
Adey, W. R. Computer analysis in neurophysiology. In
Stacy, R. W. and Waxman, B. D. (Eds.). Computers in
biomedical researoh. Vol. I. New York: Academic
Press, 1965.
Affani, J., Marchiafava, P. L., and Zeraicki, B. Higher
nervous activity in cats with mldpontive petrigeminal
transections. Soience. 1962, 137* 126-127.
Anderson, P. and Eccles, J. Inhibitory phasing of
neuronal discharge. Nature, 1962, 196, 645-664.
Arduini, A. and Plnneo, L. R. A method for the quanti
fication of tonic activity in the nervous system.
Archives Itallennes de Blologie, 1962, 100, 415-424.
(In English)
! Bagshaw, M. H., Kimble, D. P., and Pribram, K. H. The GSR
of monkeys during orienting and habituation and after
ablation of the amygdala, hippocampus, and infero-
temporal cortex. Neuropsyohologla. 1965» 3» 111-119.
Beyer, C. and Sawyer, C. H. Effects of vigilance and
other factors on nonspecific acoustic responses in
the rabbit. Exp. Neurol., 1964, 10, 156-169.
Brooks, B. A. Neurophysiological correlates of bright
ness discrimination in the lateral geniculate nucleus
of the squirrel monkey. Exp. Brain Res., 1966, 2,
1-17.
Buchwald, J. S., Halas, E. S., and Schramm, S. Progressive
changes in efferent unit response to repeated cutaneous
stimulation in spinal cats. J. Neurophysiol., 1965»
28, 200-215.
_______. Changes in cortical and subcortical activity
during behavioral conditioning. Physiol, and Beh.,
1966a, 1, 11-22.
;_______• Relationships of neuronal spike populations and
EEG activity in chronic cats. Electroenceph. clin.
Neurophysiol.. 1966b, 21, 227-23^
106
107
i
I
| Buchwald, N. A., Wyers, E. J., Lauprecht, C. ¥., and
Heuser, G. The "caudate-spindle." IV. A behavioral
index of caudate-induoed Inhibition. Electroencenh.
clin. Neurophysiol., 1961b, 13, 531-537.
Oarmel, P. W. and Starr, A. Acoustic and nonaooustic
factors modifying middle-ear muscle activity in waking
cats. J. Neurophysiol., 1963, 26, 598-616.
i
; Cowen, M. A. Elementary functional correlates of the ,
transencephalic DO circuit. Psychophysiol., 1967, 3,
262-272. '
i I
Creutzfeldt, 0. D., Lux, H. D., and Watanabe, S. Electro- i
physiology of cortical nerve cells. In The Thalamus
ed. by D. P. Purpura and M. D. Yahr. New York: j
Columbia University Press, 1966. |
t i j
Danisch, P. Uber Reizbiologie und Relzempfindlichkeit !
von Vorticella Nebubulifera. Z. P. Alg. Physiol..
1921, 19, 133-188. j
i
Diamond, S., Balvin, R. S., and Diamond, P. R. Inhibition j
and choice. New York: Harper & Row, 1963. j
i
Douglas, R. J. The hippocampus and behavior. Psych. j
Bull.. 1967, 67, 416-442.
Dunlop, C. W., Webster, W. R., and Rodger, R. S. Ampli- j
tude changes of evoked potentials in the auditory j
system of unanesthetized cats during acoustic habitua- j
tion. J. Aud. Res., 1966, 6, 47-66.
Evarts, E. V. Pyramidal tract activity associated with a
conditioned hand movement in the monkey. J. Neuro
physiol.. 1966, 6, 1011-1027. j
Galbraith, G. 0. and Pinneo, L. R. Multiple unit evoked j
response to light. 1968 (in press). j
Galeano, 0. Electrophysiologlcal aspects of brain j
activity during conditioning. A review. AOTA Neurol. ,
Latino Amer., 1963, 9, 395-415. j
Galin, D. Auditory nucleic distinctive response patterns
to white noise and tones in unanesthetized cats.
Science, 1964, 146, 270-272.
108
I Galin, D. Background and evoked activity in the auditory
pathway: Effects of noise-shock pairing. Science,
1965, 149, 761.
Gastaut, H. The role of the reticular formation in
establishing conditioned reactions. In Reticular
Formation of the Brain, W. H. Jasper, Boston: Little,
! Brown, 1967. I
: i
Glaser, E. M. The physiological basis of habituation. |
London: Oxford University Press, 1966.
Glaser, E. M. and Griffin, J. P. Influence of the j
cerebral cortex on habituation. J. Physiol. (London), i
1962, 160, 429-445.
Glickman, S. E. and Feldman, S. M. Habituation of the j
arousal response to direct stimulation of the brain
stem. Electroenoeph. clin. Neurophysiol., 1961, 13, !
703-709.
Goodman, S. J. and Mann, P. E. Reticular and thalamic j
multiple unit activity during wakefulness, sleep, and
anesthesia. Exp. Neurol.. 1967, 19, 11-24. j
!
Griffin, J. P. The role of the frontal areas of the
cortex upon habituation in man. Olin. Scl., 1963,
24, 127-134. |
Guzman, F. 0., Alcaraz, V., and Fernandez, G. A. Rapid
procedure to localize electrodes in experimental I
neurophysiology, Bol. Inst. Estud. med. biol. Mex., j
1958, 16, 29-31. !
Harris, J. D. Habituatory response decrement in the
intact organism. Psych. Bull., 1943, 40, 385-422. J
Hartline, H. K. Inhibition of activity of visual I
receptors by illuminating nearby retinal areas in the !
limulus eye. Fed. Proc., 1949, 8, 69. j
iw-imn
Heath, R. G. and Galbraith, G. 0. Sensory evoked
responses recorded simultaneously from human cortex j
and scalp. Nature, 1966, 212 (5070), 1535-1537. j
Hendrix, 0. E. Transmission of electric fields in !
cortical tissue: a model for the origin of the alpha j
rhythm. Bull. Math. Biophysics, 1965, 27, 197-213.
109
i
Hernandez-Peon, R. Neurophysiological correlates of
habituation and other manifestations of plastic
| inhibition. Electroenceph. clin. Neurophyslol.,
supp. 13, i960, 101-ll4.
Herrick, C. J. Neurological foundations of animal
behavior. New Yorks Hafner Publishing Company, 1962.
Huttenlocher, P. R. Effects of state of arousal on
chick responses in the mesencephalic reticular forma
tion. Electroenceph. clin. Neurophysiol., I960, 12,
819-827^---------- ---------------
Jarrard, L. E. Behavior of hippocampal lesioned rats in
home cage and novel situations. Physiol, and Beh.,
1968, 3, 65-70.
Jasper, H., Ricci, G.» and Doane, B. Microelectrode
analysis of cortical cell discharge during avoidance
conditioning in the monkey. (In: H. H. Jasper and
G. D. Smirnov, Eds.: "The Moscow Oolloquim on
Electroencephalography of Higher Nervous Activity.")
Electroenceph. clin. Neurophysiol., supp., I960, 13,
137-153.
Jennings, H. S. Sea anemones: modifiability of behavior.
J. Exp. Biol.. 1905, 2, 447-448.
Kandel, E. R. and Spencer, W. A. Cellular neurophysio
logical approaches in the study of learning.
Physiol. Rev., 1967, 48, 65-134.
iKasamatsu, T. and Iwana, K. Some observations on
spontaneous optic tract discharges in free-behaving
cats. Jap. J. Physiol., 1967, 17* 220-233.
Kaufman, L. and Price, R. The detection of cortical
spike activity at the human scalp. IEEE Transactions
on Biomedical Engineering. 1967* 14, 84-90.
Kimble, G. A. Hllgard and Marquis* conditioning and
learning. New York: Appleton-Oentury-Orofts, 1961.
Komlsaruk, B. R«, McDonald, P. G., Whitmoyer, D. I., and
Sawyer, 0. H. Effects of progesterone and sensory
stimulation on EEG and neuronal activity in the rat.
Exp. Neurol.. 1967, 19, 494-507.
110
Lebrandt, 0. 0. The significance of the olivo-cochlear
bundle for the adaptation mechanism of the inner ear.
AOTO oto laryng.. 1966, 59, 124-132. !
j
Lehner, G. F. J. A study of the extinction of uncondi- |
tioned reflexes. J. exo. Psychol., 1941, 29, 435- !
456. !
|
Livingston, R. B. Brain mechanisms in conditioning and j
learning. Neurosciences Res. Prog. Bull., 1966, 4, j
235-354.
Lynn, R. Attention, arousal, and the orientation re- ;
action^ New York: Pergamon Press, 1966.
Morrell, F. Microelectrode and steady potential studies !
suggesting a dendritic locus of closure. In: H. H. j
Jasper and G. D. Smirnov, Eds.: "The Moscow ;
Colloquium of Electroencephalography of Higher !
Nervous Activity." Electroenceph. clin. Neurophyslol.,j
I960, supp. 13, 65-80. |
Paintal, A. S. Block of conduction in mammalian myelinat
ed nerve fibers by low temperatures. J. Physiol.
(London), 1965, 180, 1-19.
Pavlov, I. P. Conditioned Reflexes. (Trans, by G. V. |
Andrep.) London: Oxford University Press, 1927.
Podvoll, E. M. and Goodman, S. J. Averaged neural
electrical activity and arousal. Science, 1967, 155, j
223-225. j
Rose, J. E., Greenwood, 33. 33., Goldberg, J. M., and Hind,
J. E. Some discharge characteristics of single
neurons in the inferior colliculus of the cat:
Tonotopical organization, relation of spike counts
to tone intensity, and firing patterns of single
elements. J. Neurophysiol.. 1963, 26, 294-320.
Sawyer, C. H., Everett, J. W., and Green, J. 33. The
rabbit diencephalon in stereotaxic coordinates. J.
comp. Neurol.. 1954, 101, 801-824.
I
Scheibel, M. E. and Scheibel, A. B. The response or j
reticular units to repetitive stimuli. Arch. Ital. 1
Biol.. 1965, 103, 279-299.
Ill
Schlag, J. and Balvin, R. Background activity in the
cerebral oortex and reticular formation in relation
with the electroencephalogram. Ex p . Neurol.. 1963,
8, 203-219.
Sharpless, S. K. and Jasper, H. H. Habituation of the
arousal reaction. Brain, 1956, 79, 655-680.
Siegel, S. Non-parametrlo statistics for the behavioral
sciences" New York': McGraw-Hill, 1956.
Sokolov, E. N. Perception and the conditioned reflex.
New York: Pergamon Press, 1963a.
_______. Higher nervous functions: The orienting
reflex. Ann. Rev. Physiol., 1963b, 25, 545-580.
_______. Inhibitory conditioned reflex at single unit
level. Proc. 23rd Intern. Physiol. Congr., Tokyo IV,
1965, 340-343.
Starr, A. and Livingston, R. B. Long-lasting nervous
system responses to prolonged sound stimulation in
waking cats. J. Neurophyslol., 1963, 26, 417-431.
Teas, D. 0. and Kiang, N. Y. Evoked responses from the
auditory cortex. Exp. Neurol., 1964, 10, 91-119.
Thompson, R. P. and Shaw, J. A. Behavioral correlates of
evoked activity recorded from association areas of
the cerebral cortex. J. comp, physiol. Psychol.,
1965, 60, 329-339.
Thompson, R. P. and Spencer, W. A. Habituation: a
model phenomenon for the study of the neuronal sub
strates of behavior. Psychol. Rev.. 1966, 73,
16-43.
Thompson, R. P. and Welker, W. I. Role of auditory
cortex in reflex head orientations by cats to
auditory stimuli. J. comp, physiol. Psychol., 1963,
56, 996-1002.
Thorpe, W. H. Learning and instinct in animals. London:
Methuen, 195"6^
Unger, S. M. Habituation of the vasoconstrictive orient
ing reaction. J. exp. Psychol., 1964, 64, 11-18.
112
j Ursin, H., Wester, S. and Ursin, R. Habituation to
electrical stimulation of the brain In unanesthetized
cats. Electroenceph. clin. neurophyslol., 1967* 23,
j 41-50.
I
i Walter, W. G. The convergence and Interaction of visual,
auditory, and tactile responses In human nonspecific
cortex. Ann. H. Y. Acad. Sol.. 1964, 112, 320-361.
; Wayner, M. Hypothalamic and limbic system activation
I during salt arousal of drinking. Psvchon. Sol., 1967,
7, 179-180.
: Webster, W. R., Dunlop, C. W., Simons, L. A., and Aitkin,
L. M. Auditory habituation: A test of a centrifugal
and a peripheral theory. Science. 1965* 148, 654-656.
i Welker, W. I. An analysis of exploratory and play
j behavior in animals. In D. W. Fiske and S. R. Maddi
! (Eds.), Functions of varied experience. Homewood,
111.: Dorsey, 1961.
j
Wlnokur, G«, Stewart, M., Stem, J., and Pfeiffer, E.
A dynamic equilibrium in GSR habituation: the effect
of interstimulus interval. J. Psychosom. Res., 1962,
6, 117-122.
Winters, W. D., Mori, K., Spooner, C. E., and Kado, R. T.
Correlation of reticular and multiple unit activity
with auditory evoked responses during wakefulness
and sleep. Electroenceph. clin. neurophyslol.. 1967,
23, 539-545.
Worden, F. G. and Marsh, J. T. Amplitude changes of
auditory potentials evoked at cochlear nucleus dur
ing acoustic habituation. Electroenceph. clin.
Heurophyslol., 1963, 15* 866-681.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The Modification Of Partial Reinforcement Effect As A Consequence Of Electrical Stimulation Of The Caudate Nucleus In Cats

PDF

The Use Of The Orienting Reflex To Test The Zeaman And House Theory Of Attention

PDF

Monocular Acquisition And Interocular Transfer Of Two Types Of Discriminations In Normal And Corpus Callosally-Sectioned Guinea Pigs

PDF

Ucr Diminution As A Function Of Isi

PDF

Psychomotor Performance And Change In Cardiac Rate In Subjects Behaviorally Predisposed To Coronary Heart Disease

PDF

The Effects Of Transverse Accelerations And Exponential Time-Lag Constants On Compensatory Tracking Performance

PDF

The Effects Of Prior Part-Experiences On Visual Form Perception In The Albino Rat

PDF

The Effects Of Reinforcement Pattern, Intertrial Interval Regularity, And adaptation On The Orienting Response And Classical Gsr Discrimination Conditioning

PDF

Cue-Dependent Amnesia

PDF

Verbal Reports Of Emotional States And Onsets And Offsets Of Conditioned Stimuli

PDF

Human Performance As A Function Of The Joint Effects Of Drive And Incentive Motivation

PDF

The Effect Of Discriminability On The Partial Reinforcement Effect In Human Gsr Conditioning

PDF

The Effect Of Irrelevant Environmental Stimulation On Vigilance Performance

PDF

Reward Expectancy Strength As Related To The Magnitude Of Frustration In Children

PDF

Electrophysiological Properties Of Reaction Time And Aging

PDF

The Effect Of Motor Ability Loss On Cognition And Emotion

PDF

Operant Discrimination Of An Interoceptive Stimulus In The Urinary Bladder Of Intact And Dorsal Root Transected Female Rhesus Monkeys

PDF

Stress Response Modification And Autohypnotic Techniques

PDF

Sex-Role Preferences Of Early Adolescents In Relation To Adjustment

PDF

Signal Processing Time And Aging: Age Differences In Backward Dichoptic Masking

Asset Metadata

Creator Williston, John Stoddard (author)

Core Title Habituation Of The Multiple Unit Discharge Response To White Noise Stimulation In The Unanesthetized Rabbit

Degree Doctor of Philosophy

Degree Program Psychology

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

Tag OAI-PMH Harvest,psychology, experimental

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Galbraith, Gary C. (committee chair), Grings, William W. (committee member), Meehan, John P. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-385403

Unique identifier UC11361241

Identifier 7005233.pdf (filename),usctheses-c18-385403 (legacy record id)

Legacy Identifier 7005233.pdf

Dmrecord 385403

Document Type Dissertation

Rights Williston, John Stoddard

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 au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA