University of Southern California Dissertations and Theses

The Effects Of Ether And Electroconvulsive Shock On One Trial Appetitive And Adversive Learning

(USC Thesis Other)

The Effects Of Ether And Electroconvulsive Shock On One Trial Appetitive And Adversive Learning

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 6 7 -4 0 8
HERZ, Michael Joseph, 1936-
THE EFFECTS OF ETHER AND ELECTROCONVULSIVE
SHOCK ON ONE TRIAL APPETITIVE AND AVERSIVE
LEARNING.
University of Southern California, Ph.D., 1966
Psychology, experimental
University Microfilms, Inc., Ann Arbor, Michigan
THE EFFECTS OF ETHER AND ELECTROCONVULSIVE SHOCK ON
ONE TRIAL APPETITIVE AND AVERSIVE LEARNING
by
Michael Joseph Herz
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Psychology)
August 1966
UN IV ERSITY O F S O U T H E R N C A L IFO R N IA
T H E G RA D U A TE SC H O O L
U N IV ER SIT Y PARK
LO S A N G E L E S, C A L IFO R N IA 9 0 0 0 7
This thesis, written by
.........M £ h & e l . . j0.aep.h.JEfer.£.... ........
under the direction of hiS.....Thesis Committee,
and approved by all its members, has been pre
sented to and accepted by the Dean of The
Graduate School, in partial fulfillment of the
requirements for the degree of
DOCTOR OF PHILOSOPHY (Psychology)
Dean
Date. September _3j _ _ 1966
ACKNOWLEDGMENTS
I would like to express my deep appreciation and
gratitude to Dr. E. J. Wyers, not only for the support and
guidance which he offered during the experimentation
reported in this dissertation, but also for his encourage
ment throughout my graduate career. The many discussions
which have taken place with him through the years have
greatly contributed to my training as a psychologist and
are, I trust, reflected in the work reported here. In
addition I would also like to thank the other members of
my dissertation committee for their cooperation and
assistance. I also wish to express my gratitude to my
wife, Joan, without whose patience and support during the
disappointments and negative results I might never have
completed this phase of my career.
The research reported herein was supported in part
by Public Health Service Fellowship #Fl-MH-22,26 9 (to the
author) from the National Institute of Mental Health and
by Research Grants G-18891 and GB-4363 (to Dr. E. J. Wyers)
from the National Science Foundation.
Computing assistance was obtained from the Health
Sciences Computing Facility, University of California at
Los Angeles, sponsored by National Institutes of Health
Grant FR-3, and from the Computor Sciences Laboratory,
University of Southern California.
TABLE OP CONTENTS
Page
ACKNOWLEDGMENTS................................ ii
LIST OF FIGURES ............................ vi |
Chapter
I. INTRODUCTION.............................. . 1
II. EXPERIMENT 1 ................................. 20
Method
Results
Discussion
III. EXPERIMENT 2 ............................ 57
Method
Results
Discussion
IV. EXPERIMENT 3 . ....................... 91
Method
Results
Discussion
V. DISCUSSION ................................... Ill
VI. SUMMARY........................................ 134
REFERENCES............................................ 138
Chapter
APPENDIX A
APPENDIX B
APPENDIX C
Means and Medians for Control Groups
(Experiments 1 and 2), Shock Groups
(Experiment 1) .... . ...........
Means and Medians for Water-Reinforced
Groups (Experiment 2) . . ......
Means and Medians for Experiment. 3 . .
Page
145
147
149
v
LIST OF FIGURES
Figure Page
1. Apparatus Utilized in Experiments 1, 2,
and 3 ....................................... 21
2. Mean (and Median) Latency to First Head Poke
on Test D a y ................................ 32
3. Mean (and Median) Head Poke Latency Differ
ence Score (Day 1 - Day 2) 35
4. Mean (and Median) Latency to First Cup Touch
on Test D a y ................. . 37
5. Mean (and Median) Cup Touch Latency Differ
ence Score (Day 1 - Day 2) 39
6. Mean (and Median) Frequency of Licks on Test
D a y ......................................... 40
7. Mean (and Median) Frequency of Head Pokes
on Test D a y ................................ 42
8. Mean (and Median) Frequency of Cup Touch
Sessions on Test D a y ....................... 44
9. Mean (and (Median) Duration of Head Poking
on Test D a y ................................ 45
10. Mean (and Median) Duration of Cup Touch
Sessions on Test D a y ....................... 47
11. Mean (and Median) Latency to First Cup Touch
on Test Day:................................ 69
12. Mean (and Median) Frequency of Licks on Test
D a y ......................................... 72
vi
Page
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
4
24.
Mean (and Median) Frequency of Head Pokes
on Test D a y ................................. 73
Mean (and Median) Frequency of Cup Touch
Sessions on Test D a y ...................... 75
Mean (and Median) Duration of Head Poking on
Test D a y ................................... 77
Mean (and Median) Duration of Cup Touch
Sessions on Test D a y ...................... 78
Mean (and Median) Latency to First Head Poke
on Test Day .......................... 95
Mean (and Median) Head Poke Latency Differ
ence Score (Day 1 - Day 2) 96
Mean (and Median) Latency to First Cup Touch
on Test D a y ................................ 98
Mean (and Median) Frequency of Head Pokes on
Test Day ............... 100
Mean (and Median) Frequency of Cup Touch
Sessions on Test D a y ...................... 102
Mean (and Median) Duration of Head Poking on
Test D a y .................................. 103
Per Cent Retention as a Function of Time
Intervening between Learning Trial and
Administration of Amnesic Agent on Fre
quency of Licks Measure for Water
Reinforced Groups and on Latency to First j
Head Poke on Test Day for Shocked Groups . 118 j
|
Per Cent Retention on Frequency-Duration
Measures as a Function of Time Intervening
between Learning Trial and Administration
of Amnesic A g e n t .......................... 121
vii
CHAPTER I
INTRODUCTION
1
|
j
In the past several decades numerous investigations!
have been concerned with the perseveration-consolidation i
theory of memory storage (Hebb, 1949). This theory main
tains that for a period of time following a learning
experience (usually one-half to one hour) the memory trace
representing this experience is extremely labile and sub
ject to external interference. During this period the !
neural representation of the event is hypothesized to
become increasingly stable until it is consolidated and no
longer capable of disruption. For example, if a rat is
placed on a small raised platform and receives a footshock
when it steps from the platform to an electrified grid
i
below, it will remain on the platform when placed there on !
I
a subsequent test trial and an avoidance response is con- !
sidered to have been established. If, on the other hand,
the rat is administered electroconvulsive shock (ECS)
shortly after stepping off of the platform (and receiving
1
footshock) it will leave the platform as readily on the
test trial as it did on the original shocked trial. Such
absence of retention of the avoidance response is termed
retrograde amnesia (RA). The degree of such RA produced |
by ECS typically decreases as the interval between foot- i
shock and ECS is increased until a point in time -is reached]
i where ECS has no effect on the avoidance response and the
memory of the experience is said to be consolidated.
Evidence such as this is usually offered as support for the
consolidation theory of memory.
Numerous investigations have shown that the closer
in time that ECS follows a learning trial the greater the
i degree of RA produced (e.g., Heriot and Coleman, 1962;
Weissman, 1963, 1964). Such studies (see reviews by
Glickman, 1961; Deutsch, 1962; McGaugh and Petrinovich,
i 1965) have demonstrated that as the interval between the
learning trial and ECS increases the amnesic effect
j
diminishes until, after approximately one hour, convulsed j
i
: t
subjects suffer little or no impairment of memory. |
: i
i Results utilizing other amnesic agents have been |
similar. Traumatic head injury (e.g., Russell and Nathan,
1946), hyperthermia (e.g., Otis and Cerf, 1958), cortical
I
i spreading depression (e.g., Bures and Buresova, 1963),
carbon dioxide .(e.g., Leukel and Quinton, 1964), pento
barbital anesthesia (e.g., Leukel, 1957), ether anesthesia
(e.g., Herz, Peeke, and Wyers, 1966),. thalamic stimulation
(e.g., Mahut, 1964), and single pulse caudate nucleus
stimulation (Williston, Herz, Peeke, and Wyers, 1964) all
have been demonstrated to produce some degree of RA when
applied shortly after a learning trial.
Alternative Hypotheses Explaining
Retrograde Amnesia
In the decade following Duncan's (1949) major
parametric study demonstrating ECS-produced RA for shock
avoidance learning little data on retrograde amnesia
appeared until Coons and Miller (1960) challenged Duncan's
interpretation of his results. Their research indicated
. that although RA developed as a consequence of multiple
ECS's, indications of increased fear, as measured by urina
tion and defecation rates, also increased over successive
trials (each trial followed by an ECS). To determine
whether the interference with avoidance learning resulting
; from ECS was due to fear or actual RA, Coons and Miller
i (1960) performed an experiment to separate aversive from
i
I amnesic effects. They taught rats to reverse an active
I avoidance response when an electric shock in the start-box
4
was turned off and shock introduced into the goal-box.
i Under these conditions subjects learned to avoid the goal-
box (and shock) faster when ECS followed immediately after
each trial than when longer delays intervened between a
trial and ECS (results which are opposite from those pre-
i
dieted by an amnesic hypothesis). Any amnesia produced by j
ECS immediately following shock in the goal-box appeared I
to be overridden by increased fear produced by the summa
tion of ECS and footshock.
Several additional investigations have compared
the amnesic and aversive effects of ECS. McGaugh and
Madsen (19 64), Gerbrandt and Thomson (1964), and Gerbrandt
(1965) have demonstrated that one of the important con
siderations is whether one or a series of ECS's is used.
In these studies it was observed that the aversive compo
nent became apparent only after one convulsion per day
had been administered for four days or more. As a result
s
of such findings these authors suggested that future !
; ■ i
j studies utilizing ECS should use one-trial learning situa- I
| tions and one ECS in order that the amnesic effect be j
j observed prior to the development of the aversiveness
j resulting from numerous convulsions such as were used in
j the Coons and Miller (1960) study. In addition, studies |
5
such as these raise the important point that passive
rather than active avoidance learning situations appear
better suited for separation of the amnesic and aversive
properties of ECS. If ECS does produce fear and freezing,
as suggested by Coons and Miller (1960), interference with i
j
active avoidance learning could be the result of fear
rather than amnesia. However, in the passive avoidance |
situation where learning is demonstrated by the inhibiting
of a response, aversive and amnesic effects of ECS lead to
quite different predictions. The animal which demonstrates
RA in this type of situation must make an active response,
(e.g., approach the place where painful shock was received)
a state of affairs quite incompatable with an interpreta
tion in terms of the aversiveness of ECS.
A more recent alternative interpretation of the
results of ECS experiments is the "conditioned inhibition
hypothesis" proposed by Lewis and Maher (1965). These
i
authors take the point of view that unconsciousness is an j
unconditioned response produced by ECS, and that a weakened
version of the loss of consciousness, i.e., relaxation, j
lowered activity level, becomes conditioned to the cues
present at the time of an ECS administration. Further,
they suggest that this conditioned inhibition will vary as j
6
a function of the~nearness in time and space of the cues
associated with the convulsion. This prediction can,
therefore, account for the time gradient commonly observed
in studies in which ECS is administered to different groups
of subjects at increasing intervals following the learning
trial, although such conditioning appears unlikely with
trial-ECS intervals as long as. 30 minutes or more, despite j
the "massive nature of the unconditioned stimulus" (Lewis
and Maher, 1965, p. 237).
Although this hypothesis can explain much of the
data resulting from active avoidance studies, as could the
Coons and Miller (1960) interpretation, it is difficult
to fit the data resulting from one-trial, passive avoid
ance experiments into the framework of the "conditioned
inhibition hypothesis." As pointed out by McGaugh and
Petrinovich (in press), this hypothesis relies solely on
evidence from studies in which RA is indexed by increases
1
!
in latencies. As mentioned above, passive avoidance |
learning requires that a subject make an active response |
(demonstrating decreased latency) if RA is to be demon- i
i
i
strated. It is difficult to envision a situation in which
a totally relaxed animal (one which has inhibition condi
tioned to the experimental situation as a result of ECS in
7
the terms of Lewis and Maher, 1965), would be capable of
making the response necessary to demonstrate RA. Studies
such as those by Weissman (1963), 1964) and Heriot and
Coleman (1962), in which subjects return to press a lever
for food or water after receiving an extremely punishing
shock from the lever followed by a single ECS, are diffi
cult to interpret in the terms of the "conditioned inhibi- |
tion hypothesis." Also difficult to explain with this
hypothesis are studies utilizing a one-trial, stepping
situation (e.g., Quartermain, Paolino, and Miller, 1965);
Chorover and Schiller, 1965, 1966; Herz, Peeke, and Wyers,
1966). In these studies rats or mice step through a hole
or off of a platform and receive foot shock as the step
is made. Control subjects, which do not receive ECS,
evidence increased latencies or failure to step when
tested 24 hours later while immediately convulsed subjects
demonstrate unchanged or reduced latencies. In order for
the "conditioned inhibition hypothesis" to account for the
s
unchanged or reduced latencies of ECSed subjects such as
typically occur in passive avoidance, stepping situations,
conditioned relaxation would have to produce an increased
tendency to step in convulsed animals, a condition which
is exactly the opposite of the results of active avoidance
8
studies in which the probability of moving is reduced. To
account for this discrepancy Lewis and Maher (1965) suggest
that responses which are mediated by lower levels of
' arousal, such as passive avoidance responses, will evidence
less disruption and will be more likely to reappear in
convulsed animals. j
One additional point relevant to the "conditioned \
inhibition hypothesis" should also be mentioned. Lewis
; and Maher (1965) maintain that inhibition is conditioned
to various environmental stimuli present at the time that
loss of consciousness takes place. It is of interest to
note that in many ECS studies (e.g., Corson, 19 65; Cheva
lier, 1965) significant HA is demonstrated when ECS is
administered outside of the training apparatus. In addi
tion, Leonard and Zavala (1964) and Quartermain, Paolino,
and Miller (1965) have demonstrated that the degree of RA
produced by a single ECS is independent of the location
in which the seizure takes place (inside or outside of the j
j apparatus) .
Since all of the cited studies utilize shock, in
j
j an active or passive learning situation (and therefore
i involve conditioned emotional responses), it is possible
i
| that some complex interaction between ECS and conditioned
9
emotional responses (CER’s) is responsible for the pro
duction of interference with learned behavior which has
been interpreted as RA. More specifically, as is suggested
by the "conditioned inhibition hypothesis," conditioned
relaxation may reflect a situation in which the conditioned
emotional component of the avoidance response is attenu
ated, resulting in an absence of the freezing which, is
characteristic of animals which have received foot shock.
Indeed, results of studies such as those performed by Hunt
and Brady (1951) support such a contention. These experi
menters conditioned rats to stop bar pressing at the
presentation of a tone which had previously been paired
with shock (conditioned suppression). Such conditioned
suppression was labeled a conditioned emotional response.
Several days after the last conditioning trial, when the
CER was strong, subjects were administered a series of
21 ECS's (3 a day for 7 days). Two days after the last
ECS subjects were again presented with the CS while lever J
pressing. ECSed animals evidenced little conditioned
suppression, while control (non-ECSed) subjects stopped
pressing when the CS was sounded, suggesting that ECS's
had attenuated the CER but had not affected lever pressing.
However, the opposite relationship between the formation
of a CER and the acquisition of an avoidance response was
observed in a recent study by Gerbrandt (1965). In this
investigation conditioned emotional responding, as measured
by urination and defecation ra.tes, increased as rapidly
for animals receiving only ECS as a consequence of respond
ing as for subjects receiving response-contingent foot-
shock. Avoidance responding also increased rapidly in
the foot-shocked animals, whereas no avoidance responses
were observed during early trials in the ECS group. These
results suggest that ECS can produce dissociation between
conditioned emotional responses and avoidance responses,
with the memory for the response component but not the
motivational component being disrupted.
One final series of studies will be mentioned in
this section, not because they are inconsistent with the
perseveration-consolidation hypothesis, but because the
ECS-produced RA gradient is significantly briefer than
j
that observed in other studies, a difference which is
extremely important in terms of the neurophysiology of
memory. Investigations by Chorover and Schiller (1965,
1966) and Quartermain, Paolino, and Miller (1965), utiliz
ing a stepping, passive avoidance situation, have demon
strated unimpaired retention with foot-shock-ECS intervals
11
exceeding 30 seconds although clear RA was produced with
briefer intervals. The issue is further confused when it
is noted that Hudspeth, McGaugh, and Thomson (1964) have
observed RA with shock-ECS intervals up to 30 minutes in
a similar passive avoidance learning situation.
One point relevant to the present discussion
emerged from Chorover and Schiller's most recent investi
gation (1966), which utilized the paradigm of Bures and
Buresova (19 63). Rats were first placed in the larger of
a two compartment box and the time required to enter the
smaller compartment recorded. Subjects were then removed
and later confined in the latter compartment where they
were administered unavoidable foot-shock. Bures and
Buresova (1963) found that ECS administered up to two
hours after the foot-shock impaired avoidance of the
smaller compartment when subjects were tested 24 hours
later. Chorover and Schiller (1966) replicated this find
ing but in addition demonstrated that confinement during
the administration of foot-shock was responsible for the
long interval ECS effect. When subjects were foot-shocked
in the small compartment, but allowed to escape, no ECS
interference effect resulted, even with convulsions
occurring one minute after foot-shock. In the same
I 12
| investigation, but using a step-down, passive avoidance
! apparatus, it was observed that ECS, following confinement
in a small compartment immediately after the foot-shock,
! produced reliable attenuation of the passive avoidance
i response, while ECS alone produced no such interference.
The results of these studies appear to indicate that some
i degree of situational dependency may be involved in the
production of ECS interference effects with certain types
i of apparatus. It would appear that the different temporal
gradients of ECS interference resulting from the many
studies of passive avoidance can be accounted for in part
by differences in the apparatus and procedures employed.
Retrograde Amnesia with Ether Anesthesia
The situation with regard to anesthetic-produced
; RA appears to be less controversial? as yet no one has
argued that the use of such agents may produce aversive as
well as amnesic effects on retention. Although no direct
: investigation of their aversive properties has been under
taken, direct comparison of the effects of diethyl ether
and ECS resulted in complete RA with both agents in a one-
J trial, passive step-through avoidance situation (Herz,
; Peeke, and Wyers, 1966). Both ether and ECS administered
13
within one minute of foot-shock resulted in test trial
latencies which were indistinguishable from non-shocked
control mice. This study, when examined in conjunction
with the investigations of Herz (19 62) and Alpern and
Kimble (in press) suggests that the duration of etherization
and temperature at which ether anesthesia is induced are
critical variables in the production of RA. Herz (1962)
found that it was necessary to etherize subjects for 120
seconds in order to produce results suggestive of RA at
22° C. Herz, Peeke, and Wyers (1966) found no RA with
40 second duration of ether anesthesia at 30-31° C. and
complete RA with 70 second anesthetization at the same
temperature. Alpern and Kimble (in press) found that only
15 second etherization was necessary to produce RA with
potentiated ether (heated to 43-46° C.j . The results of
these studies point to interaction between temperature and
duration, the physiological mechanism for which is at
present unknown.
I
Additional studies utilizing ether anesthesia as
an amnesic agent and passive avoidance tasks have demon
strated an amnesic gradient similar to that found with ECS.
Abt, Essman, and Jarvik (1961) , Essman and Jarvik (1961) ,
and Pearlman, Sharpless, and Jarvik (1961) observed
14
significant RA with shock-ether intervals up to 10 or 15
minutes. In addition, a recent investigation by Pearlman
(1966), in which both ether and cortical spreading depres
sion resulted in comparable degrees of interference in a
step-down, passive avoidance situation, suggests that
ether-produced RA may also be the result of disruption of
neocortical activity. (Pearlman, 1966, infers this from
evidence presented by Pribram, 1960, and others, which
appears to indicate that simple tasks, such as ones
involving latency measures in passive avoidance learning,
are mediated principally by cortical function.) Although
no direct investigation has been made of the aversive
properties of etherization, as pointed out above, the
evidence at present appears to support an amnesic inter
pretation. Since all of the cited studies utilized a
passive avoidance situation, the conditioned relaxation
explanation of Lewis and Maher (1965) does not appear
relevant.
Won'-aversive Learning Situations
Because of the possible interaction between shock
and ECS, non-aversive learning situations would appear to
be ideal for the study of RA. Although an incomplete
report of the amnesic effect of ether anesthesia on
imprinting has appeared (Gutakunst and Youniss, 1963),
there appears, to be only one careful investigation of the
effect of an amnesic agent (ECS) in a one-trial, appeti
tive learning situation. Tenen (1965a) placed rats in a
small compartment to which a small, short tunnel was
attached. Over subsequent 4-minute sessions subjects
significantly reduced the number of times that they poked
their heads into the tunnel. A water spout was then
inserted into the tunnel and animals were permitted to
drink for a 10 second period. When tested 24 hours after
the water-reinforced session, control subjects evidenced
a significant increase in head pokes to well over the
level observed when they were originally introduced into
the situation. Subjects which received ECS immediately
after the drinking experience, on the other hand, showed
no such increase but were comparable to animals which had
received no water on the preceding day. (Subsequent data
referred to by Tenen, 196 5b, suggests RA in this situation
with water-ECS intervals as long as 5 hours.) Since this
appetitive learning situation appears to involve no CER
properties, the previous interpretations of shock-avoidance
learning RA in terms of an interaction between shock and
16
ECS do not apply. •
From the preceding review of the data and inter
pretations of retrograde amnesia investigations several
conclusions can be drawn:
1. Although there is some evidence to support a
"conditioned inhibition" interpretation of some of the
investigations utilizing multiple ECS's and/or active
avoidance learning situations, a more parsimonious inter
pretation of the large body of data from RA experiments can
be made in terms of the perserveration-consolidation
hypothesis.
2. Since multiple (or single) ECS's combined with
an active avoidance learning situation may confound
aversive and amnesic properties of ECS, more direct evi
dence of the effect of amnesic agents on short-term
memory can be gained from studies involving one passive
avoidance learning trial and one administration of the
amnesic agent.
3. Because of the possible interaction of shock
and ECS, and the possibility that ECS may differentially
affect memory for the emotional and the response compo
nents of the learning experience, situations involving
appetitive learning and amnesic agents other than ECS
: (such as ether) might well result in less controversial
I
i results than many of the experiments cited above.
4. Great differences exist among the apparatuses,
! shock levels, ECS intensities and durations, etherization
; durations and temperatures, subjects, and other general
procedural variables in the many investigations of retro
grade amnesia. An experimental design which utilizes
i ECS- and ether-produced retrograde amnesia for both appe
titive and aversive one-trial learning in the same
experimental setting should resolve some of the discrepen-
cies among different studies and may also rule out some
of the alternative explanations of retrograde amnesia data
The purpose of the present series of investiga
tions is to examine the effects of ether and ECS on one-
trial appetitive and aversive learning in the same
experimental situation. Since previous studies have
utilized different apparatuses for investigating the two
i different types of learning little data exists concerning
i
; the similarities and differences in the manner in which
i
j memories for positively and negatively reinforced habits
| are affected by amnestic agents. In addition, since there
1 does not appear to be any evidence that ether is aversive,
i any interference produced can be attributed to its effects
; 18
: on learning rather than on performance as has been sug-
i
gested with ECS. (On the basis of previous experiments,
; ether, although expected to produce RA, should be less
i
■ effective than ECS.) Multiple dependent measures will be
utilized in the present investigations in order that addi
tional information can be gained concerning the pervasive
ness of the amnesic effect. Should the measures be
j differentially affected by the amnestic agents, explicit
i
hypotheses concerning the consolidation of the memories
for various experiences can be made.
The design of the present investigation is such
that the predicted results can only support the persevera-
: tion-consolidation hypothesis. Neither the "conditioned
inhibition hypothesis" of Lewis and Maher (1965) nor the
conditioned fear hypothesis of Coons and Miller (1960) can
be supported if retrograde amnesia results in all situa
tions. Both of the latter hypotheses would predict
interference with approach responses, and therefore no RA
I in the shock-ECS situation, although Lewis and Maher (1965)
| do suggest that fear may reappear if emotional responsive-
I ness has been reduced. Since no apparent CER is expected
: to develop in the appetitive paradigm, interference with
| emotional responding can not be invoked to explain
19
approach to the place where ECSed subjects previously
found water. If ECS produces fear (Coons and Miller) or
reduces activity (Lewis and Maher) subjects convulsed as
long as an hour after the appetitive experience should
reduce their performance as much as animals ECSed immedi
ately after the appetitive tiral. Graded effects, such as
those predicted by the perseveration-consolidation hypoth- j
esis, should rule out this interpretation.
CHAPTER II
EXPERIMENT 1
The purpose of the first experiment was to deter
mine the effect of electroconvulsive shock and ether
administered at varying intervals after a foot-shock in a
one-trial passive avoidance learning situation. Subjects
were placed in a narrow start-box (see Figure 1) and when
the door was opened were allowed to enter a larger chamber.
Directly opposite the entrance to this chamber was a small
opening leading to a brightly lighted cul-de-sac. After
slight hesitation animals usually approached the cul,
poked their heads into it, and contacted the empty drink
ing cup recessed in the floor. Upon touching the cup
subjects received a painful shock resulting in withdrawal
from the cul, and removal from the apparatus 60 seconds
I
later. Animals in different groups were administered
ether or ECS either immediately after removal or 15 minutes
later. All subjects were tested for retention of the
shock experience 24 hours later.
20
21
' n f r t f ' - V A
Efiaasa
22
Method
! Subjects
Two hundred thirty-nine male Swiss-Webster albino
mice, 60 days old (30 - 40 gms.) were used in this study.
The subjects were obtained from a commercial vendor
(Simonson Laboratories, Gilroy, California) in batches of
120 and 160 animals each. Upon receipt subjects were 1
ear-punched and assigned to group cages, each containing
12-14 animals. All subjects were placed on a 23-hour water
deprivation schedule following housing, and were allowed
access to a water bottle (one per cage) one hour per day
for 5 days prior to the initiation of experimentation.
Dry food pellets were available at all times.
Apparatus
The apparatus (see Figure 1) consisted of a 10.8
cm. x 10.8 cm. x 6.1 cm. high translucent polyvinyl cham-
; ber. Centered on one wall of the chamber, 10 mm. above
I the base, was a 2.7 cm. x 2.2 cm. opening to which was
| attached a 4.5 cm. deep cul-de-sac which was just large
enough to permit the insertion of a mouse's head. A
!
i photocell detection circuit was placed 1.0 cm. from the
. to detect the head-pokes (photo cell - GE#A33). Throughout
the experiment an empty 1.3 cm. diameter stainless steel
i
j drinking cup, 1 cm. deep, protruded 3 mm. into the cul at
the center of the cul floor. A start box, 11.1 cm. long
x 2.2 cm. wide x 6.1 cm. high) was attached to the center
: of the wall opposite the cul. It was separated from the j
experimental chamber by an opaque guillotine door which
operated almost inaudibly. The chamber was set on a sheet
aluminum floor which was connected to one side of a
Grason-Stadler Drinkometer, the other side of which was
attached to the cup to allow the recording of licking. An
Applegate Constant Current shocker was also attached to
the aluminum floor and to the cup by means of a relay
circuit which permitted bypassing the Drinkometer when the
shock source was activated. Two ma., d.c. current was
delivered to animals in the appropriate groups when the
cup was contacted. Both the photocell and Drinkometer
circuits were attached to an event recorder so that latency
to the first head-poke and cup touch, frequency and dura
tion of all head-poke and cup touch sessions, and frequency
of discrete licks could be recorded.
Ether anesthesia was induced in a 1000 ml. glass
container into which subjects in the appropriate groups
24
were placed until they had reached respiratory arrest (or,
in some cases, the stage of faint breathing which just
preceded it). The chamber rested in a water bath which
was maintained at 30-35° C. Cotton at the bottom of the
chamber, separated from the subjects by a hardware-cloth
platform, was saturated with diethyl ether prior to the j
introduction of the first subject and periodically resatu
rated after each few subjects had been etherized.
Electroconvulsive shock was induced by passing
770 v. (60 cycle sine wave, 18.5 ma.) through the animal
via ear clips (covered with electrode paste) for 800 msec.
All etherized and ECSed subjects were artificially respi- i
i
t
rated with a small plastic squeeze bottle until breathing
appeared stable.
Procedure
The subjects from each batch were assigned to
8 groups, each of which ultimately contained 22 subjects.
Initially, assignment to groups was random (i.e., each
subject's group was determined by the order in which its
number appeared in a table of random numbers). However,
the high mortality rate which resulted from the amnesic
treatment (37% with ether and 18% with ECS) necessitated
25
the institution of haphazard assignment in most of the
later batches. In three of the five batches used in this
experiment subjects were assigned to all of the groups.
Procedural difficulties made it necessary to run only half
of the groups in the two earliest batches, although all
baseline control groups were represented in these two
batches.
This study was concerned with the effects of ECS
and ether on one-trial, passive avoidance learning, and
the groups used were as follows:
1. SN— the baseline avoidance learning group which
received at least one mouth shock and no addi
tional treatment.
2. SE^— treated the same as group SN, but ether
ized 1 minute after receiving an initial shock
from the drinking cup.
3. SECS^— same as SE^, but received ECS at one
minute. !
4. SE^g— same as SE^, but received ether at 15
minutes.
5. SECS^g— same as SE^, but received ECS at 15
minutes.
26
6. NN~-the baseline control group for the avoid
ance learning group. Subjects were in the
experimental chamber for 60 seconds following
initial contact of the drinking cup, but
received no shock or other treatment.
7. NE— same as group NN except that subjects were
I
etherized after removal from the apparatus to j
assess the possible aversive effects of ether
ization.
8. NECS— same as group NE but received ECS after
removal from the apparatus to assess the
possible aversive effects of ECS.
(Since the control procedures for this and the
following study were identical, and since both studies
were run concurrently, groups NN, NE, and NECS served as
the control groups in both investigations.
Subjects in group SN were placed in the startbox,
the door raised, and the cumulative recorder started when
I
the animal was oriented toward the experimental chamber.
(In cases where the subject turned around after being
placed in the startbox, the door remained closed until
the animal was properly oriented. If the subject persisted
in facing away from the door, it was removed and replaced
with the proper orientation.) The door was closed gently
after the animal had left the start box. Subjects assigned
to shock groups (SN, SE , SECS , SE , and SECS )
■ 1 1 15 15
received a 2 raa. shock when the empty drinking cup was
contacted. Shock group animals received additional shocks j
!
if they attempted to contact the cup again during the j
remainder of the 60 second period which followed the
initial shock. At the end of this one minute period they
were removed from the chamber. (Approximately half of the
subjects in each of the groups received only one shock and
41 of the remaining 44 animals were administered two
shocks. Of the other three subjects, two received three \
i
j
and the last, 4 shocks.) Animals in the immediate ether
and ECS groups (SE^ and SECS^) were administered the
appropriate amnesic agent within 25 seconds of their
removal from the apparatus (85 seconds after the initial
shock) and, after recovery (usually within 35-9 0 seconds),
were returned to their home cages. Subjects in the 15
i
minute ECS and ether groups (SE^ and SECS^) were placed j
in holding cages, three or four to a cage, during the time
between their removal from the apparatus and administration
i
of the appropriate agent. j
Animals in the control groups (NN, NE, and NECS)
28
were introduced into the apparatus and treated in the same
manner as the shock groups, except that the shock source
was disconnected. These subjects were allowed 60 seconds
in the chamber following their initial contact with the cup
and were then removed. Group NN animals were then returned
to their home cages while group NE and NECS subjects were
etherized or convulsed in the same manner as the experi
mental groups. Subjects failing to contact the drinking
cup within the first 60 seconds in the chamber were dis
carded from the experiment. (Seven subjects were elimi
nated for this reason.)
One hour after the last subject in each cage had
been run, one hour's access to water was allowed. (The
duration of the watering period was based on the observa
tion that all of the animals in a cage appeared to have
drunk at least once within this time and at the end of an
hour little or no activity was observed at the drinking
spout.)
I
The subjects were re-introduced into the start box j
i
24 hours after the experimental session and were allowed
access to the chamber for a three-minute test period.
During the test session the following measures were
recorded:
29
1. Latency to First Head Poke—-this consisted of
the amount of time taken from the opening of
the start box door until the photocell beam
across the opening of the cul was broken.
2. Latency to First Cup Touch— this was the amount
i
j
of time required from opening of the start box j
door to the first time that the drinking cup j
was contacted.
3. Frequency of Licks— this measure consisted of
the total number of discrete licks recorded by
the Drinkometer.
4. Frequency of Head Pokes— this was the number j
of times that the photocell beam was crossed
by the animal.
5. Frequency of Cup Touch Sessions— one session
consisted of at least one cup touch during a
head poke.
j
6. Head Poke Duration— this was the total amount j
i
of time spent with the photocell beam broken, j
7. Cup Touch Session Duration— this measure con- j
sisted of the amount of time spent from the
first to the last cup touch within each head
poke session, totaled for all of the Cup Touch
Sessions on the test day. Two additional
measures were derived from the latency scores.
8. Head Poke Latency Difference Score— this con
sisted of the difference between the Latency
to Cup Touch on the experimental treatment day
and the same measure taken on the test day.
9. Cup Touch Latency difference Score— this score j
was analogous to that of the previous measure,
but was based on initial Cup Touch Latencies
on the first and second days of the experiment.
Results
Since most of the subjects in the baseline shock
group (SN) and in the etherized groups (SE^ and SE^,.)
failed to explore the cul within the three-minute test
period (no score on any of the frequency or duration
measures could be recorded) these animals were arbitrarily
assigned latency scores of 180 seconds, the total duration
I
of the test session. For this reason nonparametric j
|
analyses were used throughout this investigation.
Kruskal-Wallis one-way analyses of variance, performed
across all groups for the treatment day Latencies to Head
Poke and to Cup Touch, yielded non-significant values.
31
This indicated that the 8 groups were comparable prior to
the introduction of the experimental treatments. (Means
and medians for all groups on all measures are presented
in Appendix A.) In addition, an analysis of variance
across the 5 batches of animals, disregarding treatment
group, also yielded a non-significant H-value, suggesting j
that the different batches of animals were comparable.
Additional analyses of variance comparing the
8 groups on the 9 independent measures all yielded H-values
significant beyond the .001 level, indicating that the
treatments were successful in separating the originally
comparable groups. Because the pattern of group compari
sons (Mann-Whitney U tests) differed with the various
dependent measures, each variable will be considered
separately.
Latency to First Head Poke—
H = 76.41, p < .001
On the test day the difference between the baseline
shock group (SN, X = 134.54, mdn. = 180.0; see Figure 2^) j
and the non-shocked control group (NN, x = 22.39,
1-For this and all subsequent graphs the top of
each bar represents the group mean while the solid line
crossing the bar represents the group median.
Latency to First Head Poke (seconds)
32
SN SEX SECS;l SE15 SECS15 NN NE NECS
Groups
Figure 2.— Mean (and median) latency
to first head poke on test day. For this and
all of the following bar graphs, the vertical
bars represent the group means and the hori
zontal bars denote group medians.
33
mdn. = 12.0) was highly significant (U = 37%, p< .00006),
indicating that the latency to the First Head Poke was
reliably longer for the shocked, as compared with the
control subjects. (Approximately 2/3 of the subjects in
groups SN, SE , and SE did not head poke on the test j
1 15 j
day, and approximately 3/4 of the animals in the same I
groups did not touch the drinking cup in this session.)
Both groups SECS^ (x = 73.68, mdn. = 62.0) and SECS15 (x =
95.04, mdn. = 106.25) had significantly shorter latencies
than the shock control group (U's = 109%, p < .002, 154%,
p< .042, respectively). However, both of these groups gave
evidence of significantly longer latencies than group NN j
'
(U's = 83, p< .0004 and 38, p< .00006, respectively). The
immediate ECS group (SECS1) had a significantly shorter
latency than group SECS^^ (U = 154%, p< .042), Neither of
the etherized groups differed from group SN on this or
subsequent measures. However, there was no difference
I
I
between groups SE^ and SECS^,-. There was no significant j
difference between the ECSed control group (NECS) and
group NN, nor between groups NE and NN, but the difference
between group NECS and the etherized control group (NE)
was significance (U = 157%, p< .05).
34
Head Poke Latency Difference Score—
H = 70.52, p < .001
The pattern of results with this measure was the
same as that observed with the previous measure, with the
exception that there was no difference betwen the immedi-
i
ate and 15 minute ECS groups (see Figure 3). The differ
ence between the baseline shock group, SN (x = 117.02, j
mdn. = 155.75), and the non-shocked control group (x = 4.07,
mdn. = -2.0 was highly significant (U =24, p <.00006).
Both groups SECS1 (x = 61.30, mdn. = 45.5) and SECS15
(x = 80.5, mdn. = 88.5) differed significantly from group
SN (U's = 121, p< .003, and 1593s, p< .026, one-tailed test,
respectively). However, both groups SECS^ and SECS^^
differed significantly from group NN (U's = 80, p< .0002
and 593s, p<.00006, respectively). The difference between
the two ECS groups did not approach significance. Althoughj
there was no difference between group NN and the ECSed
control group (NECS), group NE was significantly different j
. i
j
from group NN (U = 135, p < .0 2). Neither of the etherized j
!
groups (SE^ or SE^,.) differed significantly from the shock
control group (SN). However, group SE^ also was no differ
ent from SECS^,_.
Head Poke Latency Difference Score (seconds)
35
168 A
144 ■
120 -
96 -
72
48 -
24
0 H
-8
m
i SN SE-j^ SECS-j^ SE-^ SECS15 NN NE NECS
Groups
Figure 3.— Mean (and median) head poke
latency difference score (day 1 - day 2).
Latency to First Cup Touch—
H = 95.88, p < .001
As with the other measures group SN, the baseline
shock group (x = 160.15, mdn. = 180.0, see Figure 4),
differed from the non-shocked control group, NN (x =35.86
mdn. = 24.25), at a highly significant level (U = 74,
p <.0001). While the difference between group SN and the
immediate ECS group (SECS-^, x = 120.46, mdn. 136.0)
approached significance (U = 187, p <.099, one-tailed
test) , the difference between groups SN and SECS-j^ (x =
144.46, mdn. = 180.0) did not. Group SECS^, however, had
a significantly shorter latency than group SECS^,. (U = 160
p < .027, one-tailed test). Group SE^ (x = 148.14, mdn. =
180.0) did not differ from group SECS^, the only case with
all the measures where the immediate ether group did not
differ significantly from the immediate ECS group. All of
these experimental groups differed significantly from the
non-shocked control group (NN). None of the differences
among the control groups (NN, NE, and NECS) approached
significance. The difference between groups SECS^ and
SECS^5 was significant (U = 160, p <.027, one-tailed test)
However, group SECS^ did not differ from group SE-^.
Latency to First Cup Touch (seconds)
37
192 i
168 -
144 -
120 -
96 -
72 -
48 -
24 -
0 - I
SN SE-l SECS-l SE15 SECS15 NN NE NECS
Groups
Figure 4.— Mean (and median) latency to
first cup touch on test day.
38
Cup Touch Latency Difference Score—
H = 78.60, p < .001
Consistent with all of the other measures, the
difference between the baseline shock group, SN (x =
136.48, mdn. = 151.59), gave evidence of a reliably greater
difference score than group NN (x = 13.39, mdn. = 3.75;
U = 83, p< .0003, see Figure 5). The immediate ECS group,
f
SECS^ (x = 104.66, mdn. = 112.5) differed significantly
i both from group SN (U = 122, p < .005) and from group NN
(U = 53, p < .0002) . Group SECS^^ did not differ from
group SN, but the difference between the former and group
SECS^ approached significance (U = 181, p< .06, one-tailed
test). However, group SECS^ also did not differ from
group SE-^. There were no significant differences among
the three non-shocked control groups (NN, NE, and NECS).
Frequency of Licks—
H = 108.92, p < .001
The difference between group SN (x = 1.32, mdn. =
0.0, see Figure 6) and the group NN (x = 15.41, mdn. =
11.5) was again highly significant (U = 32^, p< .00006). j
: As with the other measures, the difference between groups j
i
; SECSj (x = 4.91, mdn. = 3.0) and SN was significant (U =
j 92.5, p < .0005), as was the difference between the former
I and group NN (U = 103, p < .002). With this measure group
39
168 H
144 -
t o
'd
a
o
u
a j
03
u 120
o
o
w
c u
o
c
03
U
a >
m
iw
•H
Q
O
fl
( I )
4-1
< 0
Hi
rC
o
P
o
ft 24
P
o
0 - I
96
72 -
48 -
SN SE, SECS SECS,
! SE1 5 15
Groups
NN NE NECS
Figure 5.— Mean (and median) cup touch
latency difference score (day 1-day 2).
40
16
12 -
w
M
U
-H
m
o
> i
0
g
a >
P
01
a )
n
P m
8 -
SN SE-j^ SECS! SE-^ SECS^
Groups
NN ' NE NECS
Figure 6.— Mean (and median) frequency
of licks on test day.
; 41
; SECS^g (x = .73, mdn. = 0.0) did not differ reliably from
j the baseline shock group (SN), although it did differ from
group SECS^ (U = 98, p < .001) . Group SECS15 also did not
differ from group SE^. There were no differences among
the three non-shocked control groups (NN, NE, and NECS),
all of which differed significantly from the baseline
■ shock group (SN).
Frequency of Head Pokes—
H = 95.07, p < .001
Following the consistent trend within the other
measures, the difference between the shocked baseline
group, SN (x = 1.27, mdn. = 0.0, see Figure 7), and the
non-shocked control group, NN (x = 7.5, mdn. = 7.0), was
highly significant (U = 18, p< .00006). Both the imme
diate ECS group, SECS^ (x = 3.59, mdn. = 3.0) and the
15 minute ECS group, SECS-^ (x = 1.68, mdn. = 1.5) differed
reliably from group SN (U's 95, p< .0006 and 168, p < .041,
: one-tailed test, respectively), while group SECS-^ gave
evidence of significantly more Head Pokes than group
| SECS^g (U = 13433, p< .012) and group NN (U = 84, p <
I
I .00014). However, group SECS did not differ from group
15
! SE^. There were no differences among the three non-
shocked control groups (NN, NE, and NECS).
42
6 -
c o
CD
M
O
P h
c d
(D
«
o
>i
o
f l
0 )
s
D1
CD
U
4 '
2 ‘
0 -
: ■
SN SE j SECS! SE15 SECS25 NN NE NECS
Groups
Figure 7.— Mean (and median) frequency
of head pokes on test day.
43
Frequency of Cup Touch Sessions—
H = 10:5.04, p < .001
As with the other measures the shocked baseline
group, SN (x = .45, mdn. =0.0, see Figure 8) gave evidence
of reliably fewer Cup Touch Sessions than either the non
shocked control group, NN (3.41, mdn. = 4.0) or the imme-
I
diate ECS group, SECS, (x = 1.41, mdn. = 1.0; U's = 40, I
X \
p <.00006, and 118, p< .006, respectively). Group SECS^
also differed reliably from group NN (U = 89, p< .0004).
Although the median of group SECS^,- (x = .45, mdn. = 0.0)
was no different from that of group SN, it was signifi
cantly below that of group SECS^ (U = 123%, p < .006).
However, group SECS^5 did not differ from group SE^.
There were no significant differences among the three non
shocked control groups (NN, NE, and NECS).
Duration of Head Poking—
H = 105.09, p < .001
In addition to its difference from group NN (x = j
18.46, mdn. = 18.75, see Figure 9), the shocked baseline 1
group, SN (x = 1.80, mdn. =0.0, U = 18, p < .00006)
differed significantly from both group SECS^ (x = 5.36,
mdn. =s 4.12, U = 85%, p < .00032) and group SECS._ (x =
15
1.82, mdn. = 1.0; U = 158, p <.026, one-tailed test).
44
Groups
Figure 8.— Mean (and median) frequency
of cup touch sessions on test day.
I
j
45
20 -
NN NE NECS
Groups
Figure 9.— Mean (and median) duration
of head poking on test day.
46
Both of the ECS groups, SECS., and SECS.._, also differed
- L
reliably from the non-shocked control group, NN (U's =
65, p< .00006, and 30^, p< .00006, respectively). In
addition, group SECS^ spent significantly more time Head
Poking than group SECS15 (U = 110, p < .002). There were
no reliable differences among the three non-shocked con
trol groups.
Duration of Cup Touch Sessions—
H = 106.30, p < .001
As with all of the preceding measures the shocked
baseline group, SN (x = .89, mdn. = 0.0, see Figure 10),
differed reliably from the non-shocked control group, NN
(x = 6.98, mdn. = 5.75; U = 37, p< .00006). In addition,
the immediate ECS group, SECS^ (x = 1.78, mdn. = 1.25)
differed significantly from both groups SN and NN (U's =
109, p< .002, and 82, p< .0002, respectively). There was
no difference between group SN and the 15 minute ECS
group, SECS^(-, while the difference between the latter
group and group SECS was significant (U = 110, p< .002).
Group SECS^ did not differ from group SE-^. There were no
reliable differences among the three non-shocked control
groups (NN, NE, and NECS).
Duration of Cup Touch Sessions (seconds)
47
7.5 -
5.0 -
2.5
0 -
1
SN SE-j^ SECS1 SE1S SECS25
Groups
NN NE NECS
Figure 10.— Mean (and median) duration
of cup touch sessions on test day.
48
Discussion
The most consistent finding in the present study
was that the shock experience on the first experimental
day was sufficient to produce one-trial, passive avoidance
learning with all of the measures utilized, i.e., highly j
significant differences between the baseline shock group j
and the non-shocked control group. In all 9 of the
dependent measures the difference between these two groups
was well beyond the .0003 level of significance, indicat
ing that the shock experience successfully differentiated
the originally comparable groups.
In all but one of the measures (First Latency to
Cup Touch), electroconvulsive shock administered to ani
mals within 85 seconds of the punishing mouth shock (or
shocks) resulted in significant decreases in latency
measures and increases in frequency and duration measures,
indicating that ECS had effectively attenuated the effects
of shock. Although ECS greatly interfered with the j
I
establishment of the passive avoidance habit, it did not
lead to nearly complete attenuation, as in the following
experiment where immediately ECSed subjects were indis
tinguishable from control levels of response on several of
the measures. With all of the measures in the present
49
investigation the shocked, immediately convulsed animals
differed from the non-shocked controls at highly signifi
cant levels, indicating partial retention of the avoidance
; habit. In addition to producing a strong group effect, i
. . |
immediate ECS following the shock-avoidance trial resulted j
l
: in 20 of the 22 subjects in that group spending some time
exploring the cul on the test day. This was done by only j
j
9 of the 22 subjects in the baseline shock group, SN.
Since there were no differences between the non-
shocked control group (NN) and the ECSed control group
(NECS) with any of the measures, the effectiveness of
immediate ECS in attenuating the shock experience can not
be attributed to performance changes brought about by
electroconvulsive shock. If the effectiveness of ECS was
the result of performance changes alone (such as would be
! reflected by decreased latencies and increased frequency
and duration indexes in ECSed subjects), the effect would
i
be detectable in convulsed control animals. j
: , i
With the exception of the First Latency to Cup j
i
i
; Touch measure, which failed to show an effect, the attenu- j
' I
: j
| ation resulting from immediate ECS was consistent and j
\ ‘ |
relatively uniform across all of the measures. This
| finding suggests that the administration of
electroconvulsive shock within 8 5 seconds of a shock
avoidance trial interferes with most of the response
components associated with the aversive experience. The
failure to obtain reliable attenuation of the shock
effects with the one latency measure is difficult to
assess. The act of traversing the distance from the
startbox to the drinking cup was further removed in time
and space from the mouth shock than such acts as licking.
From this one might surmise that the effects of ECS are
stronger on events the more proximate they are to the
aversive experience. However, Latency to the First Head
Poke involved events further removed temporally and
spatially from the aversive experience than did Latency to
the First Cup Touch, and nevertheless showed a significant
ECS effect. This argues against any suggestion that the
attenuating effects of ECS are greatest on responses most
closely associated with reinforcement, an hypothesis which
appears to be supported by the results of the following
: experiment. Since the Cup Touch Latency Difference Score,
which was based on the difference between the treatment
| and test days1 latencies, was interfered with reliably by
immediate administration of ECS, the possibility remains
| that the failure to obtain significant results only with
; 51
the test day latency resulted from some peculiarity of
the relationship between the latencies for the two differ
ent sessions.
The pattern of results with ECS administered
15 minutes after the shock avoidance experience, although
less striking than those obtained with immediate ECS,
supported an interference interpretation with all of the
measures associated with head poking. ECS administered
15 minutes after the shock trial resulted in significantly
briefer First Head Poke Latency, decreased Head Poke
Latency Difference Score measures, and larger Head Poke
Frequency and Duration scores than were observed in the
baseline shock group, SN. The differences between these
groups ranged between the .025 and the .041 levels of
significance, considerably less reliable than the differ
ences obtained with immediate ECS. Further support for
the observation that electroconvulsive shock given 15 min
utes after the learning trial was less effective than
i immediate ECS is the finding that, with all measures
: except the Head Poke Latency Difference Score, the imme-
| diate group evidenced a reliably greater degree of
attenuation than the 15 minute group.
] Although the means, but not the medians, for most
52
of the measures suggest a slight attenuation with immediate
administration of ether, none of the comparisons between
the immediate ether group, SE^, and the baseline shock
group, SN, approached significance. This would appear to
indicate that immediate ether had no interfering effect
on the shock avoidance experience. However, there were
also no reliable differences between the immediate ether
group and the 15 minute ECS group, SECS^j-, on any of the
measures, despite the fact that all of the measures
related to Head Poking showed an attenuation effect with
15 minute ECS. This finding serves only to support the
trend noted in the means for the immediate ether group,
i.e., that with administration of ether within 85 seconds
of the shock experience a non-significant trend toward
attenuation is observed.
With etherization 15 minutes after receipt of
shock not even a trend toward interference was observed.
In fact, if any trend exists among the means of the j
measures for group SE15, it is in the opposite direction.
In every case the mean of the 15 minute ether group tended
toward the direction of facilitation of learning of the
passive avoidance response. The means for the latency
measures were larger, the difference scores greater, and
the frequency and duration scores smaller than those
obtained with the baseline shock group, SN, suggesting
that shock produced greater changes in behavior when fol
lowed 15 minutes later by ether than if no ether is
administered. However, it should be noted that none of
these differences approached statistical significance.
Although there was little or no effect on learning
with ether, there were several measures on which the
etherized and non-etherized control groups differed. This
effect was most marked with Head Poke Latency Difference
score, where the etherized control group gave evidence of
significantly greater latencies on the test day, (as
compared to the treatment day), than the normal control
group. With the Head Poke Frequency and Duration measures,
the differences, although not reaching the 10% level of
significance ( p <.125 with two-tailed tests), were in the
same direction, i.e., etherized controls gave lower scores
on both measures. One interpretation of these findings is
j
that the failure to obtain attenuation of the shock- I
t
avoidance experience with ether is in part due to the
performance effects of this agent which appear to have
worked against the predicted outcome. However, since the
apparent performance effect of ether was slight, reaching
54
significance only with one measure, this explanation does
not appear sufficient to account for the failure of imme
diate etherization to produce any interference with
’ learning.
The results of the present experiment support the
predictions only with electroconvulsive shock. Although
it was anticipated that administration of ECS immediately
(within 85 seconds) following the shock-avoidance experi-
; ence would result in more complete attenuation, i.e.,
result in no differences between the immediately ECSed and
control subjects, immediate ECS did produce a reliable
degree of interference with learning as indexed by the
: significant differences between group SECS^ and the base
line shock group on 8 of the 9 measures. Since pilot work
indicated that ECS administered 15 minutes after the
learning trial produced no evidence of attenuation, it
was decided not to run additional groups with longer
shock-ECS intervals. The results of the present investi-
i gation, however, indicate that more than 15 minutes must
intervene between shock and ECS to insure that the af.ter-
I effects of the learning experience are impervious to
! attenuation by electroconvulsive shock. Although only two
points in time were used, the results appear to support
55
the predicted graded effect. Immediate ECS produced
attenuation with 8 of the 9 measures taken while 15 minute
ECS resulted in interference on only 4 of these measures.
In addition, the significance levels obtained as a result
j
j
of immediate ECS (i.e., for comparisons between groups
SECS-^ and SN) were much more reliable than those resulting :
from comparisons between the baseline shock group and the j
15 minute ECS group.
The results obtained with ether in the present
investigation are somewhat surprising in view of the out
comes of several previous investigations in which rela
tively complete attenuation of a one-trial, passive
avoidance experience was obtained (e.g., Herz, Peeke, and
Wyers, 1966; Pearlman, 1966). The absence of any clear
evidence for ether-produced interference is open to at
least two interpretations:
1. It is possible that factors related to the
apparatus or procedure in the present study j
preclude any demonstration of attenuation with i
ether.
2. The results of this experiment may lead to the
conclusion that the interference produced by
ether in other investigations may be an artifact
56
of procedures or apparatuses used in such
previous studies.
Although the first interpretation appears to be the most
tenable, consideration of both possibilities will be under
taken in the final discussion following the presentation
i
of the remaining two studies.
CHAPTER III
EXPERIMENT 2
The purpose of this experiment was to examine the
effects of electroconvulsive shock and ether on one-trial
appetitive learning within the framework of an investiga
tion which replicated, as closely as possible, the condi
tions of the first experiment. Subjects were placed in
the same apparatus as was used in the previous study and
were allowed 60 seconds access to water in a drinking cup
following the initial contact with the cup. Animals in
different groups were administered ether or ECS immedi
ately, 15, 30, or 60 minutes after removal from the
apparatus and tested for retention of the habit 24 hours
later.
Method
Subjects
Three hundred sixty-four male Swiss-Webster albino
mice, part of the batches described in the first
57
experiment, were used in this study. Subjects were
; handled, housed, and deprived in the same fashion and at
the same time as the animals in Experiment 1.
Apparatus
The apparatus was the same as that utilized in the
previous experiment.
Procedure
The subjects utilized in the present study were
run at the same time as those of the first study. In most
cases subjects from each cage in each batch were assigned
to the groups of both Experiment 1 and Experiment 2. The
animals from each batch were assigned to 12 groups, each
of which ultimately contained 22 subjects. Initially,
assignment to groups was random (i.e., each subject's
group was determined by the order in which its number
appeared in a table of random numbers). However, the high
mortality rate which resulted from the amnesic treatments
(33% with ether and 17% with ECS), necessitated the
| institution of haphazard assignment in most of the later
; batches. In three of the five batches used, subjects were
assigned all of the groups. Procedural difficulties made
i it necessary to run only two of the four experimental
59
groups in the two earliest batches, although the baseline
control groups were represented in both of these batches.
The groups used were as follows:
1. WN-.-the baseline appetitive learning group
which was allowed free access to water from
the drinking cup and received no further treat
ment .
2. WE^— treated the same as group WN but etherized
1 minute after initial contact with the drink
ing cup.
3. WECS^— same as WE^, but received ECS at 1
minute.
4. W E — same as WE , but received ether at 15
15 1
minutes.
5. WECS^j.— same as WE^, but received ECS at 15
minutes.
6. WE3q— same as WE^, but received ether at 30
minutes.
7. WECS3q— same as WE^, but received ECS at 30
minutes.
8. WE(jo— same as WE^f but received ether at 60
minutes.
60
9. WECSgQ— same as WE-^, but received ECS at 60
minutes.
10. NN— the baseline control group for the appe
titive learning group. Subjects were in the
experimental chamber for 60 seconds following
initial contact of the empty drinking cup,
but received no further treatment.
11. NE— same as group NN except that subjects
were etherized after removal from the appara
tus, to assess the possible aversive effects
of etherization.
12. NECS— same as group NE, but received ECS
after removal from the apparatus, to assess
the possible aversive effects of ECS.
(As noted above, since Experiments 1 and 2 were
run concurrently and since the control procedures for both
studies were the same, groups NN, NE, and NECS served as
control groups in both investigations.)
After subjects in group WN were introduced into
the startbox, the door was raised and the cumulative
recorder was started when the subject was oriented toward
the door. The door was closed gently after the subject
had left the startbox. Each subject was allowed access
61
to the water in the drinking cup for 60 seconds after its
first contact with the cup. (During this interval an
average of 35-40 seconds were spent drinking and approxi
mately 3/4 c.c. of water consumed.) At the completion of
: this period, the subject was removed from the apparatus
and returned to its home cage as in Experiment 1. Latency
to first head poke (breaking of the photocell beam),
latency to the first drinking cup touch, frequency and
duration of head pokes and cup touches, licks, and total
duration of drinking were all recorded automatically on
the event recorder. Subjects in the other groups which
received water (i.e., WE -WE,.„ and WECS..-WECS^.„) were
1 60 1 60
treated in the same fashion, but received amnesic treat
ment before being returned to their home cages. Animals
in group WE-^ were removed from the apparatus after one
minute access to water and were placed in the etherization
chamber within 15 seconds. Subjects in the other etheri
zation groups were placed in holding cages (3 or 4 to a
cage) until the appropriate time interval had passed and
were then placed in the etherization chamber. After
| recovery from ether (usually 60-90 seconds) animals were
' returned to their home cages. Subjects in group WECS-^
i
| were removed from the apparatus, had ear clips applied,
62
and were convulsed within 20 to 25 seconds. Animals in
the other ECS groups were placed in the holding cages
until administration of seizure and were returned to their
home cages after recovery (35 to 45 seconds). Subjects in
groups NN, NE, and NECS were introduced into the apparatus
in the same manner as in Experiment 1. (A different
drinking cup was used for all subjects which did not
receive water to eliminate the possibility of small drops
of water being present). Subjects in these groups were
left in the chamber for 60 seconds following their first
contact with the empty drinking cup, making the time they
spent in the apparatus equivalent to that of the water-
reinforced groups. Subjects were then removed from the
chamber and returned to their home cages. Animals in
groups NE and NECS were administered ether or ECS follow
ing their removal from the apparatus and were returned to
their home cages after recovery from treatment. Any sub
ject that did not contact the drinking cup within 60 j
i
I
seconds of opening of the startbox door was discarded from j
the experiment. (Nine subjects were eliminated for this j
I
reason.)
One hour after the last subject in each cage had
been run, one hour's access to water was allowed. (The
63
animals in this experiment were run under the same condi
tions of water deprivation as those of Experiment 1
so that the same control groups, NN, NE, and NECS, could
be used in both studies. Comparison of the groups used
in the present study with those of Experiment 1 indicated j
that the Day 1 Head Poke and Cup Touch Latencies were '
comparable for all groups, i.e., the F-ratios were non- !
significant.) j
!
Twenty-four hours after the first experimental
session (approximately 21 to 22 hours after the end of the
watering period) subjects were reintroduced into the
startbox and allowed access to the chamber for a three
minute period. The same measures were recorded as on the
previous day, but no water was available during this
session.
Results
In order to determine the comparability of the
12 groups utilized in this study prior to the analyses of
the dependent variables, over-al.1 analyses of variance
were performed on the initial latency measures to the
first head poke and to the first drinking cup touch, as in
the first study. For both measures the F-ratios were
smaller than that required at the 25% level of signifi
cance, indicating that the group mean latency measures
were well within the limits of one parent population.
Because this experiment was performed on 5 separate
batches of animals, pretreatment differences due to j
f
batches might also have existed. To examine this possi- !
I
: bility over-all analyses of variance were performed across
batches (disregarding treatment group) for the two initial
■ latency measures. For both of these analyses the result
ing F-ratios were considerably below that required at the
25% level of significance, suggesting comparability among
the different batches of subjects. The s2 /s2. tests
J max./ m m .
for homogeneity of variance performed within each of the
four comparability analyses indicated that in all but one
of the cases the ratios were well within the acceptable
range. Only in the case of the treatment groups compara
bility analyses for the first latency to drinking cup
touch was the acceptable level exceeded, and there only by
j r
t
; 20%. Parallel analyses performed with scores transformed j
! i
s to both square root and log scales yielded non-significant
i
! F-ratios. The analysis for which the siL„ /s2j„ ratio
j J max. m m .
| had exceeded the acceptable level was within acceptable
limits and resulted in non-significant F-ratios with both
65
transformations. An analysis of the amount of time spent
drinking by the reinforced groups (WN, WE^ WECS^ WE15,
WECS^gz WE^q/ WECS^q, WE5q^ an^ WECS60^ revealed no sig
nificant over-all difference further supporting the
observation that the groups were comparable.
i
Additional one-way analyses of variance comparing
the 12 groups on the 9 dependent measures all yielded i
i
I
F-ratios significant beyond the .005 level. Since the
2 2
/sm . ratios for most of the comparisons far exceeded
max. m m . ^
the acceptable level, parallel analyses were performed
with square root (and in some cases log) transformations.
The resulting F-ratios were strikingly similar and, in all
cases, significant at the same level, but extreme non-
homogeneity still existed. Although the assumption of
homogeneity of variance can be violated in analyses based
2 2
on equal subgroup sizes, some of the s* „ /si;. ratios in
iuaX • m m •
the present situation (even the transformed ones) exceeded
the acceptable level by factors of 10 to 20 times. It
therefore appeared more appropriate to use nonparametric
analyses. Parallel nonparametric analyses (Kruskal-Wallis
one-way analyses of variance for multiple group compari
sons and Mann-Whitney U tests between pairs of groups)
were therefore performed with all of the data and the
66
results compared with parametric analyses of variance and
;
; t-tests. Nonparametric tests revealed significant differ
ences in all cases where such differences had been detected
by parametric tests and in many cases where parametric ;
i
i
analyses had not revealed differences. In all cases the j
probability levels obtained with Mann-Whitney analyses
i
; were considerably lower than with t-tests, indicating that j
j
with the present data the distribution free analyses were
more powerful. It was decided, therefore, to present only
the results obtained with the nonparametric tests. ^
Since the pattern of group comparison (U tests)
differed with the various dependent measures, each variable
will be considered separately.
Latency to First Head Poke—
H = 40.43, p < .001
On the test day the difference between the water
reinforced baseline group, WN {x = 9.34, mdn. = 8.25, see
tabled means and medians in Appendices A and B) and the j
i non-reinforced control group, NN (x =22.39, mdn. = 12.0)
' was significant (U = 147, p< .026), suggesting that the
| ----------------------------------------------------------------------------
| -^Nonparametric analyses performed for all of the
above-mentioned comparability comparisons yielded similar,
| i.e., non-significant, results.
67
60 second access to the drinking cup on the first experi
mental day was sufficient to produce a latency to the first
head poke on the test day which was significantly shorter
than that evidenced by the non-reinforced control group
j
(NN). None of the experimental groups differed from the j
I
i
water-reinforced group (WN), suggesting that the amnesic
treatments following water reinforcement had no effect on i
head poke latency. While the non-reinforced ECS control
group, NECS (x = 13.48, mdn. = 11.0), did not differ from
group NN, it also did not differ from the reinforced con
trol group (WN), suggesting the possibility that ECS may
have slightly facilitated this latency measure. Although
the etherized, non-reinforced control group, NE (x = 35.66,
mdn. = 17.0), was not reliably different from group NN, it
was significantly slower than group NECS (U = 157%, p< .05),
raising the possibility that etherization may have had
some aversive quality which served to slow the subjects.
j
Head Poke Latency Difference Score— i
H = 25.36, p < .01 j
For this measure the principal comparison, i.e.,
between groups NN (x = 4.07, mdn. = -2.0) and WN (x = 2.09,
mdn. = 2.50) failed to reach significance, leading to the
conclusion that no one-trial learning was in evidence.
Although there was no difference between the NN and NECS
L groups with this dependent measure, the difference between
groups NECS and NE approached significance (U = 162, p <
; .07).
Latency to First Cup Touch—
H = 57.64, p < .001
With this latency measure the comparison between
, the non-reinforced control group, NN (5c = 35.86, mdn.
24.25, see Figure 11), and the reinforced group, WN
(5c = 12.25, mdn. = 12.25), was significant (U = 134%,
p <.012). There were no differences among the three non-
reinforced control groups (NN, NE, and NECS). With this
measure the amnesic agents applied shortly after rein
forcement successfully interfered with one-trial learning,
as evidenced by the finding that the latency for group WN
was significantly lower than the latency for both group
WE^ (U = 99, p < .001) and group WECS^ (U = 139, p< .016) .
. Further support for this conclusion is the observation
j that neither of these two groups (WE^ or WECS-^) , which did
; not differ from each other, was significantly different
|
j from any of the non-reinforced control groups (NN, NE, or
j
i NECS). Finally, of the remaining experimental groups,
j only WE-^5 (5c = 21.43, mdn. = 12.75) appears to be of
Latency to First Cup Touch (seconds)
60 ■
50 -
40 -
30 -
20
10 -
WN WE
i wecsx we15 wecsb we30 wecs^ q we60 wecs60
NN NE NECS
Groups
Figure 11.— Mean (and median) latency to first cup touch
on test day.
OS
VD
(
f
70
interest. The other groups differed from group NN and did
not differ from each other or from group WN. Group WE^5,
however, had a significantly higher latency than the
remaining groups (e.g., comparison of WE-^ and WECS-^
yielded U = 156%, p < .022) . Group WE-^ also did not
differ significantly from the non-reinforced control group
(NN) (U = 168%, p < .08).
Cup Touch Latency Difference Score—
H = 43.50, p < .001
The basic comparison between groups NN (x = 13.38,
mdn. = 3.75) and WN (x = -.41, mdn. = -.75) approached
significance (U - 177%, p< .066, one-tailed test). Group
WN did, however, differ from group NE (x = 43.02, mdn. =
10.0; TJ = 152, p< .035), although not from groups NN or
NECS. With this measure group WE-^ (x = 12.91, mdn. = 5.75)
had a significantly larger latency difference than group
WN (U = 144, p< „03). There were no differences among the
control groups (NN, NE, and NECS).
j
! Frequency of Licks—
| H = 105.06, p < .001
In terms of differentiation among the various
| experimental groups, this measure was the most sensitive.
| The baseline comparison between groups NN (x = 15.41,
71
mdn. = 11.5, see Figure 12) and WN (x = 61.96, mdn. 55.5)
was highly significant (U = 72%, p < .0001). There were no
. differences among the three non-reinforced control groups
(NN, NE, and NECS), all of which differed significantly
from the group WN. Both of the shortest delay amnesic
treatment groups (WE^, x = 31.91, mdn. = 30.0; WECS^,
x = 21.86, mdn. = 18.0) differed significantly from group
i WN (WN vs. WEX— U = 91%, p< .0005; WN vs. WECS-j^— U = 47,
: p <.00006) , but the two groups did not differ from each
other. Group WECS^ did not differ from any of the non-
reinforced control groups while group WE-^ differed sig
nificantly from all three of these groups. With the
exception of group WECS3Q (x = 46.27, mdn. = 38.5) which
differed significantly from group WN (U = 143, p < .02) but
also from group NN (U = 49%, p <.00006), none of the
other experimental groups differed from the reinforced
control group (WN).
Frequency of Head Pokes—
! H = 90.93, p <.001
Although this measure successfully differentiated
I between the reinforced and non-reinforced baseline groups,
; i.e., WN (x = 12.68, mdn. = 13.0, see Figure 13) and
j NN (x = 7.5, mdn. 7.0; U = 59, p < .00006) , none of the
Frequency of Licks
60
50 -
40
30 -
20 -
10 -
1
j •
■
u
WN WE^ WECS1 W E^ WECS^ W E^ WECS^ WEg0 W ECSgQ
Groups
NN NE NECS
Figure 12.— Mean (and median) frequency of licks on
test day.
'j
N>
WN m 1 WECS^ WE^ WECS-^ WE30 WECB^ WEg0 WECSg0 NN NE
Groups
NECS
Figure 13.— Mean (and median) frequency of head pokes
on test day.
■vl
CO
experimental groups differed significantly from the water-
reinforced control group (WN). However, the difference
between groups WN and WECS-^ approached significance (U =
182, p < .075, one-tailed test). There were no differences
among the three non-reinforced control groups (NN, NE, and
i
NECS), all of which differed significantly from the group j
!
!
WN.
Frequency of Cup Touch Sessions
H = 90.56, p < .001
For this measure the baseline comparison, between
groups WN (x = 7.77, mdn. = 7.5, see Figure 14) and NN
(x = 3.41, mdn. = 4.0) was again highly significant (U =
20h , p<.00006). As with the other frequency measures,
there were no significant differences among the three non-
reinforced control groups (NN, NE, and NECS). Although
group WECS^ (x = 5.36, mdn. 5.0) had significantly fewer
cup touch sessions than group WN (U = 92, p< .0005), it
had significantly more sessions than group NN (U = 120,
)
p < .0042) . Group WE^ (x = 7.41, mdn. 8.0) did not differ
significantly from group WN, nor from group WECS-^.
Duration of Head Poking—
H = 96.98, p < .001
The baseline comparison between groups WN (x =
Frequency of Cup Touch Sessions
4 -
WN WE-l WECSj^ WE15 WECS^ WE3Q WECS^ WEg0 WECSgg NN NE NflyCS
Groups
Figure 14.— Mean (and median) frequency of cup touch
sessions on test day.
76
36.40, mdn. = 36.5, see Figure 15) and NN (x = 18.46,
mdn. = 18.75) for this measure was highly significant (U =
52%, p< .00006). Again, there were no differences among
the three non-reinforced control groups (NN, NE, and NECS),
all of which differed significantly from group WN. Group
W E ^ (x = 2 9.45, mdn. = 28.0) spent significantly less j
; time head poking than group WN (U = 162, p < .031, one-
: tailed test), but significantly more time than group NN
: (U = 131%, p < .01). Comparison of group WEj with groups
NE and NECS revealed similar significant differences.
Group WECS^ (x = 23.73, mdn. = 23.0) also differed
significantly from group WN (U = 96%, p < .0006), but not
from group NN. Group WECS^ did, however, differ from
group NECS (x = 15.07, mdn. = 14.75; U = 127%, p< .008),
but not from group WE^. None of the other comparisons
with the remaining experimental groups were significant,
although the comparison between groups WN and WECS^,.
approached significance (U = 175, p < .06, one-tailed test).j
i i
! |
Duration of Cup Touch Sessions— I
! H = 71.90, p< .001
For this measure the baseline comparison between
the group WN (x = 19.46, mdn. = 17.0) and group NN (x =
[
I 6.98, mdn. = 5.75, see Figure 16) was again highly
Duration of Head Poking (seconds)
WN WE., WECS WE _ WECS.. WECS*, WE„ WECS„
1 1 15 15 30 TO 60 60
NN NE NECS
Groups
Figure 15.— Mean (and median) duration of head poking
on test day.
Duration of Cup Touch Sessions (seconds)
18
12 -
6 -
WN WEX WECS1 WE^ WECS^ W E ^ Q WECS^ WE^ WECSgQ
Groups
NN NE NECS
Figure 16.— Mean (and median) duration of cup touch
sessions on test day.
- j
00
79
significant . (U = 49, p <.00006). There were also no
| differences among the three non-reinforced control groups,
all of which differed significantly from group WN. Group
i WE^ (x = 13.33, mdn. = 11.0) differed significantly from
group WN (U = 144, p < .04) but differed significantly
from groups NN and NE as well. Group WECS^ (x = 9.54,
; mdn. = 9.0) also differed from group WN (U = 96%, p<
• .00 06). However, the comparisons of group WECS^ with
‘ groups NN and NECS were not significant. As in the case
of the Licks measure, group WECS^q ( x = 15.36, mdn. =
12.75) spent significantly less time in contact with the
cup than group WN (U = 165% p < .04, one-tailed test).
: However, as was also the case with the Licks measure, the
difference between group WECS^q and NN was highly signifi
cant (U = 97%, p < .001).
Discussion
The results strongly support the prediction that
; the 60 second access to water on the treatment day of the
i
i experiment would result in one-trial appetitive learning.
; In 6 out of the 9 dependent measures the water-reinforced
!
j baseline group (WN) differed from the non-reinforced
control group at a high level of significance. The three
measures from which a significant baseline comparison
difference failed to emerge were latency measures, while
5 of the 6 successful measures were based on frequencies
and durations of responses which were more closely related
both temporally and spatially, to the water reinforcement.
The one latency measure which successfully differentiated
between reinforced and non-reinforced groups, i.e.,
Latency to First Cup Touch on Test Day, was more closely
tied to reinforcement history than the other latency
measures which did not differentiate the groups.
t
With all but one of the measures {i.e., Head Poke
Frequency) with which one-trial learning was revealed,
the subjects which were administered ECS within 25 seconds
of the completion of the 60 second reinforcement period,
i.e., WECS^ gave evidence of interference with that learn-
when tested 24 hours later. With all of these measures
immediate ECS had the effect of attenuating changes which
were used as indexes of learning. The latency to the
first cup touch on the test day for immediately ECSed
subjects was significantly slower than the same latency
for reinforced, non-ECSed subjects. Since ECS had no
latency slowing effect on non-reinforced subjects (the
comparison of this latency for the non-reinforced control
81
group, NN, and the convulsed control group, NECS, was non
significant) , the reduced latency for immediately ECSed
subjects could not be attributed to an ECS performance
effect. If any performance effect of ECS existed in this
situation it appeared to have worked in the opposite
direction, since the ECSed control subjects had a shorter
i
latency to the first cup touch than the non-ECSed animals, j
1
j
In terms of the Lewis and Maher (1965) "conditioned
inhibition" hypothesis, it is difficult to determine how
the "general muscular relaxation, lowered level of
activity and the like," (Lewis and Maher, 1965, p. 234)
which are hypothesized to exist in the presence of the CS
associated with ECS (the apparatus cues), could result in
briefer latencies for ECSed control animals.
The attenuating effect of immediate ECS was
demonstrated most clearly in the case of the Licks measure.
ECS within 80 to 85 seconds of the initiation of water
reinforcement resulted in test day performance similar to j
that observed in non-reinforced animals (NN and NECS),
the inference being that the memory for the drinking
experience had been partially abolished. It should also
be noted that although the previous discussion, as well as
all of the statistical analyses, are based on group data,
82
the attenuation of such one-trial appetitive learning was
: seen in the majority of individual subjects. In the case
! of the Licks measure, the number of licks on the test day
; for 19 of the 22 animals in the immediately ECSed group j
i
!
(WECS.,) was approximately one-half of the mean value for j
I
the reinforced, non-ECSed subjects.
For the remaining measures which revealed signifi- !
cant attenuation of the appetitive learning experience
with immediate ECS, the effect was not as dramatic.
Although group WECS^ had significantly fewer cup touch
sessions and spent reliably less time head poking and cup
touching than group WN and did not differ from group NN
on the last two measures, the mean and median values for
. the ECSed group were all slightly above that of the con-
. trol group levels (NN, NE and NECS). This suggests that
some memory for the reinforcing experience was retained.
The absence of any amnesic effect with the head
poke measure was somewhat surprising. In the only other
; study of the effect of ECS on one-trial appetitive learn- j
| ing, Tenen (1965a), whose apparatus was the model for that
|
I used in the present experiment, found that frequency of
i head pokes, and not latency or duration measures, was the
j only variable with which statistical significance was
83
obtained. The principal difference between the two experi
ments, other than the subjects used (Tenen used rats) was
that Tenen habituated head poking during several sessions
prior to the introduction of water reinforcement. It is
possible that his technique had the effect of greatly
reducing curiosity and exploratory behavior associated j
|
with the cul, thereby reducing the probability of non-
reinforced animals poking their heads into the cul on the
test day. Since subjects in the present experiment had
not been in the apparatus prior to the session in which
they received reinforcement, they had not had the oppor
tunity of extinguishing curiosity and exploratory behavior
associated with the cul. In addition, the memory for any
habituation of head poking which may have taken place on
the first day of the present experiment should have been
partially interfered with by the presentation of ECS after
removal from the apparatus.
i
Although the comparison between the 15 minute ECS j
|
group (WECS • ) and the reinforced control group (WN)
JL5
approached significance for both frequency and duration of
head poking (p< .06 and p < .075, respectively, one-tailed
tests), there was no clear attenuation of appetitive
learning with 15 minute ECS. This was particularly
84
surprising in view of the finding that 30 minute ECS
effectively interfered with learning with several of the
measures, and also because numerous other studies have
found ECS interference at that interval. Tenen (1965b)
I
reports attenuation of appetitive learning with ECS i
[
1
administered up to one hour after the learning trial (and \
at four intervals, between one minute and one hour, includ
ing 15 minutes). It is interesting to note that, although
group WECS-^,- did not differ significantly from the water-
reinforced group (WN), it also did not differ from the
30 minute ECS group (WECS^q) on either the Licks or
Duration of Cup Touch Sessions measures, where group
WECS2q was significantly inferior to group WN. However,
such comparisons only support impressions gained from
inspection of the medians (group WECS^j-'s median is
clearly smaller than that of group WN) and do not warrant
the conclusion that 15 minute ECS interfered with appeti
tive learning.
The situation with the 30 minute ECS group is much |
more in keeping with pre-experimental predictions. As
pointed out in the preceding paragraph, the number of test
day licks and the duration of cup touch sessions for
group WECS^q were significantly below the levels attained
85
by group WN. This leads to the conclusion that some memory
interference resulted from convulsive treatment at this
interval. However, this group differed significantly from
group NN on both of these measures, suggesting that
interference with learning after a learning-ECS interval
of 30 minutes was less than that observed when ECS was
i
administered immediately. Thirty minute ECS did not result
in subjects being statistically undifferentiatable from
non-reinforced animals as was the case with ECS adminis
tered immediately after the learning trial. In addition,
statistically reliable evidence for interference with
learning at this interval was obtained with only two of
the 9 measures.
Electroconvulsive shock administered one hour
after water reinforcement had no effect on test day per
formance with any of the measures utilized in this study.
This finding, when viewed in light of the results with
immediate and 30 minute administration of ECS, strongly
supports the conclusion that a graded effect was obtained
with the intervals used, although the absence of any clear
interference with 15 minute ECS detracts somewhat from
this argument. Immediate ECS had the greatest attenuating
effect on learning, as evidenced by the levels of
! 86
significance obtained and in terms of. the number of
measures with which interference was obtained. Although
; ECS administered 15 minutes after the learning trial had
little effect, 30 minute ECS produced results interme
diate between immediate and 6 0 minute administration. In
this group interference was obtained with only two of the
measures and the significance levels, although reliable,
; were not as striking as those obtained with immediate ECS.
! The period of time necessary for the memory trace of the
appetitive learning experience used in this study to
become stable enough that it can withstand the trauma
produced by electroconvulsive shock would appear to lie
somewhere between 30 and 60 minutes. Since the magnitude
of the interfering effects of ECS delivered after 30 min
utes appears quite small when compared with the effects of
: ECS administered immediately, whatever central nervous
system processes required to protect memory traces from
disruption should be completed closer to 30 than to 60
minutes after the learning event.
I
The pattern of results with immediate ether,
| although supporting an interference hypothesis, were less
I
j striking than those obtained with immediate ECS. With
| ether, the levels of statistical significance were
87
somewhat higher and differences between this group and the
reinforced baseline group (WN) were obtained on only three
of the five frequency-duration measures, as well as with
the First Latency to Cup Touch on the test day. (The
comparison between groups WE^ and WN for Frequency of Cup
Touch Sessions did not reach significance.) Although
statistical significance was obtained only in the case of
the Head Poke Latency Difference Score, and was approached
(p< .06) with Head Poke Duration, the non-reinforced,
etherized control group (NE) was somewhat depressed on many
of the measures, suggesting the possibility that part of
the apparent interfering effect of ether might be due to i
an aversive or a performance effect of ether. But since
the only significant reduction in control level perform
ance was with a latency measure which did not yield a
significant learning effect, it can be argued that the
performance effect of ether is minimal, and was over
shadowed by ether's interfering effect on memory.
i
!
The observation that immediate etherization j
attenuated learning less effectively than immediate ECS is
further supported by the fact that group WECS^ obtained
lower scores than group WE^ (indicating greater degrees
of attenuation) on the Cup Touch Frequency measure. (The
i
Licks and Cup Touch Duration measures both approached
significance for this comparison; p< .053 and p< .061,
I
respectively.) In addition, immediate etherization did
not reduce performance to levels statistically indistin
guishable from baseline control (NN) levels as did imme-
' I
i
; diate ECS. (For all of the frequency and duration measures
; the performance levels for group WE1 were higher than
those of group NN by very significant amounts, i.e., all
p values <.01.) All of these findings appear to indicate
that the effectiveness of immediate ether in attenuating
one-trial appetitive learning is only slightly greater
than that obtained when ECS is given 30 minutes following
the learning trial.
None of the comparisons between the water-
: reinforced control group (WN) and the remaining ether
groups (WE^5, WE^q, or WE^) reached significance, suggest
ing that ether has little interfering effect on appetitive
: learning unless administered very shortly after the
I learning trial.
From the foregoing discussion several important
j
| points can be made. First, particularly in the case of
! the latency measures, ether appears to have had an effect
89
over and above its interference with learning. Although
the differences were not significant with the frequency
and duration measures, in every case the means and medians
for the etherized control group (NE) were below those of
the baseline, control group (NN) . Whether the effect of
either on behavior is debilitating or aversive is diffi
cult to determine from this study, but it would be of
interest to attempt a direct examination of this problem
in future studies utilizing ether. It is possible that
none of the previous investigations using ether as an
amnesic agent have detected such effects because none of
them have used more than two or three measures at most.
Also, since previous studies have utilized ether only in
aversive situations, reductions in latency (the measure
most frequently used in one-trial passive avoidance
learning studies) resulting from performance effects of
ether would have gone unnoticed, since they would have
A
worked in the same direction as retention of the avoidance
habit, making it appear that ether was not as effective an
amnesic agent as ECS. The curious fact remains, however,
that in previous passive-avoidance investigations utiliz
ing ether, (e.g., Herz, Peeke, and Wyers, 1966; Pearlman,
1966) shocked subjects treated with the amnesic agent gave
90
evidence of significantly reduced latencies when compared
with shocked but unetherized animals. It is difficult to
determine how ether could partially interfere with laten
cies in the present study, but result in shorter latency
in conditioned avoidance investigations.
The failure of the latency measures to success
fully differentiate between reinforced and non-reinforced
groups (WN and NN) and to show any interference effect,
except in the case of the Latency to Cup Touch on the
Test Day is curious in that many of the above-mentioned
one-trial passive avoidance studies have obtained a sig
nificant retrograde amnesia gradient with latencies. The
fact that the present investigation utilized latency
measures with appetitive learning, which has not been
previously done, suggests the possibility that decreased
latencies following one-trial water reinforcement are more
difficult to obtain than latency increases following one-
trial shock avoidance learning. In addition, the inter
action of habituation of curiosity and exploratory
behavior with reinforcement in the present study may have
served to increase latencies which would have decreased
had the separation of these variables been achieved.
CHAPTER IV
EXPERIMENT 3
Since the previous experiment failed to result in
: reliable attenuation of one-trial appetitive learning with j
l
i ether administered more than 85 seconds after the learning
. trial or with convulsive seizures induced 15 minutes after
reinforcement, it was of interest to determine whether the
use of these agents at a time interval intermediate between
one and 15 minutes after reinforcement would result in
interference effects greater than those observed with 15
minute administration, but less than those resulting with
immediate application. In addition, the present investi
gation also permitted replication of the baseline water-
reinforced and the non-reinforced control groups.
I
i
i
Method i
------------------------------------- i
I
: I
: Subjects
|
One hundred twenty male Swiss-Webster albino mice,
! obtained from the same source, and housed and deprived in
92
the same manner as in the first two experiments, were used
in this study.
Apparatus
The apparatus was the same as that utilized in the
first two experiments.
Procedure
The procedure was the same as that of the previous
investigation. The subjects were assigned to the follow
ing groups which contain unequal numbers of subjects:
1. WN (N =23)— the baseline, water-reinforced
group treated in the same manner as group WN
in Experiment 2.
2. WE,. (N = 24)— treated the same as group WN,
but etherized 5 minutes after initial contact
with the drinking cup.
3. WECS-. (N = 26)--treated as group WE_, but
b d
administered ECS after 5 minutes.
4. NN (N = 24)— control group for the effects of
water-reinforcement and treated in the same
manner as group NN in Experiment 2.
Subjects in this study were run exactly as in
Experiment 2, the only difference being the use of the
93
5 minute .learning—-ECS or ether interval. Since in the
preceding study the ECSed control group did not differ
from group NN and since group NE differed significantly
from group NN on only one measure, these two groups were
not run in the present study. As before, the experiment
took place on two days, and the same measures were used.
Results
As with the preceding studies, comparability
analyses were performed on the experimental treatment day
latencies to First Head Poke and First Cup Touch. The
non-significant H values demonstrated as with the two
previous experiments, that the groups were comparable
prior to experimental treatment. (Further analyses also
indicated comparability with the earlier studies.) In
addition, an analysis of the amount of time spent drinking
from the cup by the reinforced groups (WN, WE,., and WECS,.)
on the first day of the experiment also yielded a non-
i
significant H value, further indicating the comparability, j
I
i
Since the pattern of results with the different measures
varied as in the previous experiments, each measure will
be considered separately. (Non-parametrie analyses were
used in all cases.)
94
Latency to First Head Poke—
H = 23.74, p < .001
The difference between the baseline water-rein-
forced group (WN (x = 6.98, mdn. = 6.0, see Figure 17 and
tabled means and medians in Appendix C) and the non
reinforced control group NN (x = 18.48, mdn. = 15.5) was
highly significant (U = 101%, p < .00022). In addition,
both of the experimental groups, WE^ (x = 13.02, mdn. =
10.25) and WECS,. (x = 8.62, mdn. = 7.75), differed signifi
cantly from group WN (U's = 10 0%, p < .0001, and 181,
p <.02, respectively). However, while there was no differ
ence between groups WE5 and NN, group WECS^ evidenced a
significantly briefer latency than the non-reinforced j
j
control group, NN (U = 173, p< .05). Finally, the differ
ence between the two experimental groups was significant
(U = 152, p< .001) with the ether group having the longer
latency.
i
I
Head Poke Latency Difference Score— i
H = 14.85 , p <.01
The results with this measure were similar to those
obtained with the preceding one. Groups WN (x = -7.61,
mdn. = -6.0,see Figure 18) and NN (x = 3.23, mdn. = 1.0)
differed reliably (U = 141%, p < .0042). Group WE5 (x =
95
20 ,
WN WE5 WECS5 NN
Groups
Figure 17.— Mean (and median) latency
to first head poke on test day.
96
01
fl
O
O
CD
W
CD
S - t
o
o
cn
CD
o
£
CD
U
CD
MH
< 4 -1
■H
Q
>i
O
£
CD
-P
i d
t£
CD
M
O
P L ,
i d
CD
K
0 -
-4
- 8 -I
WN WE5 WECS5
Groups
NN
Figure 18.— Mean (and median) head poke
latency difference score (day 1 - day 2).
97
1.29, mdn. = 1.25) differed from group WN (U = 136, p <
; .0014) and group WECS5 (x = -5.14, mdn. = -4.5, U = 146%,
p < .002) but not from group NN. However, the 5 minute ECS
| group, WECS,., differed reliably from group NN (U = 173,
p < .005), but not from group WN.
Latency to First Cup Touch-- j
i H = 16.24, p < .01 ;
The pattern of results with this measure was
essentially the same as that obtained above. Group WN
(x = 12.78, mdn. = 8.5, see Figure 19) evidenced a sig
nificantly briefer latency than group NN (x = 30.08, mdn.
= 17.5; U = 131, p< .002). The 5 minute ether group,
WE,., differed reliably from group WN (U = 175, p < .016) and
from group WECS (x = 13.75, mdn. = 11.0; U = 199, p< .03),
5
but not from the non-reinforced control group, NN, although
the latter difference approached significance. Group
WECS5 differed from group NN (U = 142, p < .001) but not
from group WN. j
i
! Cup Touch Latency Difference Score—
| H = 4.81, N. S.
j
I Although the difference between groups WN and NN
j significant with this measure (p < .04, one-tailed test),
| the over-all H value was non-significant and the results
98
WN WEC WECS
Groups
Figure 19.— Mean (and median) latency
to first cup touch on test day.
will not be presented.
Frequency of Licks—
H = 5.07, N. S.
Despite the fact that Group WN appeared to give
significantly more licks on the test day (p< .04. one
tailed test) than group NN, the over-all analysis of
variance did not reach significance.
Frequency of Head Pokes—
H = 25.96, p <.001
As with the previous measure, the difference
between the reinforced baseline group, WN (x = 14.22, mdn.
= 14.0, see Figure 20), gave evidence of reliably more
Head Pokes than the control group, NN (U = 56, p< .00006).
In addition, both groups WE^ (x = 13.0, mdn. = 11.50) and
WECSg (x = 11.15, mdn. - 11.0) were significantly differ
ent from group WN (U's = 200, p< .03, one-tailed test, and
150%, p< .003). Both of the experimental groups gave
significantly more Head Pokes than group NN (NN vs.
WE5— U = 132, p< .0014; NN vs. WECSg— U = 174%, p< .008),
but did not differ from each other.
Frequency of Cup Touch Sessions—
H = 26.70, p < .001
The pattern of results with this measure was the
Frequency of Head Pokes
15 -
10
5 -
0
WN WE5 WECS5 NN
Groups
Figure 20.— Mean (and median) fre
quency of head pokes on test day.
; same as that in the preceding one. Groups NN (x - 4.21,
mdn. = 4.0, see Figure 21), WE^ (x = 6.62, mdn. = 6.0) ,
and WECScj (x = 6.15, mdn. = 6.0) differed significantly
from the reinforced baseline group, WN (x = 7.83, mdn. =
l 8.0; U's = 73, p< .00006, 187, p < .04, and 178, p < .016,
respectively). Group NN had significantly fewer Cup Touch
Sessions than either group WE^ (U = 138, p< .002) or group
WECS^ (U = 173, p< .008), but the latter two groups did
not differ.
Duration of Head Poking—
H = 10.00, p < .02
As with the previous measures, the difference
between the baseline group, WN (x = 31.85, mdn. = 28.5,
see Figure 22), and the non-reinforced control group,
NN (x = 21.23, mdn = 20.5), was significant (U = 160,
p < .02). Both of the experimental groups, WEg (x = 24.19,
mdn. = 23.75) and WECS^ (x = 25.31, mdn. = 24.0) differed
from group WN (U's = 178, p < .02 and 203, p< .03, one-
tailed test, respectively), but not from group NN nor from
each other.
Duration of Cup Touch Sessions—
H = 6.39, N. S.
Although group WE5 differed significantly from
102
WN we5 wecs5
Groups
NN
Figure 21.— Mean (and median) fre
quency of cup touch sessions on test day.
103
WECS
Groups
Figure 22.— Mean (and median) duration
of head poking on test day.
104
group WN, the difference between groups NN and WN was not
significant.
Discussion
Consistent with the previous experiment, the
difference between the baseline reinforced group, WN, and
the non-reinforced control group was highly significant on
most of the measures. This indicates the reliability of
the one-trial appetitive learning situation developed for
this investigation. Although the actual test day means
and medians for the two studies differed, in some cases
even significantly, the reliability of the technique for
differentiating between previously comparable groups was
demonstrated. The principal difference between the
results of the two studies was obtained with Frequency of
Licks,^ which, in Experiment 2, was one of the most
-*-In the case of the Frequency of Licks measure, the
difference between the baseline reinforced groups in the
two studies (x = 61.96, and mdn. = 55.5 in Experiment 2 and
x =31.70, mdn. = 28.0 in the present study) was signifi- j
cant, suggesting that sampling error may have strongly |
influenced the actual frequencies obtained in either the
original study, the present replication, or, more likely,
both. Further support for this argument is the result of
a replication of the present study (total sample size = 66)
in which group WN (x = 69.67, mdn. = 65.0) gave more Licks,
although not significantly more, than in the original
experiment.
105
sensitive measures, and Duration of Cup Touch Sessions,
which was highly correlated with Licks (r = .873). In the
present study groups WN and NN did not differ on Cup Touch
Duration and differed only at the .04 level (one-tailed
test) on the Licks measure (despite this difference the !
|
over-all H value was non-significant), while these two
groups differed beyond the .00006 level in Experiment 1.
The lack of consistency between the two studies, in terms
of measures which reflected learning as a result of the
brief experience in the apparatus, can be interpreted in
several ways. It is possible that the different batches
of animals differed in terms of some variable which was
related to one or several of the dependent measures.
Although the batches appeared comparable in terms of
several pre-treatment measures, the comparability variables
may not have been correlated highly enough with the depend
ent measures. This possibility is further supported by
the previously noted fact that the dependent measure data
was highly variable, necessitating nonparametrie analyses.
It is also possible, although less probable, that the
variability between batches of animals resulted in differ
ent things being learned by various animals on the treat
ment day. However, the fact remains that in both studies
106
reliable learning was reflected in most of the dependent
measures.
It is of interest in the present study that for
three of the four test day latency measures the difference
between groups NN and WN was significant. Reference to
i
j
the previous study indicates that significance between I
groups WN and NN was obtained with only one of the latency I
measures, Cup Touch Latency, although the difference for
Cup Touch Difference Score approached significance (p< .07,
one-tailed test).
In the present experiment, ether administered
5 minutes after water reinforcement greatly interfered
with performance on all of the measures in which learning
was demonstrated, i.e., group WE^ differed from group WN
on all of these measures. However, since no etherized
control group was run in the current investigation, it is
difficult to determine whether the reduction in performance
j
level evidenced by this group, especially with the latency I
i
i
measures, is a reflection of interference with learning or j
I
a performance effect. In Experiment 2 it was noted that
etherized control subjects gave evidence of significantly
larger latencies when compared with non-reinforced and
ECSed controls. Since only one of the latency measures in
that study revealed a significant learning effect, the
performance effects of ether were considered only briefly.
However, since significant learning and a reliable ether
effect resulted with three of the measures in the present
experiment, the possibility of ether-produced performance
effects must be more carefully considered. In the absence
of etherized control subjects with which to compare the
non-etherized controls, it is impossible to accept the
reliably greater latency measures in the 5 minute ether -
group as unequivocal evidence for interference with learn
ing. But since a significant interference effect was also
observed on the frequency and duration measures in Experi
ment 2, where no significant performance effect was
observed with etherized control animals, attenuation of
learning, as measured by latencies, cannot be rejected as
a possibility.
In addition to differing from the reinforced base
line group on all of the measures with which learning was
|
demonstrated, the 5 minute ether group differed signifi- j
cantly from the control group NN on two of the measures |
(Frequency of Pokes and Cup Touch Sessions), while the
differences between these two groups did not reach signifi
cance on the remaining measures. This suggests that
108
relatively complete attenuation of the drinking experience
was produced by ether in the present study. Such a finding
is somewhat surprising in light of the fact that the
immediate ether group in the previous investigation dif-
; fered-significantly from the non-reinforced control group
; j
of all but one on the measures with which learning was
i
; demonstrated. It would appear, therefore, that adminis- j
; tration of ether 5 minutes after the drinking experience
i is at least as effective, if not more so, than immediate
etherization.
The situation with convulsions induced 5 minutes
after the learning experience is more in keeping with the
, expected graded effect. With ECS significant differences
indicative of interference were obtained on only four of
the measures {First Latency to Head Poke, Frequency of
Head Pokes, Frequency of Cup Touch Sessions, and Head Poke
Duration) while the difference for Frequency of Licks
approached significance (p< .053). However, on all but j
: i
■ the Licks and Poke Duration measures, the 5 minute ECS |
i
i
i
; ;
i group also differed reliably from the control group, NN, i
! suggesting that although some attenuation was produced,
I the performance of ECSed animals was not reduced to near
! control level performance as was the case with immediate
109
ECS in the previous experiment.
Examination of the differences between the two
experimental groups, WE^ and WECS,., reveals that although
; ether appeared to produce significantly greater interfer
ence effects than ECS on the latency measures, where the
differences between the two groups were significant, no j
I
differences between the two experimental groups were
observed with any of the frequency or duration measures.
Because of the possible performance effects of ether on
the latency measures, it would appear premature to conclude
that ether resulted in greater interference with learning
than ECS on the basis of the above significant differences.
In terms of the number of measures on which inter
ference was obtained and the levels of significance
reached, the results of the present investigation would
appear to warrant the conclusion that reliable interference
effects are produced when ether or ECS is administered to
animals within 5 minutes of a one-trial appetitive learn-
i ing trial. Further, these findings suggest that both
agents produce a greater degree of attenuation when admin- I
i istered 5 minutes, as compared with 15 minutes, after the
; learning experience. In fact, it would appear that appli-
i
j cation of ether within 5 minutes of the learning trial
i n o
produces effects as great as those observed with immediate
administration. Although the results of ether on the
latency measures can be attributed, to some degree, to the
I effects of this anesthetic on performance, the significant
interference produced with the frequency and duration
measures strongly suggests the ether's attenuating proper- |
ties are due to its effect on memory. Since not even a j
]
trend toward a performance effect of ECS was detected in
i the comparisons between the normal and ECSed control groups
in the previous study, the significant interference effect
of ECS in the present study can also be attributed to its
amnesic properties.
CHAPTER V
DISCUSSION
'
The results of the present investigations illus
trate the effectiveness of the one-trial learning situa
tions used, both with positive and negative reinforcement.
As was pointed out in the Introduction, paradigms utilizing
one learning trial and one administration of amnesic
agents appear to be most appropriate for separating the
amnesic from the performance effects of such treatments.
Highly reliable evidence for learning, as evidenced by
significant differences between reinforced baseline groups
and non-reinforced control groups, was demonstrated with
all of the measures in the aversive situation (Experiment
1), with 7 of the measures in the original appetitive
learning paradigm (Experiment 2), and with 6 of the meas
ures in the replicated water reinforcement study (Experi
ment 3). Although some unaccounted for differences exist j
between the effectiveness of the various measures in the
latter two experiments, consistent results were obtained
111
112
; on most of the variables.
The findings obtained in the current studies, in
which interference with both appetitive and aversive
; learning were observed, do not appear interpretable in
terms of ECS-produced fear or "conditioned inhibition"
3
]
hypotheses. Both of these explanations of the data from
previous retrograde amnesic studies appear forced to
predict that administration of amnesic agents in the pres
ent experiments, both with approach and avoidance learning,
would have similar effects. The "conditioned inhibition
hypothesis," for example, maintains that the interference
resulting from administration of amnesic agents, especially
in the case of electroconvulsive shock, is produced by
relaxation conditioned to the cues of the experimental
apparatus. Although amnesic treatments were administered
: outside of the apparatus in the present studies, making it
difficult to establish conditioned relaxation, it would be
anticipated from this theory that, regardless of the j
; l
; reinforcement conditions (positive or negative), ECS or i
| j
ether should effect the dependent measure in the same way. j
: i
I However, the results produced by these treatments in the
current appetitive and aversive learning situations are
j in the opposite directions. Whereas the agents
113
administered following water reinforcement increased
latencies and decreased frequency and duration measure
scores, their application following shock decreased laten
cies and increased scores obtained on the other variables.
Lewis and Maher (1965, p. 236) also suggest that
an interaction exists between level of arousal and the
degree of retrograde amnesia produced in a given experi- j
mental situation: "Responses that are mediated by some-
i
what lower levels of arousal will be less disrupted and
will more readily reappear when the 'emotional' response
has been antagonized." This 'axiom1 appears to have been
added to their basic argument to account for the amnesic
effects obtained in passive avoidance situations (as
opposed to more arousing active avoidance paradigms) such
as utilized in Experiment 1, where ECS led to interference
with the avoidance response. Extension of their logic
leads to the prediction that in a water-reinforced approach
situation, such as that used in Experiments 2 and 3, the j
i
level of arousal should be lower than that observed in the j
shock avoidance situation and therefore should result in
less disruption. In both of the latter experiments, how
ever, greater degrees of retrograde amnesia were obtained
with immediate administration of ECS than in the passive
114
avoidance situation of the first experiment. It would
appear, therefore, that the "conditioned inhibition
hypothesis" can not predict the type of results obtained
in the present series of studies without the addition of
still further postulates. In addition, the absence of
any differences between the normal and ECSed control
groups on any of the measures further complicates inter
pretation in terms of the "conditioned inhibition hypoth
esis." According to this hypothesis, it would be expected
that ECSed control subjects would have relaxation condi
tioned to cues associated with the experimental apparatus
and, as a result, perform differently than non-ECSed
control subjects.
The results of the present studies therefore
appear to support an interpretation based on the consoli
dation hypothesis, in terms of which the observed inter
ference effect was the result of interruption of the
process or processes involved in the establishment of
permanent memory. However, it should be emphasized that
this theory, as it is usually interpreted, maintains that
"memory," as an entity, is interfered with, while the
results of the present investigation, where differential
amnesia was produced with the different measures, suggest
115
that certain aspects of the memory of the learning experi
ence are more completely attenuated than others.
Although these experiments offer little information
regarding the mode or site of action of the amnesic agents
utilized, they do present data concerning differences in !
i
l
the effect of such agents on appetitive and aversive learn-:
ing. Comparisons among the three studies indicate that j
although electroconvulsive shock administered immediately
after water-reinforcement resulted in almost complete
amnesia for the reinforcement, as indexed by the absence
of differences between the immediately ECSed group and
the non-reinforced control group on the test day on several
of the measures, convulsions induced 30 minutes after
reinforcement had only a slight, although significant,
effect. The results in the aversive situation were quite
different with immediate ECS resulting in only partial
attenuation of the passive avoidance habit. However, ECS
administered 15 minutes after shock reinforcement resulted
I
I
in almost as complete interference as immediate administra-j
:
tion with several of the latency measures. It should also
be noted that 15 minute administration of ECS resulted in
amnesia for the shock experience with four measures, while
15 minute ECS administration in the appetitive situation
resulted in no reliable attenuation, although some inter-
; ference was obtained with 30 minute ECS on two of the
measures in the same experiment.^ (Despite the failure of
; the replication, Experiment 3, to produce as reliable
evidence for learning as was obtained with several of the
: I
measures in Experiment 2, the interfering effects of ECS |
administered 5 minutes after water reinforcement appear to |
lie intermediate to those produced by immediate and 30 min
ute induction of ECS.)
Ether was generally less effective in producing
longer interval retrograde amnesia in both appetitive and
aversive situations, as was predicted. Although immediate
and 5 minute administration of the anesthetic produced
"^The absence of any significant amnesic effect
with 15 minute administration and reliable effects on two
measures with 30 minute administration is somewhat sur
prising. Although the result could be due to sampling
error, it is also possible that such a break in the con
solidation gradient reflects changes in the processes
responsible for the permanence of memory. The possibility
that at least two types of memory exist (short and long I
term, and perhaps an intermediate stage) has been consid
ered by a number of authors (Barondes and Cohen, 1966;
Flexner, Flexner, and Stellar, 1963; McGaugh, unpublished i
manuscript). In addition, in a recent study of the |
effects of spreading depression (SD) on one-trial passive
avoidance learning (Rabedeau, 1966), no amnesia was ob
tained with immediate SD but relatively complete interfer
ence resulted at 16 minutes, further supporting the
possibility of memory consisting of more than one process.
117
2
almost as complete amnesia as immediate and 5 minute ECS,
i.e., no difference-between the two groups was obtained on
these measures, ether produced no effect on appetitive
learning when administered at greater intervals. More
important, not even a suggestion of an amnesic effect was
obtained with either immediate or 15 minute anesthetization
in the aversive situation.
A graphic comparison of the effectiveness of the
two amnestic agents in the present appetitive and aversive
learning situations suggests that in spite of the failure
to obtain reliable interference at some of the learning
trial-amnesic agent intervals, strong evidence for a graded
effect was obtained. Figure 23 presents a representation
of the amnesic effect obtained with the most sensitive
measure in each of the two major studies (Experiments 1
and 2). In the appetitive situation, the Frequency of
Licks measure produced the greatest evidence for retrograde
2
Five minute administration of ether, compared with
5 minute ECS, resulted in slightly greater interference on
two of the frequency and duration measures, and signifi
cantly greater attenuation of three of the latency measures.
Although the effect of ether on the latency measures can
be questioned in terms of possible performance effects,
as mentioned above, no readily available explanation
exists for the slight superiority of ether on other
measures.
Per Cent Retention
100 H
80-
60 -
"fl Water + Ether
40-
□ --------□ Water + ECS
A* Shock + Ether
A Shock + ECS
1 15 30 60
Time between Reinforcement and Amnesic Treatment (Minutes)
Figure 23.— Per cent retention as a function of time interven
ing between learning trial and administration of amnesic agent on
frequency of licks measure for water reinforced groups and on
latency to first head poke on test day for shocked groups.
119
amnesia, while First Latency to Head Poke on the test day
reflected the greatest degree of attenuation in the pas-
: sive avoidance paradigm. The Per Cent Retention measure
graphed in this figure is the ratio between the level of
performance achieved by each experimental group and that
obtained by the baseline water- or shock-reinforced group |
i
j
for the appropriate study and was derived as follows:
1. The mean for the non-reinforced control group,
NN, was subtracted from the mean of the base
line group, WN or SN, to arrive at a measure
reflecting the learning which had taken place
as a result of reinforcement.
2. The mean for group NN was subtracted from the
mean of the experimental group to create a
similar measure, but here also reflecting the
effect of the administration of amnestic agent
on learning.
' — 3. The resulting numbers were made into a ratio, |
i.e., 2/1, which was an expression of the
retention of the learning experience by each
of the experimental groups.
(For example, the mean for group NN was 15.41 which, when
; subtracted from the mean of group WN, 61.96, yielded 46.55.
i
; Subtraction of the mean of group NN from the mean of
I group 31.91, yielded 16.50, and the resulting ratio,
16.50/46.55, resulted in a retention or savings score of
| 35%.)
Although reflecting less complete amnesic effects,
j
: a similar graded^effect is seen in Figure 24 which was
1
I
! created by computing similar ratios on the group medians
I for all of the frequency-duration measures on which reli-
t
i able learning was obtained in all three studies. (Per
cent retention measures were computed for all reliable
measures for each group and the median percentage plotted.)
The over-all raising of the gradient in the second figure
reflects the fact that different degrees of interference
were produced with the various measures.
It is significant that in the present investiga
tions the degree of retrograde amnesia produced at most
intervals was considerably less than that predicted on the
; basis of previous experiments. Only in the appetitive
1
| learning situation, with immediate ether and ECS, was
almost complete amnesia produced. With intervals longer
! than 5 minutes,attenuation was relatively slight following
I
| positive reinforcement. In the aversive situation no
interference resulted even with immediate administration
Per Cent Retention
100 H
80-
/ /
//"
60 -
Water + Ether 40 _
□ -------- □ Water + ECS
-A Shock + Ether 20 -
■ A Shock + ECS
15 30
Time Between Reinforcement and Amnesic Treatment (Minutes)
Figure 24.— Per cent retention on frequency-duration measures
as a function of.time intervening between learning trial and
administration of amnesic agent. (See text for explanation.)
! 122
; of ether. Other investigations have obtained almost com-
f
j plete amnesia for an aversive shock with immediate ether
| and some attenuation with intervals as long as 15 to 20
i
! minutes (Abt, Essman, and Jarvik, 1962; Pearlman, 196 6;
I Herz, Peeke, and Wyers, 1966). Similarly, more complete
and longer duration interference than that obtained in the
i present aversive situation has been reported with ECS
! (e.g., Heriot and Coleman, 1962; Weissman, 1964). In
I addition, Tenen (19 65a, b) obtained some amnesia with ECS
' delivered as long as 5 hours after water reinforcement with
an apparatus very similar to the one used in the present
studies.
The fact that different apparatuses and procedures
were used in each of the above studies, all of which
differed from that of the current experiments, suggests
i that the inconsistencies are a result of some factor or
factors unique to the present studies. Perhaps the most
; prominent feature which differentiates the current experi-
I
! ments from those mentioned above is the fact that in all
i
of the other studies subjects were removed from the appa
ratus immediately following reinforcement. This is true
particularly in the passive avoidance studies, where other
i
experimenters removed subjects from the apparatus as soon
123
as they had received a shock. In Experiment 1, subjects
were left in the experimental chamber for one minute
following the receipt of shock in order to maintain com-
... . 3
parability with the appetitive learning paradigm. Even
in the case of Tenen's appetitive situation, animals were
removed from the apparatus after only 10 seconds of drink- |
T- ' !
ing, whereas 60 seconds elapsed between a subject's first j
j contact with the drinking cup and his removal in Experi-
ments 2 and 3. It would appear to follow, therefore, that
the time interval immediately following initiation of
reinforcement is a critical one for the establishment of
memory for the experience. Comparison of the present and
previous studies suggests that if the organism is removed
■ from the apparatus, and consequently from the cues associ
ated with reinforcement, immediately after reinforcement
occurs, ECS- or ether-produced interference with memory of
i
the reinforcing conditions will be more complete at any
interval than if the animal is allowed a longer period of
time in the stimulus conditions associated with reinforce-
! ment. It therefore appears that lengthening the single
S trial {the period during which the subject is in the
j O
°In addition, almost half of the subjects in the
I present study received two shocks.
124
situation containing reinforcement cues) results in better
: learning which is more resistant to interference by amne
sic treatments. Further support for this hypothesis comes
i from the results of extension of an experiment on the
effects of brief stimulation of the caudate nucleus fol
lowing a foot-shock administered to rats pressing a lever j
for water (Williston, Herz, Peeke, and Wyers, 1964). In I
that study subjects were left in the apparatus for 30
seconds following the foot-shock, and although relatively
complete RA resulted from caudate stimulation delivered
100 msec, after foot-shock, longer interval RA was less
apparent.
An additional related factor might also account
for some of the discrepancies between the present and
earlier studies. The over-all size of the apparatus was
such that during the interval between reinforcement and
removal from the apparatus, particularly in the case of
the aversive learning situation, subjects could never move
i
1 I
i I
| very far from the cul where shock had been received. This !
i
i
! may well have resulted in increasing the importance of the
'i
f
| cues most closely associated with shock, i.e., the cul and
I
the drinking cup, during the proposed critical period,
j
; i.e., immediately following reinforcement.
125
The use of multiple dependent measures in the
present series of investigations resulted in several curi
ous observations. With one of the latency measures in the
aversive situation, and with one of the latency and one of
the frequency measures in the appetitive paradigm, signifi
cant learning but no reliable ECS-produced RA was obtained.
i
This suggests that different response components are j
differentially attenuated as a result of amnesic treatment.
Since the frequency and duration measures appear to have !
reflected more complete amnesia, than the latencies, it is
conceivable that a temporal gradient exists in terms of
the interval preceding reinforcement which can be affected ]
by amnesic agents. The latency measures appear to involve
events further removed in time from reinforcement than the
frequency-duration ones. Since the latter are interfered
with to a greater degree, it is possible that the degree
of retrograde amnesia produced is dependent upon the
)
temporal interval separating each response component from j
reinforcement as well as the time intervening between
i
reinforcement and administration of the amnesic treatment.
Further support for this hypothesis is the obser
vation that one of the most sensitive measures, in terms
of the degree of interference produced with both appetitive
126
and aversive learning, was the Frequency of Licks, which
appears to be more closely associated with reinforcement
than any of the other measures.^
The effects of the two amnesic agents in the three
experiments should be viewed in the context of the mor
tality rate resulting from their use. In each of the
three experiments, approximately 35% of the animals ether- j
ized to (or close to) respiratory arrest failed to recover.
With ECS the fatalities were approximately half of the
value for ether, 18%. It is difficult to determine whether
these figures are unusually high, since most studies
utilizing the two amnesic agents do not, as a rule, report
mortality rates. In addition, many of the previous inves
tigations have utilized rats, which appear to withstand
the effects of ECS, and perhaps ether, better than mice.
However, the above mortality rates are similar to those
reported in at least one study utilizing mice. Herz,
Peeke, and Wyers (1966) report that 18% of ECSed and 25%
!
4
One exception to this was in the case of the
final study (Experiment 3) where the Licks measure did not
result in reliable evidence of learning. Although no
explanation for this finding is readily available, it
should be reiterated that a replication of this study ob
tained both reliable learning on 7 of the measures,
including Licks, and a significant 5 minute ECS interfer
ence effect.
of etherized subjects failed to recover. It should be
noted, however, that following ether and ECS treatments,
! subjects in the present investigations were artificially
respirated until normal breathing was restored, while no
! • I
artificial respiration was utilized in the earlier study. !
i
, The higher ether death rate in the present investigation
; may reflect the fact that the endpoint for etherization j
I
was individually determined by each subject's response to I
' i
ether. When the experimenter noted that breathing had
become very faint or had ceased entirely, the subject was
removed from the etherization chamber. (The median dura
tions of etherization for all groups were between 45 and
50 seconds.) By contrast, the previous investigation
utilized a fixed 70 seconds of etherization following
which the subject was removed from the chamber, regardless
of the stage of respiration which had been reached. It is
likely that the probability of recovery from anesthesia is
I
greatly reduced if animals are etherized to the point of |
! i
I respiratory arrest, and this factor may account in part j
■ \ \
| for the higher rate of ether mortality in the present |
! experiment. Although the conditions were similar, and the
1
! subjects were of the same age and strain (and from the
same supplier) in the two experiments, the 45-50 second
128
median duration of ether administration in the present
investigation (where 35% mortality resulted) was consider
ably lower than the 70 second etherization duration used
by Herz, Peeke, and Wyers (1966) (where only 25% of the
subjects failed to recover). One discrepency between the
two experiments which may have affected susceptibility to
i
ether is that animals in the present study were 22-24
i
hours water-deprived. Another explanation of the differ
ent mortality rates is that shipments of subjects from the
same supplier at different times may be differentially
susceptible to ether. Support for this hypothesis was
obtained from analyses of etherization data in which sub
jects from each batch (shipment) were grouped together.
Although the mortality rate did not appear to differ among
the various batches, duration of time to removal from
etherization revealed significant differences among the
batches. The results of this analysis raise the possibil
ity that although the batches appeared comparable behav- j
j
iorally, as assessed by the non-significant over-all j
analysis of variance of the experimental day latencies to
the first head poke and first cup touch, they may have
differed on some underlying physiological variable, a pos
sibility already considered in the context of the j
129
differences between Experiments 2 and 3. Whether such
: underlying differences exist and, if so, had any effect
on the dependent variables studied in the current investi
gation is at present a moot point subject to elucidation
by future experimentation.
Finally, the foregoing discussion of mortality
rates with the two amnesic agents suggests that a severe
degree of physiological trauma is required to produce
attenuation of both appetitive and aversive habits.
Although this is the case with many of the conditions
which produce interference with learning, e.g., traumatic
head injury or concussion, severe anoxia, hypothermia, it
is also possible to produce attenuation with cortical
spreading depression (Bures and Buresova, 1963) and with
small amounts of electrical current delivered to the
caudate nucleus (Williston, Herz, Peeke and Wyers, 1964)
or to the thalamus (Mahut, 1964), all of which appear to
result in only mild changes in ongoing, overt behavior.
; In terms of the present investigation it is of interest
I that greater attenuation was produced with the amnesic
i
| agent which resulted in the lowest mortality rate.
The failure of ether to produce retrograde amnesia
; in the aversive situation, as well as the absence of any
130
effect beyond 5 minutes after reinforcement in the appeti
tive learning paradigm when contrasted with the relatively
complete amnesia produced with ether by other experimenters
(e.g., Herz, Peeke, and Wyers, 1966; Pearlman, 1966) may
be due to several factors. First, it is extremely diffi-
!
cult to maintain a constant ether environment in the
absence of a clinical anesthesia apparatus. Temperature, j
humidity, frequency of recharging the ether chamber, and
relative airtightness of the chamber, to mention several
possibilities, may all influence the etherizing atmosphere.
In addition, the condition of the animal, i.e., suscep-
tability to respiratory diseases, schedules of food and
water deprivation, may also determine the effect of
anesthesia. Although many of these variables are not
specified in other studies, comparison of the present
aversive learning study with that of Herz, Peeke, and
Wyers (1966) reveals the only apparent difference to be the
use of water deprivation in the present study. Whether j
this factor could have influenced the degree of amnesia
i
produced remains to be determined by future research. |
j
Finally, consideration should be given to the
performance effect of ether revealed by the significant
difference between the etherized and normal control groups
131
(and the several additional cases where ether reduced
performance). Although the possibility exists that the
: attenuation produced by ether in the appetitive situation
I could have been contributed to by this effect, the magni
tude of the immediate ether effect was such that it could
not have been entirely the result of a performance-produced;
; I
response decrement. [
The results of the current series of investigations
; suggest that interference with memory with amnesic agents
is more complex than the impression warranted by many
retrograde amnesia studies involving only one measure,
usually latency. The fact that different patterns of
results were obtained with the several different measures
in the present investigations appears to indicate that
only certain aspects of the memory for an appetitive or
- aversive learning situation, specifically those most
closely associated in time with reinforcement, can be
completely abolished. In addition, the present research \
• i
indicates that appetitive and aversive learning are differ-
! entially affected by ECS and ether. Since previous
i investigations have examined only one or the other learning
; situation, with quite different types of experimental
i apparatuses and procedures, it has not previously been
i
132
possible to simultaneously examine the effects of amnesic
treatments on both. Although parametric studies are
tedious to perform, future research examining the course
of consolidation for appetitive and aversive learning,
with amnesic agents administered at numerous intervals
between immediately and one hour after reinforcement, I
would further elucidate similarities and differences.
Such studies would answer many questions relative to breaks
in the consolidation gradient such as appeared at 15 min
utes in the present experiment on appetitive learning and
as observed by Rabedeau (196 6) who obtained a reliable
i
16 minute effect but no immediate effect of spreading
depression on passive avoidance learning.
Finally, future studies of retrograde amnesia will
add greatly to our understanding of memory if they concen
trate on the temporal intervals preceding as well as
following the reinforcing event. The relationship between
the behavior, both of the central nervous system and of
the organism as a whole, which precedes and follows rein-
i
forcement and the formation of neural representations of
these events is obviously a complex one, but careful
examination of these relationships would greatly enhance
our knowledge of the process or processes which we call
memory.
CHAPTER VI
SUMMARY
The consolidation theory of memory maintains that
the memory for an experience, encoded as neural activity, j
is extremely labile and subject to external interference
for a period of time following the event. Previous
research has demonstrated that the administration of elec-
troconvulsive shock (ECS) or ether anesthesia within one
half hour of an event greatly interferes with the memory
for that experience. The present series of experiments
was designed to compare the effects of these two amnesic
agents on one-trial appetitive and aversive learning in
the same experimental apparatus.
The first experiment was concerned with the effects
i
of the two agents on the memory for a painful shock |
administered to mice when they contacted an empty drinking
i
cup in a small cul-de-sac attached to the experimental
chamber. Animals were introduced into the apparatus on
the first experimental day and the Latency to First Head
134
135
Poke and First Cup Touch recorded. Shock was received
when the cup was touched and, after a one-minute interval,
animals were removed from the chamber. Control subjects
received no shock and were either returned to their home
cages or administered immediate ECS or ether. Animals in
the experimental groups were given ether or ECS immediately
i
or 15 minutes later. On the test day, 24 hours later,
subjects were reintroduced into the chamber and the same
latencies, as well as frequencies and durations of Head
Pokes and Cup Touches and frequency of Licks recorded
during the three-minute test period. Reliable learning
was demonstrated on all of the measures, i.e., latencies
were greater and frequencies and durations lower for
shocked than for control subjects. In addition, signifi
cant interference with this learning was observed on all
but one of the measures in subjects which had received
electroconvulsive shock immediately after removal from
i
the apparatus on the preceding day. Animals administered j
i
ECS 15 minutes after the shock also showed significant [
interference on 4 of the measures. No interference was
produced with ether at either interval.
The second experiment examined the effects of the
same agents on one-trial water-reinforced learning.
Subjects were run as in the first study and the same
measures recorded, but were allowed access to water in
the drinking cup for one minute following their initial
contact with it. Control subjects received no water.
Different groups of experimental animals were administered
either ECS or ether immediately, 15, 30, or 60 minutes
after the drinking experience. Twenty-four hours later
subjects were returned to the apparatus for a three minute
test period with no water available. Reliable learning .
was observed on 6 of the measures. Almost complete
amnesia was observed on several of the measures with both
of immediate groups, but in none of the remaining ether
groups. Thirty minute administration of ECS, however,
resulted in partial attenuation of the memory for the
drinking experience on two of the measures. A slight, but
non-significant interference effect was obtained on one
of the measures with 15 minute ECS.
The final experiment replicated the baseline con
ditions of the previous study, but ether and ECS were
administered 5 minutes after removal from the apparatus.
Despite failure to obtain reliable learning on several of
the measures on which learning had been observed in the
second experiment, significant learning and reliable
137
interference effects were obtained with both experimental
groups.
The results of the experiments support a consoli
dation theory interpretation and present data on the time
gradient involved in the fixation of experience. Differ
ences among the various measures in the appetitive and
aversive learning situations were obtained, suggesting
the possibility that the consolidation of these two types
of learning may differ. Consideration was also given to
the importance of the events both preceding and following
reinforcement and their role in the establishment of
memory.
REFERENCES
REFERENCES
Abt, J.P., Essman, W.B., and Jarvik, M.E. Ether-induced
retroactive amnesia for one-trial learning in mice.
Science, . 1961, 133, 1477-1478.
Alpern, H.P. and Kimble, D.P. The retrograde amnesic
effects of diethylether and Bis(trifluorethyl)ether.
Journal of Comparative and Physiological Psychology,
in press.
Barondes, S.H., and Cohen, H.D. Puromycin effect on
successive phases of memory storage. Science, 1966,
151, 594-595.
Bures, J., and Buresova, 0. Cortical spreading depression
as a memory disturbing factor. Journal of Comparative
and Physiological^ Psychology, 1963, 56, 268-272.
Chevalier, J.A. Permanance of amnesia after a single
post-trial electroconvulsive seizure. Journal of
Comparative and Physiological Psychology, 1965, 59,
125-127.
Chorover, S.L., and Schiller, P.H. Short-term retrograde
amnesia in rats. Journal' of Comparative and Physio
logical Psychology, 1965, 59, 73-78.
Chorover, S.L. and Schiller, P.H. Re-examination of pro
longed retrograde amnesia in one-trial learning.
Journal of Comparative and Physiological Psychology,
1966, 61, 34-41.
Coons, E.E. and Miller, N.E. Conflict vs. consolidation
of memory traces to explain "retrograde amnesia" pro
duced by ECS. Journal of Comparative and Physiologi
cal Psychology, 1960, 53, 524-531.
139
140
Corson, J.A. Memory as influenced by a single electro-
convulsive shock. Journal of Psychiatric Research,
1965, 3, 153-158.
Deutsch, J.A. Higher nervous function: The physiological
bases of memory. Annual Review of Psychology, 1962,
24, 259-286.
Duncan, C.P. The retroactive effect of electroshock on
learning. Journal of Comparative and Physiological
Psychology, 1949, 42, 32-44.
j
Essman, W.B. , and Jarvik, N.E. Impairment of retention
for a conditioned response by ether anesthesia in mice.
Psychopharmacalogia, 1961, 2, 172-176.
Plexner, J.B., Flexner, L.B., and Stellar, E. Memory in
mice as affected by intracerebral puromycin. Science,
1963, 141, 57-59.
Gerbrandt, L.K. Dissociation of conditioned emotional and
avoidance responses due to ECS. Psychonomic Science,
1965, 2, 385-386.
Gerbrandt, L.K., and Thompson, C.W. Competing response
and amnesic effects of electroconvulsive shock under
extinction and incentive shifts. Journal of Compara
tive and Physiological Psychology, 1964, 58, 208-211.
Glickman, S.E. Perseveration, neural processes and con
solidation of the neural trace. Psychological
Bulletin, 1961, 58, 218-233.
Gutakunst, R., and Youniss, J. Interruption of imprinting !
following anesthesia. Perceptual and Motor Skills, j
1963, 16, 348.
Hebb, D.O. Organization of Behavior. New York: Wiley,
1949.
Heriot, J.T., and Coleman, P.D. The effect of electrocon
vulsive shock on retention of a modified "one-trial”
conditioned avoidance. Journal of Comparative and
Physiological Psychology, 1962, 55, 1082-1084.
141
Herz, M.J. The effects of ether on the retention of a
one-trial avoidance response. Unpublished M.A. thesis,
San Francisco State College, 1962.
Herz, M.J., Peeke, H.V.S., and Wyers, E.J. Amnesic
effects of ether and electroconvulsive shock in mice.
Psychonomic Science, 1966, 4, 375-376.
Hudspeth, W.J., McGaugh, J.L., and Thompson, C.W. Aversive
and amnesic effects of ECS. Journal of Comparative
and Physiological Psychology, 1964, 57, 61-64.
i
!
Hunt, H.F., and Brady, J.B. Some effects of electrocon
vulsive shock on a conditioned emotional response
("anxiety"). Journal of Comparative and Physiological
Psychology, 1951, 44, 88-98.
Leonard, D.J. and Zavala, A. Electroconvulsive shock,
retrograde amnesia and the single shock method.
Science, 1964, 146, 1073-1074.
Leukel, F. A comparison of the effects of ECS and anes
thesia on acquisition of the maze habit. Journal of
Comparative and Physiological Psychology, 1957, 50,
300-306.
Leukel, F., and Quinton, E. Carbon dioxide effects on
acquisition and extinction of avoidance behavior.
Journal of Comparative and Physiological Psychology,
1964, 57, 267-270.
Lewis, D.J., and Maher, B.A. Neural consolidation and
electroconvulsive shock. Psychological Review, 1965,
72, 225-239.
McGaugh, J.L. Time dependent processes in memory storage, j
Unpublished manuscript, 196 6. j
I
McGaugh, J.L., and Madsen, M.C. Amnesic and punishing
effects of ECS. Science, 1964, 144, 182-183.
McGaugh, J.L., and Petrinovich, L. Effect of drugs on
learning and memory. International Review of Neuro
biology , 1965, 8, 139-196.
142
McGaugh, J.L. , and Petrinovich, L. Neural consolidation
and electroconvulsive shock reexamined. Psychological
Review, in press.
i Mahut, H. Effects of subcortical electrical stimulation
on discrimination learning in cats. Journal of Com
parative and Physiological Psychology, 1964, 58,
390-395.
Otis, L.S., and Cerf, J.A. The effects of heat narcosis
on the retention of a conditioned avoidance response
in goldfish. American Psychologist, 1958, 13, 419
(abstract).
Pearlman, C.A., Jr. Similar retrograde amnesic effects of
ether and spreading cortical depression. Journal of
Comparative and Physiological Psychology, 1966, 61,
306-308.
Pearlman, C.A., Sharpless, S.K., and Jarvik, M.E. Retro
grade amnesia produced by anesthetic and convulsive
agents. Journal of Comparative and Physiological
Psychology, 1961, 54, 109-112.
Pribram, K. A review of theory in physiological psychol
ogy. Annual Review of Psychology, 1960, 11, 1-40.
Quartermain, D., Paolino, R.M. and Miller, N.E. A brief
temporal gradient of retrograde amnesia independent of
situational change. Science, 1965, 149, 1116-1118.
Rabedeau, R.G. Retrograde amnesia due to spreading corti
cal depression: Paradoxical effect of shock— SD
interval. Psychonomic Science, 1966, 5, 113-114.
: Russell, W.R., and Nathan, P.W. Traumatic amnesia. Brain,
; 1946, 69, 280-300.
j Tenen, S.S. Retrograde amnesia from electroconvulsive
shock in a one-trial appetitive learning task.
| Science, 1965, 148, 1248-1250. (a)
: Tenen, S.S. Retrograde amnesia. Science, 19 65, 149,
1521. (b)
143
Weissman, A. Effect of electroconvulsive shock intensity
and seizure pattern on retrograde amnesia in rats.
Journal of Comparative and Physiological Psychology,
1963, 56, 806-810.
Weissman, A. Retrograde amnesia effect of supramaximal
electroconvulsive shock on one-trial acquisition in
rats: A replication. Journal of Comparative and
Physiological Psychology, 1964, 57, 248-250.
Williston, J.S., Herz, J.J., Peeke, H.V.S., and Wyers, E.J.
Disruption of short-term memory by caudate stimula
tion. American Psychologist, 1964, 19, 502 (abstract).
APPENDICES
A. MEANS AND MEDIANS FOR CONTROL GROUPS (EXPERI
MENTS 1 AND 2), SHOCK GROUPS (EXPERIMENT 1)
B. MEANS AND MEDIANS FOR WATER-REINFORCED GROUPS
(EXPERIMENT 2)
C. MEANS AND MEDIANS FOR EXPERIMENT 3
APPENDIX A
MEANS AND MEDIANS FOR CONTROL GROUPS (EXPERIMENTS
1 AND 2), SHOCK GROUPS (EXPERIMENT 1)
1U6
APPENDIX A
MEANS AND MEDIANS FOR CONTROL GROUPS (EXPERIMENTS 1 AND 2) , SHOCK GROUPS (EXPERIMENT 1)
Head Poke Latency Cup Touch Latency Licks Head CuP Touch Poke Cup Touch
Group j - j ay ^ Qay 2 Difference Day 1 Day 2 Difference Pokes Sessions Duration Duration
a
X 18..32 22..39 4 .0 7 22..46 35.86 13.39 15,.41 7,.50 3..41 18.46 6 .98
a
m dn. 15.,50 12.,00 - 2 .0 0 19..75 24.25 3.75 11.,50 7..00 4.,00 18.75 5,.75
H
a
X 12..14 35..66 23.52 17.,23 60.25 43.02 14,,04 6,.00 3..32 13.75 5,.95
m dn. 11.,00 17..00 4.75 13.,25 24.75 10.00 9..50 6..00 3.,00
13.75 5,.25
w
u
X 14..25 13.,48 - .77 17.,36 . 35.50 18.14 16,.96 7,.41 3..95 15.07 7,.48
w
a
m dn. 11.,50 11.,00 - 0 .2 5 12..75 15.50 4.50 16..50 7,.00 4..00 14.75 6,.25
X 17..52 134..54 117.0 2 23.,68 160.15 136.48 1.,32 1,.27 0.,45 1.80 0,.89
S
cq
mdn. 15.,50 180..00 155.75 23.,50 180.00 151.59 0..0 0,.0 0..0 0.0 0..0
i —1
X 17..41 116.,14 98.73 19.,59 148.14 128.54 1.,14 1..77 0..41 2.32 0.,69
W
C O
mdn. 16.,50 180..00 153.25 17.,75 180.00 158.50 0,,0 0..0 0..0 0.0 0..0
i — !
CO
o
X 12.,39 73..68 61.30 15..77 120.46 104.68 4.,91 3.,59 1.,41 5.36 1..78
M
C O
m dn. 10.,25 62..00 45.50 12..50 136.00 112.50 3..00 3..00 1..00 4.12 1..25
in
rl
X 17.,93 144..84 126.91 21.,90 172.73 150.82 0,.18 0..73 0..09 0.67 0.,08
w
C O m dn. 14..00 180..00 160.25 20..00 180.00 160.00 0..0 0..0 0..0 0.0 0..0
m
i —i
CO
X 14.,54 95..04 80.50 20..73 144.46 . 123.73 0..73 1.,68 0..45 1.82 0.,28
u
H
C O
m dn. 13..50 106,.25 88.50 19,.25 180.00 156.25 0..0 1.,50 0..0 1.00 0.,0
APPENDIX B
MEANS AND MEDIANS FOR WATER-REINFORCED GROUPS
(EXPERIMENT 2)
APPENDIX B
MEANS AND MEDIANS FOR WATER-REINFORC]
Head Poke Latency Cup Touch Latency
Group
Day 1 Day 2 Difference Day 1 Day 2 Dif ferenci
X 11.43 9.34 -2.09 12.66 12. 25 -0.41
1
mdn. 9 .75 8.25 -2.50 9.75 12.25 -0.75
i —1
M
X 14.04 1 1 . 0 0 -3.04 17.20 30.10 12.91
&
mdn. 12.00 10.50 -2.50 14.25 20 .50 5.75
i —i
Ui
u
X 17.27 10.36 -6.91 22.18 28 .64 6.46
mdn. 17.75 9 .00 -8.25 20.50 20.00 -3.75
m
i —I
!s
X 16 .96 9. 89 -7.07 18.54 21.43 2.89
mdn. 13.75 9.50 -1.75 14. 00 12.75 -1.75
in
i —I X 15.50 10.93 -4.57 17.38 11.59 -5. 80
WECS
mdn. 12.50 6.50 -5.75 14.25 8.50 -5.75
o X 15.29 7.98 -7. 32 17.84 11.54 -6. 30
mdn. 12.75 5.00 -7.75 15.25 9.25 -3.50
o
ro
X 11.91 8.14 -3.76 15.70 12.93 -2.77
W
u
w
&
mdn. 10 .50 5.25 -3.75 14.25 11.50 -3.50
o
VO
m
3=
X 15.84 8.39 -7.45 18.46 10 . 81 -7.64
mdn. 12. 25 8.25 -2.00 14.25 10.00 -2.75
o
VO
X 14.86 7.32 -7.54 17.75 11. 25 -6.50
WECS
mdn. 13.00 7.00 -5.75 15.75 9 .00 -4.75
APPENDIX B
?OR WATER-REINFORCED GROUPS (EXPERIMENT 2)
• Touch Latency
Licks
Head Cup Touch Poke Cup Touch
Day 2 Difference
Pokes Sessions Duration Duration
12. 25 -0.41 61.96 12.68 7.77 36.40 19.46
12.25 -0.75 55.50 13.00 7.50 36.50 17.00
30.10 12.91 31. 91 12.96 7.41 29.45 13.33
20 .50
5.75 30.00 13.00 8.00 28.00 11.00
28 .64 6.46 21.86 11.45 5.36 23. 73 9.54
20.00 -3.75 18.00 10 .00 5.00 23.00 9.00
21.43 2.89 56.91 12.96 6.91 31.91 15. 75
12.75 -1.75 48.50 14.00 7.00 34.00 14.25
11.59 -5. 80 53.77 11.36 7.00 31. 80 16.91
8.50 -5.75 45.50 11.00 7.50 31.00 16.25
11.54 -6. 30 53.09 14.41 8.04 35.02 15.48
9.25 -3.50 53.50 13.00 8.00 32.50 15.75
12.93 -2.77 46. 27 13.09 7.18 33. 27 15. 36
11.50 -3.50 38.50 12.00 7.50 32.50 12.75
10 . 81 -7.64 51. 27 14.32 7.86 34.48 17.99
10.00 -2.75 31.00 15.00 7.50
35.00 17.25
11. 25 -6.50 57.50 12.91
6.86
38.70 18.42
9 .00 -4.75 48. 00 13.00 7.00
36.00 16.50
APPENDIX C
MEANS AND MEDIANS FOR EXPERIMENT
150
APPENDIX C
MEANS AND MEDIANS FOR EXPERIMENT 3
Group
Head Poke Latency Cup Touch Latency
Licks
Head Cup Touch Poke Cup Touch
Day 1 Day 2 Difference Day 1 Day 2 Difference
Pokes Sessions Duration Duration
X 14.59 5.98 -7.61 16.35 12.78 -3.56 31.70 14.22 7.83 31.85 1 1 .0 0
mdn. 12.25 6.00 ~~ -6.00 13.25 8.50 -3.00 28.00 14.00 8.00 28.50 10.50
in
X 11.73 13.02 1.29 16.25 17.75 1.50 22.25 13.00 6.62 24.19 7.02
W
is
mdn. 1 1 .0 0 10.25 1.25 13.25 15.25 0.75 21.00 11.50 6.00 23.75 7.50
L D
C O
X 13.75 8.62 -5.14 15.90 13.75 -2.15 24.08 11.15 6.15 25.31 9.85
u
w
&
mdn. 13.25 7.75 -4.50 14.25 1 1 .0 0 -3.00 22.50 1 1 .0 0 6.00 24.00 8.00
a
a
X 15.25 18.48 3.23 20.02 30.08 10.06 23.80 8.25 4.21 21.23 10.77
mdn. 12.75 15.50 1.00 14.50 17.50 5.25 21.00 8.00 4.00 20.50 10.75

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

PDF

Some Factors Affecting Ucr Diminution

PDF

The Effect Of Irrelevant Environmental Stimulation On Vigilance Performance

PDF

Human Gsr Classical Conditioning And Awareness Of The Cs-Ucs Relation

PDF

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

PDF

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

PDF

Habituation Of Aggression In The Three-Spined Stickleback (Gasterosteus Aculeatus L.)

PDF

Dominance And Contiguity As Interactive Determinants Of Autonomic Conditioning

PDF

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

PDF

The role of cues in the arousal of anxiety

PDF

Effects Of Task Performance Upon The Acoustic Reflex

PDF

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

PDF

An Effect Of Verbal Conditioning On Nonverbal Behavior

PDF

Noxious Auditory Environment And Psychomotor Performance

PDF

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

PDF

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

PDF

Clinical Implications Of Verbal Learning Without Awareness: An Experimental Study

PDF

Classical Discrimination Conditioning As A Function Of Probability Of Reinforcement

PDF

The Effect Of Motor Ability Loss On Cognition And Emotion

Asset Metadata

Creator Herz, Michael Joseph (author)

Core Title The Effects Of Ether And Electroconvulsive Shock On One Trial Appetitive And Adversive Learning

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 Wyers, Everett J. (committee chair), Grings, William W. (committee member), Zimmer, Russel L. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-227605

Unique identifier UC11360260

Identifier 6700408.pdf (filename),usctheses-c18-227605 (legacy record id)

Legacy Identifier 6700408.pdf

Dmrecord 227605

Document Type Dissertation

Rights Herz, Michael Joseph

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