Effects of food-deprivation on free-operant avoidance behavior.

MISSING IMAGE

Material Information

Title:
Effects of food-deprivation on free-operant avoidance behavior.
Physical Description:
vii, 48 leaves : ill. ; 28 cm.
Language:
English
Creator:
Leander, John David, 1944-
Publication Date:

Subjects

Subjects / Keywords:
Avoidance (Psychology)   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis--University of Florida.
Bibliography:
Bibliography: leaves 45-47.
Statement of Responsibility:
By John David Leander.
General Note:
Manuscript copy.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 000558970
notis - ACY4414
oclc - 13430003
System ID:
AA00004125:00001

Full Text












Effects of Food-Deprivation on Free-Operant
Avoidance Behavior











By

JOHN DAVID LEADER


A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA IN PARTIAL
FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY








UNIVERSITY OF FLORIDA
1971






























Dedicated to my Mother and Father,
Wife and Son.













ACKNOWLEDGEMENTS
I gratefully thank Dr. E. F. Malagodi for his guidance

during my graduate school career and for the direction, sup-

port, and facilities to conduct the research in this dis-

sertation.

I also express my appreciation to Dr. Robert L. King,

Dr. Henry S. Pennypacker, Dr. Charles J. Vierck, and Dr.

Richard D. Willis for being my teachers and for serving on

my committee.

I would also like to thank many others who have con-

tributed immeasurably to my education and professional

development. These people include Dr. Merle E. Meyer, Dr.

Louis Lippman, Dr. Donald A. Dewsbury, Dr. Michael A. Milan,

Thomas Waddell, and my wife, Kathleen.


iii










TABLE OF CONTENTS

ACKNOWLEDGEMENTS:. ... iii

LIST OF FIGURES. . . .

ABSTRACT . . ii

INTRODUCTION . . 1

EXPERIMENT 1. FOOD-DEPRIVATION EFFECTS ON AVOIDANCE
BEHAVIOR MAINTAINED BY SHOCK-POSTPONEMENT IN RATS 8

METHOD.. 8

RESULTS . 10

DISCUSSION . . 16

EXPERIMENT 2. FOOD-DEPRIVATION EFFECTS ON BEHAVIOR
MAINTAINED BY A CONJOINT AVOIDANCE FIXED-INTERVAL
TIE-UT SCHEDULE . .. 20

METHOD . . 21

RESULTS 22

DISCUSSION. . 24

EXPERIMENT 3. FOOD-DEPRIVATION EFFECTS ON AVOIDANCE
BEHAVIOR MAINTAINED BY THE SHOCK-FREQUENCY-REDUCTION
PROCEDURE IN RATS . 27

METHOD. . . .. 27

RESULTS . . 29

DISCUSSION. . 33

EXPERIMENT 4. FOOD-DEPRIVATION EFFECTS ON AVOIDANCE
BEHAVIOR MAINTAINED BY SHOCK-POSTPONEMENT IN
HAMSTERS. . . 37

METHOD. . . .. 37

RESULTS AND DISCUSSION. .... ...... 37

GENERAL DISCUSSION . 40

REFERENCES . . o 45

BIOGRAPHICAL SKETCH. . .. 48
iv













LIST OF FIGURES

Figure Page

1. Response rates, shock rates, and body weights for
rat CR-3 for each session of Phase 1. . 11

2. Response rates, shock rates, and body weights for
rat CR-4 for each session of Phase 1 . 11

3. Mean response rates and ranges of the 5
criteria sessions for each half-hour interval
of the 4 hr session for CR-3 in Phase 1 .. 13

4. Mean response rates and ranges of the 6
criteria sessions for each half-hour interval
of the 4 hr session for CR-4 in Phase 1 13

5. Response rates, shock rates, and body weights for
rat CR-3 for each session of Phase 2. . 15

6. Response rates, shock rates, and body weights for
rat CR-4 for each session of Phase 2. 15

7. Mean response rates of the criteria sessions
for each half-hour interval of the 4 hr sessions
of rats CR-3 and CR-4 during Phase 2. . 17

8. Representative cumulative response records from
CR-3 during ad lib and deprived conditions. 18

9. Representative cumulative response records from
CR-4 during ad lib and deprived conditions. 18

10. Response rates, shock rates, and body weights for
rat T-l for each session of the Conjoint Avoid-
ance FI TO schedule . .. 23

11. Response rates, shock rates, and body weights for
each of the last 5 sessions of the avoidance-
alone condition for rat T-l . .. 25

12. Conditional probabilities of interresponse
intervals for rat T-l in the avoidance-alone
condition . . 25








Figure Page

13. Response rates, shock rates, and body weights for
rat R-65 for each session .. 30

14. Response rates, shock rates, and body weights for
rat R-66 for each session . 30

15. Representative cumulative records for rat R-65
during ad lib and 70% ad lib conditions 32

16. Representative cumulative records for rat R-66
during ad lib and 70% ad lib conditions 34

17. Frequency of interresponse intervals for rat
R-65 during ad lib and food-deprivation conditions. 35

18. Frequency of interresponse intervals for rat
R-66 during ad lib and food-deprivation conditions. 35

19. Response rates, shock rates, and body weights for
each of the last 5 sessions of ad lib and the food-
deprived conditions for hamsters HM-3 and HM-10 38






Abstract of Dissertation Presented to the
Graduate Council of the University of Florida in Partial Fulfillment
of the Requirements for the Degree of Doctor of Philosophy

EFFECTS OF FOOD-DEPRIVATION ON FREE-OPERANT
AVOIDANCE BEHAVIOR

By

John David Leander

August, 1971

Chairman: E. F. Malagodi
Major Department: Psychology

In Exp. 1, lever pressing was maintained by Sidman's

shock-postponement procedure under conditions of free-feed

and food deprivation with rats. In Exp. 2, responding was

maintained conjointly by Sidman's shock-postponement pro-

cedure and by response-dependent time-out from the avoidance

contingencies under the same food conditions as Exp. 1. In

Exp. 3, the effects of food deprivation were determined for

responding maintained by Herrnstein and Hineline's shock-

frequency-reduction procedure. Exps. 1-3 used rats as the

experimental animals. In order to extend the species gen-

erality of the food-deprivation effect, lever pressing by

hamsters was maintained with Sidman's shock-postponement

procedure under conditions of free-feeding and food depriva-

tion in Exp. 4. In all 4 experiments, food deprivation

resulted in a decrease in the rate of lever pressing to

avoid electric shock. Reinstatement of free-feeding con-

ditions resulted in a gain in body weight and an increase in

response rates. This food-deprivation effect was not depen-

dent upon any particular free-operant avoidance procedure

and occurred with both rats and hamsters as experimental

animals. vii
vii













INTRODUCTION
Since the advent of Sidman's (1953) free-operant

avoidance paradigm, there have been several studies using

this procedure under conditions necessitating food depri-

vation. One example is the multiple schedule procedure

employed by Herrnstein and Brady (1958). During the first

stimulus condition, food was presented on a 5 min fixed-

interval schedule (FI 5). During the third 5 min stimulus

condition, responses initially postponed shock for 20 sec

(Sidman avoidance). The second and fourth stimulus con-

ditions were Time-Out (TO) conditions and responses had no

programmed consequences. Varying the shock-postponement

interval produced changes in the response rates of both rats

during the avoidance and the two TO conditions. As the shock-

postponement interval was shortened, response rates in these

three components increased. The rate of responding during

the FI 5 condition decreased for one rat and remained un-

changed for the other. Another example is the concurrent

food-presentation and shock-avoidance procedure (Hearst,

1962; Catania, Deegan,& Cook, 1966). In Hearst's experi-

ment, chain pulling by a Rhesus monkey was reinforced on a

variable-interval 2 min schedule of food reinforcement and

lever pressing postponed shock for 10 sec. Hearst found

that stimulus generalization prior to discrimination training







was greater for the shock-avoidance-maintained response than

for the food-maintained response. After extensive discrim-

ination training the gradients could be made virtually iden-

tical. Inrthe experiment by Catania et al., a press on one

lever postponed shock and a press on a different lever re-

sulted in food presentation on a fixed-ratio schedule.

Catania et al. reported that the overall response rates main-

tained by each schedule were not systematically affected

when one or the other schedule was removed. Thus the two

concurrent behaviors were maintained relatively independent

of each other.

Both the multiple avoidance-appetitive and the concur-

rent avoidance-appetitive schedules require food deprivation

as a necessary setting-condition for the maintenance of

behavior by response-dependent food. This use of food depri-

vation suggests that it may be an important variable to mani-

pulate during chronically-maintained avoidance behavior.

Some theoreticians (Hull, 1943; Kendler, 1965) have

suggested that food deprivation should facilitate shock

escape-avoidance behavior. Hull's (1943) drive summation

hypothesis predicts that the "appetitional drive" summates

with the "drive" arising from the shock, and that this

increase in the sum of "drives" should facilitate escape-

avoidance behavior. Amsel (1950) conducted an experiment

to test this prediction. Rats were confined in a starting

box at one end of a five-foot alley with an electrifiable

grid floor. A light was turned on for 3 sec, and then,







simultaneously with opening of the start box door, the floors

of the start box and alley were electrified. Opening the

door allowed the rat to escape to a shock-free area at the

opposite end of the alley. Ten trials separated by 5 min
intertrial intervals were conducted for two days. During

training Amsel did not obtain a statistically reliable dif-

ference in running time between the group that had been food

deprived 22 hours and the non-deprived group. When the shock

was discontinued and the animals were tested in extinction,

the food-deprived group ran significantly faster than the

non-deprived group. Amsel concluded that the "appetitional

drive" summated with a conditioned drive (anxiety) to pro-

duce this effect during extinction. Ley (1965) exposed non-

deprived rats to 35 pairings of a 6 sec light illumination

with 2 sec of electric shock. The light and shock were ter-

minated simultaneously. The rats were subsequently main-

tained on a limited feeding schedule with free access to

food and water for 1 hr daily. On day 6, the animals were

tested in a hurdle-jumping task. Ten sec after being placed

in the start box, the door was opened and the light was

presented. The shock was never presented on these test

trials. When the rat jumped the hurdle, the light was ter-

minated. Day 7 was another test day. Only on the second

day of testing (day 7) were the hurdle-jump latencies for the

food-deprived group shorter than for the non-deprived group.

Thus both Amsel (1950) and Ley (1965) found that food depri-

vation facilitated an escape-avoidance response only when








tested in extinction (no-shock condition). Franchina (1966)

studied the effect of 0 or 21 hrs of food deprivation com-

bined orthogonally with 40 or 70 v of shock during a single

shock-escape trial. The rats had to jump a hurdle to escape
from the electrified start box to the shock-free compartment.

On the single trial, the 21 hr food-deprived rats had shorter

latencies than the non-deprived rats. Thus Amsel (1950), Ley

(1965), and Franchina (1966) all report food deprivation to
facilitate escape-avoidance behavior under certain conditions.
However, a recent study by Meyer, Adams, and Worthen

(1969) has reported that food deprivation up to 48 hr increases
shock-escape latencies. The rats were tested in an operant

conditioning chamber. Upon shock onset, they could terminate
the shock by depressing the lever protruding into the chamber.

After the animals had acquired the escape response, the level
of deprivation was manipulated. The response latencies were

shortest for the 0 hr deprivation condition, and longest for

the 48 hr food-deprivation condition. The latency for the
24 hr deprivation condition was intermediate between the

other two values.
While investigating the effects of inescapable shock

on the subsequent acquisition of a lever-press response to

escape electric shock, Dinsmoor (1958) deprived half of the
rats of food. He observed no effect of food deprivation on

acquisition of the escape response.
Kendler (1965), with a different theoretical approach
than Hull (1943), suggested that the effect of food depri-







vation is to lower the sensory thresholds, thus resulting

in an increased effectiveness of a given shock intensity.

Griffiths (1962) investigated the effect of food and water

deprivation on shock tolerance. The rats were placed in

the shock chamber and the DC shock voltage was increased

in 2 v steps ". until the animals indicated by vigorous

movements, vocalization, and/or excessive urination and

defecation, that the shock was noxious (p. 164)." This

shock value was determined 3 times per day for 5 days. The

results showed that 24 hr food-deprived rats tolerated higher

intensities of shock before engaging in the above behaviors

than non-deprived rats. It was suggested by Blanchard and

Blanchard (1966) that the result of Griffiths' study was

due solely to a decreased probability of shock-elicited

vocalization in the food-deprived condition. Blanchard and

Blanchard repeated Griffiths' study determining the pro-

bability of shock-elicited jumping for 12 different shock

intensities. The order of shock presentation was randomly

varied between 0.05 ma to 2.0 ma. They found no differences

between the deprived and the non-deprived groups on the

flinch or jump response measures. The shock-elicited vocal-

ization response was decreased in probability over the entire

shock-stimulus range. The deprived rats emitted fewer vocal-

ization responses at each shock intensity.

Misanin and Campbell (1969) have recently hypothesized

that the effect of various motivating conditions is to alter

the sensory capability of an organism to detect and react to







specific stimuli. In three experiments they measured the

shock-detection thresholds, shock-aversion thresholds, and

the amount of activity elicited by inescapable shock under

free-feeding and food-deprivation conditions. The shock-

detection threshold was defined as the shock intensity that

served on 50% of its presentations as an effective conditioned

stimulus for avoiding a more severe shock. In the test for

shock aversion, the rats were placed in a cage that pivoted

on an axis, and the shock was delivered when the rat was on

one side of the cage. The shock-aversion threshold was

defined as the level of shock that the rat avoided 75% of

the time. The same pivotal cage was used to determine re-

activity to electric shock. Scrambled shock was contin-

uously administered for 15 min and the number of cage cros-

sings was recorded. Misanin and Campbell concluded that

food deprivation had no effect on either the shock-detection

threshold or the aversion threshold. There was also no dif-

ference between food-deprived and non-deprived rats in their

reactivity to inescapable shock. In a fourth experiment they

manipulated the degree of food deprivation during acquisi-

tion of an avoidance response. Five seconds after presen-

tation of a light and clicker, shock was delivered until

terminated by the rat's movement to the other end of the

shuttle box. If the animal responded during the light and

clicker, the shock was avoided and the two stimuli were ter-

minated. The data showed that food deprivation had no effect

on acquisition of the avoidance response.








A synthesis of the above literature yields three reports

of negative results. The first is that food deprivation did
not affect shock-detection or shock-aversion thresholds

(Misanin & Campbell, 1969). The second is that food-depri-
vation did not modify the probability of shock-elicited acti-
vity (Misanan & Campbell, 1969), or shock-elicited flinching

and jumping (Blanchard & Blanchard, 1966). The third is
that food deprivation did not affect the acquisition of an

escape-avoidance response (Amsel, 1950; Dinsmoor, 1958;

Misanin & Campbell, 1969). The literature review also yields
three reports of positive results. First is that food depri-
vation did decrease the probability of shock-elicited vocal-

ization (Blanchard & Blanchard, 1966). Second, food depri-
vation did facilitate behavior during extinction of an avoid-
ance response (Amsel, 1950; Ley, 1965). The third is that

food deprivation of rats exhibiting chronically maintained
shock-escape behavior resulted in a decrement in escape
behavior (increased escape latencies) (Meyer et al., 1969).

The purpose of this series of experiments was to assess the
effects of food deprivation on free-operant avoidance re-
sponding chronically maintained by Sidman's (1953) shock-

postponement paradigm and Herrnstein and Hineline's (1966)
shock-frequency-reduction paradigm.













EXPERIMENT 1. FOOD-DEPRIVATION EFFECTS ON
AVOIDANCE BEHAVIOR MAINTAINED BY SHOCK POSTPONEMENT IN RATS

The purpose of this experiment was to assess the effect

of food deprivation on shock-avoidance behavior maintained

by Sidman's (1953) free-operant shock-postponement procedure.

The effect was first assessed with this procedure because

this is the most commonly used avoidance procedure utilized

in multiple avoidance-appetitive (Herrnstein & Brady, 1958)

and concurrent avoidance-appetitive schedules (Catania, Deegan,

& Cook, 1966).

Method

Subjects

Two male hooded rats, CR-3 and CR-4, approximately one

year old and 450 g in body weight were used. Both rats had

extensive training (230 hrs) on shock-postponement schedules

prior to the start of this experiment. Water was continuously

available in each animal's home cage, but unavailable during

experimental sessions. Purina lab chow was continuously

available in the home cage during the free-feeding conditions

of the experiment, but was restricted during the deprived

conditions of the experiment. The animals were fed 5.0 g per

day during food deprivation until they reached the desired

percentage of free-feeding body weight, at which time the

daily ration of food was adjusted to maintain the desired

body weight.








Apparatus

The experimental space was a Lehigh Valley Electronics

rat test cage (No. 1417) housed within a sound attenuating

chamber (No. 1417C). A Gerbrand's rat lever requiring 30 g

of force to operate was positioned 4.2 cm off the floor and

5.5 cm to the left of the stimulus panel center. Scrambled

shock was delivered by a Lehigh Valley Electronic constant-

current DC shocker and scrambler (No. 1531) to the grid floor

of stainless steel rods spaced 1.9 cm center-to-center. The

houselight was on during each session and "white" noise was

supplied to the experimental room to mask the relay pro-

gramming equipment in the adjacent room.

Procedure

Sidman's (1953) shock-postponement procedure essentially

involves two intervals as basic parameters. In the absence

of responding, there is a fixed-time interval (the shock-

shock or S-S interval) between the presentation of electric

shocks. Each response postpones the impending shock for

another fixed-time interval (the response-shock or R-S in-

terval).

Experimental sessions were 4 hr long and conducted 7

days per week.

Phase l.--The rats had an extensive history (230 hrs)

with R-S and S-S intervals of 20 sec each, shocks of 2.5 ma,

and an inescapable duration of 0.5 sec. The three experi-

mental conditions for Phase 1 were ad libitum feeding (ad lib),

food deprivation to body weights which were approximately 80%







of the ad lib weight (80% ad lib), and then unlimited feeding
allowing recovery of body weights to approximately 90% of

the ad lib weight (90% ad lib).

Phase 2.--Phase 2 was a systematic replication of Phase
1. The same rats were used, but two of the parameters were

changed. Parameters were R-S = 20 sec, S-S = 10 sec, 2.5 ma,

and 0.75 sec duration. The experimental conditions for Phase

2 were ad lib, 80% ad lib, ad lib again, and 80% ad lib again.

Between the first 80% ad lib condition and the second ad lib

condition, experimental sessions were not conducted in order

to allow the rats to regain their ad lib weights.

Results
Phase 1

Fig. 1 and Fig. 2 show the results for rats CR-3 and CR-4
respectively. The data within the vertical lines define what

in this experiment are referred to as steady-state data. The

horizontal lines indicate the means of the steady-state depen-

dent-variable values. Five sessions were used for estimating

the means for CR-3, but six were used for CR-4 because of the

greater variability exhibited in response rates and shock rates.

For both rats, as the food-deprivation condition was instituted,

response rates generally declined. Compared to the 5 days prior

to food deprivation, there was increased variability of re-
sponse rates during the transition from the ad lib condition

to the 80% ad lib condition. Approximately 5 sessions after

stabilization of body weights at 80% ad lib, the response rates

for both animals stabilized. The response rates at 90% ad lib


















0000 1% 4


RAT CR-3



*.
'='' oo


*
S @ -.* *. .' .

.-I ** **
S--- ---


*@.. j .*


s0 M i I


50 60 70
SESSION


80 90 100


Fig. 1. Response rates, shock rates, and body weights
for rat CR-3 for each session of Phase 1.


RAT CR-4


**'@'~ d.


.;;- .
.


* ** -- *
-.~ *. ~. **


50 60 70
SESSION


80 90 100


Fig. 2. Response rates, shock rates, and body weights
for rat CR-4 for each session of Phase 1.


S47C

(0 390
35C

1.5C

1.2C
I.9C




z 4
I 3


530
490
C 450
S410
3
370
.80
.60
S>-.40
m ,20
z 9




It
z 6









were intermediate to the ad lib rates and 80% ad lib rates.

The shock rates for CR-3 roughly followed a pattern opposite

to the pattern of response rates. As the body weights decreased,

the shock rates increased, and then as daily weights increased

to the 90% ad lib condition, shock rates decreased. Transi-

tion periods were characterized by a great deal of variability

in shock rates.

A different pattern of shock rates was exhibited by CR-4.

As food deprivation produced a decrease in daily weights and

response rates, the shock rates also declined. This decrease

in both response rates and shock rates indicates that there

was a change in the temporal patterning of responses. The

change to the 90% ad lib condition did not result in a change

in shock rates, although the response rates did increase.

Thus food deprivation resulted in a decrease in both rats'

response rates with an associated shock-rate increase for

rat CR-3, but decrease for rat CR-4,

In order to determine if the reduced response rates

resulting from food deprivation were associated with any

particular segment of the 4 hr session, the sessions were

divided into half-hour intervals. The mean response rates

for each half-hour interval of the steady-state sessions

are plotted in Fig. 3 and Fig. 4 for CR-3 and CR-4, respec-

tively. The vertical lines in Fig. 3 and Fig. 4 indicate

the range in response rates for each mean value plotted.

Fig. 3 shows that throughout the duration of the 4 hr session,

the mean response rates for rat CR-3 at 80% ad lib were con-


















- 5

S4
z

a


C R-3



- t"
:o .t'


100r,


t. 907.
S807.


12345678
HALF-HOUR INTERVALS


Fig. 3. Mean response rates and ranges of the 5
criteria sessions for each half-hour interval
of the 4 hr session for CR-3 in Phase 1.


Il0


-7
I-
Z
W6



"4
U:
0
0.
W4
Cr


CR-4


0 iioo

0 (0
i i I 0 907.
807.


12345678
HALF-HOUR INTERVALS


Fig. 4. Mean response rates and ranges of the 6
criteria sessions for each half-hour interval
of the 4 hr sessions for CR-4 in Phase 1.








sistently below those at ad lib. The response rates at 90%

ad lib were between those at the other two values. Fig. 4

shows that the mean response rates for rat CR-4 at 80% ad lib

were consistently lower than at ad lib. The response rates

for CR-4 obtained at 90% ad lib again occupied intermediate

values between the other two.

Phase 2

Fig. 5 and Fig. 6 show the daily weights, shock rates,

and response rates for CR-3 and CR-4, respectively, during

Phase 2. As in Fig. 1 and Fig. 2, the horizontal lines indi-

cate the means of the steady-state dependent-variable values

within the vertical lines. The inverted V's plotted in the

shock rate section of Fig. 6 indicate that the shock rates

were higher than the 1 per min ceiling of the graph.

When CR-3 was food deprived for the first time in Phase

2, a decrease in response rate was not observed. Upon return

to ad lib, the response rates were higher than during the

initial exposure to ad lib.' The second exposure to 80% ad lib

resulted in a decrease in response rate. Likewise there was

little change in shock rates during.the first manipulation,

but a reliable increase in shock rates during the second

exposure to 80% ad lib.

Rat CR-4 showed a decrease in response rates and an

increase in shock rates during both exposuresto 80% ad lib.

.Fig. 7 shows the mean response,rates for each half-

hour interval throughout the 4 hr session. The numbers at

the end of each line indicate the order of the condition.



















RAT CR-3







"I"b. ... *. ** ....*j* *


8



5 .* .........., ,i


150 160 170 180 190 200
SESSION


Response rates, shock rates, and body weights
for rat CR-3 for each session of Phase 2.


RAT CR-4




.*

'.-- 'SESSION
....-""" .... "- ..... |









SESSION


Fig. 6. Response rates, shock rates, and body weights
for rat CR-4 for each session of Phase 2.


110
110



30
00
70


I-
0
3J





IiI




a.


Fig. 5.


450
410

a30
o

.80
- 60
-.40
U .20
8
i


o 5
S 4
w








The first and third conditions for rat CR-4, the ad lib con-

ditions, show consistently higher rates throughout the entire

session than the second and fourth conditions, the 80% ad lib

conditions. The rates for the first condition (ad lib) for

rat CR-3 overlap with the rates for the second and fourth

(deprived) conditions. Rat CR-3's response rates for the

first ad lib condition were also seen in Fig. 5 to overlap

the rates from both 80% ad lib conditions.

Representative cumulative records are presented in Fig.

8 and Fig. 9. Each lever press steps the pen vertically and

shocks are indicated by a downward deflection of the pen.

The records are divided into half-hour segments and the records

shown are for the fifth, sixth, seventh, and eighth segments,

going from top-to-bottom in each figure. The cumulative records

for CR-3 in Fig. 8 show a perceptible decrease in response

rates during food deprivation. Another result of food depri-

vation was an increase in the number of shocks received from

the S-S interval. Instances of this occurring are indicated

in the cumulative records by two or more shocks occurring

close together. The cumulative records from CR-4 in Fig. 9

show decreased response rates and increased shock rates as

a result of food deprivation.
Discussion

Food deprivation resulted in decreased response rates

for CR-3 and CR-4 in both phases of Exp. 1, except during

CR-3's first ad lib exposure during Phase 2. Two lines of

evidence suggest that CR-3's response rates had not stabilized





17




















SCR-4
7 *X 0

6 o us


4-

z

SR-3

W6
z /
UJ -a -, 4
J/ AD LIB

RA 5t DEPRIVED
c 4 A

12345678
I 2 3 4 5 6 7 8
HALF-HOUR INTERVALS

Fig. 7. Mean response rates of the criteria
sessions for each half-hour interval of
the 4 hr sessions of rats CR-3 and CR-4
during Phase 2.







CR3
SESSION 180
AD LIB


10 MINUTES

Representative cumulative response records
from CR-3 during ad lib and deprived con-
ditions.


C R-4


SESSION 168
DEPRIVED





z
0


0
10 MINUTES

Representative cumulative response records
from CR-4 during ad lib and deprived con-
ditions.


Fig. 8.


Fig. 9.


cc"





19

prior to the first change from ad lib to the 80% ad lib con-

dition. One line of evidence is the fact that during the

second exposure to ad lib much higher response rates were

exhibited than during the first exposure to ad lib. The

second line of evidence is the upward trend in response rates

during the last two days of the first ad lib condition.













EXPERIMENT 2. FOOD-DEPRIVATION EFFECTS
ON BEHAVIOR MAINTAINED BY
A CONJOINT AVOIDANCE FIXED-INTERVAL TIME-OUT SCHEDULE

Exp. 1 demonstrated that food deprivation resulted in

a decrease in the rates of responding maintained by Sidman's

shock-postponement procedure. The purpose of Exp. 2 was to

extend the generality of this food-deprivation effect by

manipulating the level of food deprivation during behavior

maintained by a Conjoint Avoidance Fixed-Interval Time-Out

(Conjoint Avoidance FI TO) schedule. A conjoint schedule

(Catania et al., 1965) maintains responding on one operandum

by two or more schedules operating simultaneously. With the

Conjoint Avoidance FI TO schedule each lever press postpones

shock and the same response produces a Time-Out (TO) from

the avoidance condition on a fixed-interval schedule.

Verhave (1962) conducted a preliminary analysis of a

Conjoint Avoidance FI 15 min TO 15 min schedule. Though

very little responding occurred during the TO, the FI 15

min contingency was not able to control a typical fixed-

interval positively-accelerating pattern of responding

(Ferster & Skinner, 1957), and the avoidance performance

was not facilitated by the response-dependent TO. A response-

dependent TO when programmed on a lever separate from the

avoidance lever, did maintain lever pressing but did not

engender the positively-accelerating pattern of responding

usually associated with fixed-interval schedules.

20








After manipulating the level of food deprivation

during behavior maintained by the conjoint schedule, the

FI TO contingency was deleted. This.allowed a comparison

of the food-deprivation effect with one rat in 2 different

avoidance paradigms. The avoidance-alone condition also

allowed an analysis of the temporal patterning of responding

in terms of Anger's (1963) interresponse-time-per-opportunity

(IRT/OP) measure. The IRT/OP is a conditional probability

measure of interresponse-interval frequencies.

Method
Subject

One male hooded rat, T-l, approximately a year of age

and 450 g in body weight was used. The animal care and

apparatus was the same as in Exp. 1.

Procedure

On the Conjoint Avoidance FI TO schedule, shocks were

delivered on a S-S interval, and each response during the

avoidance condition postponed shock for the duration of the

R-S interval. A lever press by the rat, besides postponing

shock, could result in a TO from avoidance on a FI 5 min

schedule. Thus the first response made after 5 min resulted

in a 1 min TO from avoidance. The R-S interval was 10 sec

and the S-S interval was 5 sec. Shocks were 2.5 ma DC with

an inescapable duration of 0.75 sec. A houselight was on

during avoidance .and the experimental cage was totally dark

during the TO. Responses had no effect during TO and no

shocks were ever presented. Sessions were 4 hr in duration





22.


and conducted 7 days per week. Experimental manipulations

with the conjoint schedule were ad lib, 80% ad lib, and 90%

ad lib.

After analysis of the food-deprivation effect on the

conjoint schedule, the rat was exposed to a history of avoid-

ance alone. The parameters were R-S = 20 sec and S-S.= 5 sec.

Shocks were 2.0 ma AC delivered by BRS-Foringer shock generating

(SG-901) and shock scrambling (SC-901) equipment with an ines-

capable duration of 0.50 sec. In the avoidance-alone condi-

tion, the manipulations were ad lib and 80% ad lib.

Results

The data for rat T-l during the Conjoint Avoidance FI

TO condition are presented in Fig. 10. When deprivation was

instituted, there was a rapid drop in body weight to 80%

ad lib, and then when feeding was increased, the body weight

recovered to approximately 90% ad lib. The change in response

rates paralleled the weight changes. As the weight decreased,

the response rates decreased, and as the body weight increased,

the response rates increased. The response rates of T-l at

90% ad lib were intermediate between the 15-16 responses per

minute exhibited at ad lib and the 11-12 responses per minute

exhibited at 80% ad lib. During the early part of the depri-

vation condition there was a transient increase in shock rates,

but the shock rates recovered to pre-deprivation levels. There

was no systematic change in the low rates of responding present

during the TO from avoidance. The responses that occurred

during the TO were usually all within 2 sec of the initiation























RAT T-I
"-'""?..... '




*' \ AVOIDANCE

e\ /.*.







""*' ".- "' \.




00 \ 0
TIME-OUT IROM AVOIDANCE

/**^ A .....*-..., / \ .. ./
__* ^__ __, 0
0/t.c" ,\
1 1'0


200 205 210 21S
SESSIONS


220 225


Fig. 10.


Response rates, shock rates, and body weights
for rat T-l for each session of the Conjoint
Avoidance FI TO schedule.


470g
4409
4101


1.2
U.

U'

.4
S.3
0 -
U ,1
2C
3E.








of the TO. Though the TO stimuli controlled a low rate of

responding during the TO, the FI 5 TO schedule did not control

a typical fixed-interval positively-accelerating pattern of

responding.

Fig. 11 shows the data from the last 5 sessions during

the ad lib condition and the 80% ad lib condition in the

avoidance-alone paradigm. Both a decrease in response rates

and an increase in shock rates were a result of food depri-

vation. Fig. 12 presents an analysis of the conditional pro-

bability of responses occurring'in 2 sec divisions of the R-S

interval. The conditional probability of responding past the

18 sec division was not plotted in Fig. 12 since .by definition

it has to be 1.00. During the deprived condition, the pro-

bability of responding increased as the duration of the inter-

response interval increased. Responding at the ad lib weight

followed the same pattern except at the 0-2 sec division.

Thus the effect of food deprivation was to decrease the pro-

bability of a response occurring within 2 sec of a previous

response.

Discussion

Exp. 2 replicated the food-deprivation effect found in

Exp. 1 with a permutation of Sidman's shock-postponement pro-

cedure. The relatively high rates of responding (15-16/min)
were decreased markedly by food deprivation without serious

disruption of the stimulus control effecting a low rate of

responding during the TO. In the avoidance-alone condition

decreased responding was shown not to be specific to the type
















RAT T-I
,- -.-.- I I


-0-
0 ~ ,


* Ii~~


435g
a 405g

375g



.10


9


8
lU


313 344
SESSIONS


Fig. 11.


Response rates, shock rates, and body weights
for each of the last 5 sessions of the avoid-
ance-alone condition for rat T-l.


Fig. 12.


RAT T-1
o0-0 ISS 344-34i i ILIVE
A---A M-31 AD LIN


k\ -


2 4 6 8 10 12 14 16 1116e
INTItRISPOnISI IITnIV

Conditional probabilities of interresponse
intervals for rat T-l in the avoidance-
alone condition.


S

N*
0


.2

U'
.n
i *
- .


30,








of electric shock used since DC shock had been used in Exp. 1

and in the Conjoint Avoidance FI TO section of Exp. 2, while

AC was used in the avoidance-alone condition of Exp. 2.

Similar to Verhave's (1962) study, the fixed-interval

schedule of TO presentation did not produce the typical

positively-accelerating pattern of responding associated with

fixed-interval schedules. Verhave (1962) and Sidman (1962)

reported that the response dependent TO only controlled

schedule-appropriate patterns of responding when programmed

on a separate operandum and on low fixed-ratio schedules of

presentation. Thus responding was more effectively controlled

by the avoidance schedule than by the fixed-interval pre-

sentation of the relatively weaker reinforcer, the TO.













EXPERIMENT 3. FOOD-DEPRIVATION EFFECTS ON AVOIDANCE BEHAVIOR
MAINTAINED BY THE SHOCK-FREQUENCY-REDUCTION PROCEDURE IN RATS

Exp. 1 and Exp. 2 demonstrated that food deprivation
resulted in a response-rate decrement during avoidance behavior

maintained by Sidman's (1953) shock-postponement procedure

and Verhave's (1962) conjoint permutation of Sidman's pro-

cedure.

Herrnstein and Hineline (1966) introduced a procedure

which demonstrated that a response-dependent reduction in
shock frequency was a sufficient condition to generate and

maintain lever-pressing. With their procedure, responding

did not postpone shocks, it only reduced the frequency of

shock. In a later report summarizing this work and comparing

it to Sidman's shock-postponement procedure, Herrnstein (1969)

concluded that this shock-frequency-reduction paradigm was

no less effective than other free-operant avoidance procedures

in generating lever-pressing in the rat.
The purpose of Exp. 3 was to assess the effect of food

deprivation on avoidance behavior maintained by the shock-
frequency-reduction procedure of Herrnstein and Hineline (1966).

Method

Subjects

Two male hooded rats, R-65 and R-66, with an extensive
history of avoidance behavior maintained by the shock-








frequency-reduction procedure were used. The animal housing

conditions were the same as in Exp. 1.

Apparatus

The apparatus was the same as in Exp. 1 except that
the shock was AC delivered from BRS-Foringer shock generating

(SG-901) and scrambling (SC-901) units.

Procedure

Herrnstein and Hineline's (1966) shock-frequency-

reduction procedure essentially involved two separate dis-

tributions of shock delivery. One distribution, the high-

probability-shock distribution, was in effect as long as the

animal failed to depress the lever. A tape reader was stepped

every 2 sec, and the pattern of shock was varied randomly but

with a specified probability of 0.3 for every 2 sec. Thus

if no responding occurred, the rat received a Normal dis-

tribution of shocks with an average rate of 9/min. The other

distribution, the low-probability-shock distribution, was in

effect dependent upon a lever response by the animal. The

low-probability-shock distribution had a shock probability
of 0.1 for every 2 sec period. Upon a lever depression the

low-probability-shock distribution was in effect until a shock

was delivered, at which time the. high-probability-shock dis-

tribution was reinstated. Thus the lowest possible shock

rate was approximately.3/min. The first response after a

shock changed the probability of shock per 2 sec period from

0.3 to 0.1.

After stable responding was achieved under ad lib, the

rats were deprived to 80% ad lib. The deprived weights were








reached by absolute food deprivation for the first week, and

then 5 g of food per day until the desired weight was attained.

During deprivation conditions, the rats were fed enough food

after the daily sessions to maintain the desired body weight.

After stable response rates occurred at 80% ad lib, the ad

lib condition was reinstated, and then was followed by depri-

vation to 70% ad lib. Thus experimental conditions were ad

lib, 80% ad lib, ad lib, and 70% ad lib.

Sessions were 100 min in duration conducted 6 days per

week. The houselight was on for the entire session and each

response produced an audible "click" from a feedback relay.

Shocks were 0.8 ma AC with an inescapable duration of 0.3 sec.

Results

Fig. 13 presents the results for rat R-65. The horizon-

tal lines indicate the mean dependent-variable values for

the last 10 sessions (steady-state sessions) in each feeding

condition. Upon initiation of the 80% ad lib condition, there
4
was a steady decrease in response rates along with an increase

in shock rates. The second ad lib condition resulted in an

increase in response rates and a decrease in shock rates.

Subsequent food deprivation to 70% ad lib and then reinstate-

ment of ad lib conditions resulted in similar changes in re-

sponse rates and shock rates. The response rates during 70%

ad lib were lower than the rates during 80% ad lib. Con-

versely, the shock rates were higher during the 70% ad lib

condition, than during the 80% ad lib condition. There was

increased variability in both response rates and shock rates

during the deprivation conditions.




















u





I

U





U


I
F
F
L-















Fig 13


70 s0 90 100 11 1
SESSIMS


Response rates, shock rates, and body weights
for rat R-65 for each session.


4s [is


RAT 66



t -


toL'


Is.. bun,,.,


**




* S


70 0 0 do i


12o 130 140 ISO


SESSIONS


Fig. 14. Response rates, shock rates, and body weights

for rat R-66 for each session.


Asi
*


S425-

S375-

325-
-

- 275-

4-

3-

16-

S14-
U

S12-

10-




6-








Fig. 14 shows the data for rat R-66. Throughout the
entire experiment, R-66 exhibited more variability in re-

sponse rates than R-65. The first change from the ad lib
to the 80% ad lib condition did not result in a decrease in

response rates. Resumption of the ad lib condition resulted

in an increase in response rates compared to the rates exhibited
during the 80% ad lib condition. Food deprivation to 70% ad
lib resulted in a decrease in response rates and then an in-

crease in rates upon reinstatement of the ad lib condition.
Shock rates for R-66 were maintained at the minimum possible

except during deprivation to 70% ad lib.

Fig. 15 shows representative cumulative records for
rat R-65 during ad lib conditions and the 70% ad lib con- ,

edition. Oblique pips on the response pen indicate shock

presentations while the event pen indicates which shock dis-
tribution was in effect at any given point in time. When

the event pen was down (post-shock), the high-probability

shock distribution was in effect and when it was up (post-

response), the low-probability shock distribution was in
effect. Inspection of the cumulative records shows that

for R-65 there was an increase in the amount of time spent
in contact with the high-probability shock distribution

during the 70% ad lib condition.

Fig. 16 presents representative cumulative records for
rat R-66. Recording specifications are the same as in Fig.
16. In contrast to R-65's records in Fig. 15, there was no

increase in the amount of time spent in contact with the






















$ISSIOII 63
WIT 474g



__


RAT 65


SISSIO 139
WSI 234e


SESSIIO 140
WIT 313.


SSS11i lI"
WIT 4491


Fig. 15. Representative cumulative records for rat
R-65 during ad lib and 70% ad lib conditions.


_ _~__~__~____i_ i






high-probability shock distribution, but there was a marked

decrease in response rates during the 70% ad lib condition.

Inspection of these records indicates that the main effect

of food deprivation was to decrease the duration of post-

shock response "bursting." Analysis of the frequencies of

interresponse intervals in Fig. 17 and Fig. 18 for R-65 and

R-66, respectively, substantiates this conclusion. The filled

symbols in Fig. 17 and Fig. 18 are the ad lib conditions and

the open symbols are the deprived conditions. Food depri-

vation for both animals resulted primarily in a decrease in

the frequencies of responses occurring within 2 sees of a

previous response.

Discussion
Exp. 3 demonstrated that the food-deprivation effect
found in Exp. 1 and Exp. 2 with Sidman's shock-postponement

procedure was not specific to that procedure. Thus, exten-

sion of the effect to Herrnstein and Hineline's (1966) shock-

frequency-reduction procedure increases the generality of

the food-deprivation effect. It also supports Herrnstein's

(1969) conclusion that the shock-frequency-reduction pro-

cedure contains many functional similarities with Sidman's

(1953) shock-postponement procedure.
The analysis of the frequencies of interresponse inter-

vals indicated that deprivation effected a decreased in re-

sponses occurring within 2 sec of a previous response. This

was the same conclusion that the IRT/OP analysis of T-l's

responding indicated during the avoidance-alone condition














SESSION lit RAT 66
WOT 431g



10 MINUTES








SESSION 141
W8T 301#













SESSION 142












SESSION 159
WOT 4109













Fig. 16. Representative cumulative records for rat
R-66 during ad lib and 70% ad lib conditions.







RAT 65


SESSIONS
D
A


63-72
97-106
110-119
135-144
151-160


-i



'a-
u I
o20



S100
z
I,

Z
,a
U


2 4 6 10 12 14 '16 18
IRT CLASSES


Fig. 17.


Frequency of interresponse intervals for
rat R-65 during ad lib and food-deprivation
conditions.


RAT 66


* SESSIONS 64-73
a 97-106
S 110-119
A .134-143
* 15-159


I-


I-


2 4 6 8 10 12 14 16 18
IRT CLASSES


Fig. 18.


Frequency of interresponse intervals for
rat R-66 during ad lib and food-deprivation
conditions.


300 -


0 pa
.-- <-^ J-i. *


001-'A


1001-


a
M 700


" 650




s. 100
0






Ml
S
11

U





36


of Exp. 2. Since the food-deprivation effect primarily re-

duced the number of responses occurring within 2 sees of a

previous response, this explains why there can be large re-
ductions from high response rates without increasing the

shock rates. Both with T-l (Exp. 2) and R-66 (Exp. 3) re-

sponse rates were reduced markedly without an appreciable

increase in shock rates.













EXPERIMENT 4. FOOD-DEPRIVATION EFFECTS ON AVOIDANCE BEHAVIOR
MAINTAINED BY SHOCK-POSTPONEMENT IN HAMSTERS

Although Exps. 1, 2 and 3 assessed the effects of food

deprivation with several different procedures, with several

different parameters, and with several different rats, the fol-

lowing question remained unanswered. Were the effects obtained

in Exps. 1, 2 and 3 specific to rats? The purpose of Exp. 4

was to assess the effect of food deprivation on avoidance

behavior maintained by Sidman's (1953) shock-postponement

procedure in a different rodent than rat--the hamster.
Method

Two adult male hamsters, HM-3 and HM-10, each weighing

approximately 140 g were used. The same apparatus and pro-

cedures as in Exp. 1 were utilized. Experimental sessions

were 60 min in duration and were conducted six days per week

during the active phase of the hamster's diurnal cycle--the

dark phase. Shocks were 1.0 ma AC for an inescapable dura-

tion of 0.2 sec. Schedule parameters were R-S = 20 and S-S =

5 sec. The experimental conditions were ad lib, 80% ad lib,

and return to ad lib. During the deprived conditions, the

hamsters were fed 5 g of food per day until the desired

weight was attained.
Results and Discussion

Fig. 19 presents the data from the last 5 sessions

during ad lib, 80% ad lib, and ad lib again. Response rates
37
















I HM 3


w


- a aI


0/
0


DRI AD LIB
DEPRIVED


S00o
S,o'
0


SHOCKS
MIN


*-O"


- |


140-

S130-

g120-

1l0






6-


AD LISB I
DEPRIVED


LAST 5 SESSIONS


Fig. 19.


Response rates, shock rates, and body weights
for each of the last 5 sessions of the ad lib
and the food-deprived conditions for hamsters
HM-3 and HM-10.


/0
000
0


m1










/IV*


V tESPONSIS
MIN


0
o.


o


o
0 0
0






AD LIS


O'A



AD LIB








are indicated by the filled symbols and shock rates are indi-

cated by the open symbols. HM-3 exhibited a poor level of

performance as indicated by the high shock rate during the

ad lib conditions. Food deprivation resulted in a small, but

reliable decrement in response rates and an increase in the

already high shock rates. Subsequent reinstatement of the

ad lib condition resulted in an increase in response rates

and a decrease in shock rates.

Food deprivation of HM-10 resulted in a small decrease

in response rates while more than doubling the shock rates.

Reinstatement of the ad lib condition resulted in an increase

in response rates and recovery of the pre-deprivation shock

rates.

Exp. 4 replicates the previous experiments with rats and

illustrates that the food-deprivation effect on free-operant

avoidance behavior is not specific to rats.













GENERAL DISCUSSION

These four experiments demonstrate that food deprivation

results in a decrease in the response rates of free-operant

avoidance behavior. The effect was demonstrated with

Herrnstein and Hineline's (1966) shock-frequency-reduction

procedure and Sidman's (1953) shock-postponement procedure,

and was not dependent on any particular procedure or set of

parameters maintaining avoidance behavior. The species gen-

erality was also assessed by demonstrating the food-depriva-

tion effect with both rats and hamsters. The decrement was

most impressive when the rates of avoidance responding were

high and response-rate decreases could occur without markedly

increasing the shock rates. Rat T-l's performance during

the conjoint schedule in Exp. 2 (Fig. 10) was the best ex-

ample of an impressive decrease in response rates without a

marked increase in shock rates. The small decrease in re-

sponse rates exhibited in Exp. 4 by hamsters HM-3 and HM-10

(Fig. 19) was probably more of a result of the poor avoidance

baseline during ad lib food than any species differences in

susceptibility to the food-deprivation effect.

This food-deprivation effect could not be due to severe

deprivation bordering on physical incapacitation because of

three lines of evidence. First, with appetitively-maintained

schedules, it is quite common to maintain rats at 70% of the








free-feeding weight or below. For hooded rats similar to those

used in Exps. 1, 2 and 3, an average running-weight for main-

taining schedule performance with food would be 300-325 g. A

second line of evidence is that Moskowitz (1959) has found in-

creases in running-wheel activity, rather than decreases in

activity, with deprivation to 60% of the ad lib-feeding weight.

The activity levels were graded and increased as weight was

decreased from 90% to 60% ad lib. The third line of evidence

against an incapacitation type of explanation is the finding

from these present experiments of a grading of response rates

as the level of deprivation was increased. For instance, in

Phase 1 of Exp. 1, for both CR-3 and CR-4, intermediate re-

sponse rates were obtained at intermediate weight levels.

This gradation of rates with body weight was also found with

rat T-l during the conjoint schedule and rat R-65 in the

shock-frequency-reduction procedure.

In the present research, level-of-food-deprivation was

defined in terms of the organism's body weight. This defin-

ition was used because it is the most common definition in

studies chronically maintaining behavior by response-depen-

dent food presentation. All of the research on food depri-

vation reviewed in the previous introduction used hours-since-

free-access-to-food as the definition of level-of-food-depri-

vation. None of the studies used more than 24 hrs of food

deprivation, except the Meyer et al. (1969) study which used

48 hrs. Twelve or even 24 hrs of food deprivation does not

result in very deprived animals when compared to a 20% reduc-








tion in body weight. This, coupled with the inherent varia-

bility present in response-acquisition experiments, could

explain the lack of positive food-deprivation results in

escape-avoidance experiments (Amsel, 1950s Dinsmoor, 19581

Misanin & Campbell, 1969). The one experiment by Meyer et al.

(1969) that found reliable effects of food deprivation on

shock-escape behavior used 24 and 48 hrs of food deprivation.

Calculations from the weight data reported show that 48 hrs

of food deprivation reduced the average weight to 85% ad lib

(from 285 g to 240 g). Thus the study by Meyer et al. sup-

ports the findings of the present research--food deprivation

that produces a weight loss results in a decrease in escape-

avoidance performance.

Catania et al. (1966) maintained their monkeys at 80%

ad lib while Herrnstein and Brady (1958) maintained their rats

at 60% of their 200-day weights. It can now be concluded that

besides guaranteeing schedule control by the schedule of food

reinforcement, the weight control prevented changes in the

level-of-food-deprivation to effect changes in the free-operant

avoidance performance.

The results of the present research have two important

implications. The first, is that any long-term study of

free-operant avoidance behavior will require the monitoring

and, if necessary, controlling of the experimental organism's

body.weight in order to reduce food-deprivation and weight-

change effects from modifying the avoidance performance.








The second implication is directed towards research with

avoidance behavior and physiological interventions in the

hypothalamus.

The lateral and ventral-medial areas of the hypothal-

amus have been shown to be very important areas for the init-

iation, maintenance, and cessation of feeding behavior (see

Morgane & Jacobs, 1969, for a review). These two hypothal-

amic areas have also been demonstrated to be very important

for shock-avoidance behavior. Lesions of the medial forebrain

bundle of the lateral hypothalamus produce an increased sen-

sitivity to electric shock (Harvey & Lints, 1965). Lesions

in this area also interrupt feeding behavior, producing animals

that for varying periods of time do not initiate feeding be-

havior and thus lose weight drastically (Anand & Brobeck,

1951). Lesions of the ventral-medial area of the hypothalamus
have been reported to facilitate two-way shuttle-box avoidance

(Grossman, 1966). These lesions also disrupt the cessation

of feeding behavior and the lesioned rats become very obese

(Hetherington & Ranson, 1939). Chemical and electrical stim-

ulation studies have also implicated both feeding behaviors

and avoidance behaviors with these hypothalamic areas (Carder,

1970; Grossman, 1966; Miller, 1957; Sepinwall, 1969).
This food-deprivation effect on avoidance behavior sug-

gests that during hypothalamic investigations, the weights of

the experimental animals will have to be controlled. Weight
control should facilitate the determination of whether avoid-

ance effects from hypothalamic intervention are secondary to







changes in feeding behavior and weight changes, or are more

primary effects of the hypothalamic interventions themselves.

The food-deprivation decrease in response rates obtained

in this study does not support Hull's (1943) drive summation

hypothesis or Kendler's (1965) suggestion that food depri-

vation lowers sensory thresholds. These data also contradict

data presented by Pare (1969) concerning aversion thresholds

in the rat. He defined the aversion threshold as that inten-

sity of shock that was avoided 75% of the time in a tilt cage.

He compared young and old, and male and female rats. The only

measure that related to the aversion threshold was the body

weight of the experimental animal. "Light rats have lower

thresholds to grid shock, whereas heavy rats have higher

thresholds (Pare, 1969, p. 217)." If lower thresholds were a

result of food deprivation, higher response rates should have

been observed, since response rate is positively correlated

with shock intensity (Riess, 1970). This discrepancy needs

to be resolved by further research.













REFERENCES


Amsel, A. The combination of primary appetitional need with
primary and secondary emotionally derived needs. Journal
of Experimental Psychology, 1950, 40, 1-14.

Anand, B. K. & Brobeck, J. R. Localization of a 'feeding
center' in the hypothalamus of the rat. Proceedings
of the Society of Experimental Biology and Medicine,
1951, 7722. 323-324 .
Anger, D. The.role of temporal discrimination in the rein-
forcement of Sidman avoidance behavior. Journal of
the Experimental Analysis of Behavior, 1963, 6 477-506.

Blanchard, R. J. & Blanchard, C. Food deprivation and reac-
tivity to shock. Psychonomic Science, 1966, 4, 317-318.

Carder, B. Lateral hypothalamic stimulation and avoidance
in rats. Journal of Comparative and Physiological
Psychology, 1970, 71, 325-333.

Catania, A. C., Deegan, J. F., & Cook, L. Concurrent fixed-
ratio and avoidance responding in the squirrel monkey.
Journal of the Experimental Analysis of Behavior, 1966,
,2 227-231.

Dinsmoor, J. A. Pulse duration and food deprivation in
escape-from-shock training. Psychological Reports, 1958,
4, 531-534.
Ferster, C. B. & Skinner, B. F. Schedules of Reinforcement.
New York: Appleton-Century-Crofts, 1957.

Franchina, J. J. Combined sources of motivation and escape
responding. Psychonomic Science, 1966, 6, 221-222.

Griffiths, W. J., Jr. Effect of food and water deprivation
on shock tolerance of albino rats. Psychological
Reports, 1962, 11, 163-166.

Grossman, S. P. The VMHi A center for affective reactions,
satiety or both. Physiology and Behavior, 1966, 1,
1-10.

Harvey, J. A. & Lints, C. E.. Lesions in the medial forebrain
bundle Effects on sensitivity to electric shock.
Science, 1965, 148, 250-252.








Hearst, E. Concurrent generalization gradients for food-
controlled and shock-controlled behavior. Journal of
the Experimental Analysis of Behavior, 1962, I, 19-31.

Herrnstein, R. J. Method and theory in the study of avoid-
ance. Psychological Review, 1969, 76, 49-69.

Herrnstein, R. J. & Brady, J. V. Interaction among compon-
ents of a multiple schedule. Journal of the Experi-
mental Analysis of Behavior, 1958, 1, 293-300.

Herrnstein, R. J. & Hineline, P. N. Negative reinforcement
as shock-frequency reduction. Journal of the Experi-
mental Analysis of Behavior, 1966, -, 421-430.

Hetherington, A. W. & Ranson, S. W. Experimental hypothala-
mico-hypophyseal obesity in the rat. Proceedings of
the Society for Experimental Biology and Medicine,
1939, 46, 465-66.
Hull, C. L. Principles of Behavior. New York: Appleton-
Century-Crofts, 1943.

Kendler, H. H. Motivation and behavior. In D. Levine (Ed.),
Nebraska Symposium on Motivation. Lincoln: University
of Nebraska, 1965.

Ley, R. Effects of food and water deprivation on the per-
formance of a response motivated by acquired fear.
Journal of Experimental Psychology, 1965, 69, 583-589.

Meyer, M. E., Adams, W. A., & Worthen, U. K. Deprivation
and escape conditioning with various intensities of
shock, Psychonomic Science, 1969, 14, 212-214.

Miller, N. E. Experiments on motivation. Science, 1957,
126, 1271-1278.

Misanin, J. R. & Campbell, B. A. Effects of hunger and
thirst on sensitivity and reactivity to shock. Journal
of Comparative and Physiological Psychology, 1969, 69,
207-213.

Morgane, P. J. & Jacobs, H. L. Hunger and satiety. World
Review of Nutrition and Dietetics, 1969, 10, 100-213.

Moskowitz, M. J. Running-wheel activity in the white rat
as a function of combined food and water deprivation.
Journal of Comparative and Physiological Psychology,
1959, 52, 621-625.







Pare, W. P. Age, sex, and strain differences in the aversive
threshold to grid shock in the rat. Journal of Compara-
tive and Physiological Psychology, 1969, 697 214-218.

Riess, D. Sidman avoidance in rats as a function of shock
intensity and duration. Journal of Comparative and
Physiological Psychology, 1970, 73, 481-485.

Sepinwall, J. Enhancement and impairment of avoidance
behavior by chemical stimulation of the hypothalamus.
Journal of Comparative and Physiological Psychology,
1969, 68, 393-399.
Sidman, M. Avoidance conditioning with brief shock and no
exteroceptive warning signal. Science, 1953, 118,
157-158.
Sidman, M. Time out from avoidance as a reinforcer: A
study of response interaction. Journal of the Experi-
mental Analysis of Behavior, 1962, 1, 423-4_3-.

Verhave, R. The functional properties of time out from an
avoidance schedule. Journal of the Experimental
Analysis of Behavior, 1962, 7, 391-422.













BIOGRAPHICAL SKETCH

John David Leander was born on April 8, 1944, at Mount

Vernon, Washington. He attended public school in Mount Vernon

and graduated from Mount Vernon Union High School in June, 1962.

In September, 1962, he enrolled in Pacific Lutheran University

in Tacoma, Washington. During the summers between 1962 and

1966, he was employed by the Stokley-VanCamp Company in Mount

Vernon. During the school years, he was employed as a bus

driver for the Clover Park School District in Tacoma, Wash-

ington. In May, 1966, he received the Bachelor of Arts degree

from Pacific Lutheran University. In September, 1966, he

enrolled in the graduate school of Western Washington State

College in Bellingham, Washington, ,and received the Master

of Arts degree from that institution in August, 1967. In

September, 1967, he enrolled in the graduate school of the

University of Florida. He served.as a teaching assistant

to Dr. C. M. Levy and as an interim instructor in Experimental
Psychology. During the last two years of graduate study, he

was the recipient of a fellowship from the Center for Neuro-

biological Sciences in the College of Medicine at the Uni-
versity of Florida. He is married to the former Kathleen

Axelson and they have one son, Sven.









I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.



E. F. Malagoda, C airman
Assistant Prof s or/of Psychology

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.



Robert L. King (\
Associate Professor of Phyiology

I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is'fully adequate in scope and quality,
as a dissertation for the degree of Doc r of Philosophy.



h ryy'. Penn packer '
Professor of Psychology


I certify that I have read this study and that in my
opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.



Aharles J. ierck
Associated rofessorof Neurosciences









I certify that I have read this study and that in my
opinion it confor- to acceptable standards of scholarly
presentation and is fully adequate, in scope and quality,
as a dissertation for the degree of Doctor of Philosophy.



Richard D. Willis'
Assistant Professor of Psychology




This dissertation was submitted to the Department of
Psychology in the College of Arts and Sciences and to the
Graduate Council, and was accepted as partial fulfillment
of the requirements for the degree of Doctor of Philosophy.


Dean, Graduate School






































UNIVERSITY OF FLORIDA
111111IIII 2 HIII 05 111511111 1 iII 11111 iII11I
3 1262 08553 6752