University of Southern California Dissertations and Theses

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

(USC Thesis Other)

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

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content t 71-27,958
| SOLDOFF, Susan, 1942-
| OPERANT DISCRIMINATION OF AN INTEROCEPTIVE
! STIMULUS IN THE URINARY BLADDER OF INTACT
| AND DORSAL ROOT TRANSECTED FEMALE RHESUS
| MONKEYS.
| University of Southern California, Ph.D., 1971
' Psychology, experimental
| University Microfilms, A XEROX C om pany, Ann Arbor, Michigan
m: , .
THIS DISSERTATION HAS BEEN MICROFILMED EXACTLY AS RECEIVED
OPERANT DISCRIMINATION OF AN INTEROCEPTIVE STIMULUS
THE URINARY BLADDER OF INTACT AND DORSAL ROOT
TRANSECTED FEMALE RHESUS MONKEYS
by
Susan Soldoff
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)
June 1971
UNIVERSITY O F SO U T H E R N CALIFORNIA
TH E GRADUATE SC H O O L
U N IV ER SITY PA RK
LO S A N G ELE S, C A L IF O R N IA 9 0 0 0 7
This dissertation, written by
............................ SUSAN. S0LP.QEZ..............................
under the direction of //.®.E..... Dissertation Com
mittee, and approved by all its members, has
been presented to and accepted by The Gradu
ate School, in partial fulfillment of require
ments of the degree of
D O C T O R OF P H I L O S O P H Y
Dean
Date JUNE..19Z1
DISSERTATION COMMITTEE
Chairman
DEDICATION
I dedicate this dissertation
to my beloved parents, Mr. and Mrs.
Charles M. Ross,
ii
ACKNOWLEDGMENTS
I wish to thank my dissertation committee, my
chairman, Dr. Gary C. Galbraith, and members, Drs. Henry
Slucki and Arnold Dunn, for their guidance and encourage
ment, Special thanks go to Dr. Slucki for his invaluable
assistance in the development and execution of the study
and for his critical readings of the manuscript, I also
wish to thank the other members of my guidance committee,
Drs. Donald J, Lewis and Milton Wolpin, for their
assistance in preparing for the dissertation.
I especially wish to express my sincere gratitude
to Dr. Robert W, Porter, Associate Chief of Staff,
Veterans Administration Hospital, Long Beach, for his
generosity in allowing me the use of his laboratory,
equipment and personnel; to Mr. George Dillard, for
technical laboratory assistance; to Mr. Frank B, McCoy and
Mr. David Wright, for their much appreciated general
assistance in carrying out the experiment; to Mr. Nathan
Fruchter, for his invaluable help in apparatus construc
tion; to Dr. Martin Krell, for performing the surgeries,
for making the post-mortem examinations, and for his
interest in my work; to Mr. Timothy Dodge, for preparation
of photographs; and to all other V. A. staff who have been
so helpful.
rii.
Also, thanks go to my friends, expecially Mrs. Leah
C. Berman, for her aid in the preparation of data for
analysis. Finally, I wish to thank my patient and
understanding husband, Steve, for his assistance and
encouragement throughout the execution of the study and
in the preparation of the final manuscript.
TABLE OF CONTENTS
Page
DEDICATION............................................. ii
ACKNOWLEDGMENT........................................... iii
LIST OF T A B L E S ........................................ vi
LIST OF FIGURES..........................................viii
ABSTRACT ............................................... x
Chapter
I. INTRODUCTION .................................. 1
II. M E T H O D ........................................ 13
III. RESULTS........................................ 26
IV. DISCUSSION.................................... 72
V. SUMMARY......................................... 95
REFERENCES............................................. 98
APPENDIX.................................................. 100
LIST OF TABLES
Table Page
1. Volumes to Be Infused and Withdrawn and
Corresponding Durations ..................... 24
2. Daily S^/S^ Ratios for Each S, Grouped
According to Stimulus Volume . .......... 28
3. SD/SA Ratios f or the Days Illustrated in
Figures 5-8, Early and Late in
Discrimination Training ..................... 33
4. Means per Volume for SD/S^ Ratio, Bladder
Pressure, Response Overshoot into S^, and
Time Latency into S® - Donna ....... 39
5. Means per Volume for S^/S^ Ratio, Bladder
Pressure, Response Overshoot into SA , and
Time Latency into S^ - Henrietta......... 40
6. Means per Volume for S^/S^ Ratio, Bladder
Pressure, Response Overshoot into S^, and
Time Latency into S^ - G i g i ............ 41
7. Means per Volume for S^/S^ Ratio, Bladder
Pressure, Response Overshoot into S , and
Time Latency into SD - Millie............... 42
8. Summary Table for Kruskal-Wallis One-Way
Analysis of Variance by Ranks, Showing df
and for S^/S^Ratio, Bladder Pressure,
O.S. and LAT., for Each S . . . . . .. . • 53
9. rs Summary Table for Change in Pressure
versus S^/S^ R a t i o ....................... . 56
10, rs Summary Table for Time in S^ versus Pres
sure, Rate versus Pressure, and Time
versus Rate, for Each Volume - Donna . . . 57
11, rs Summary: Table for Time in S^ versus Pres
sure, Rate versus Pressure, and Time
versus Rate, for Each Volume - Henrietta . 58
vi
Table
Page
12. rs Summary Table for Time in versus Pres
sure, S^ Rate versus Pressure, and Time
versus Rate, for Each Volume - Gigi .... 59
13. rs Summary Table for Time in versus Pres
sure, Rate versus Pressure, and Time
versus Rate, for Each Volume - Millie . . . 60
14. Total Means and N's for Length of O.S.,
Grouped by Previous Number of S^'s for
Each S . . . . ........................... 62
15. Summary Table for Kruskal-Wallis One-Way
Analysis of Variance by Ranks, Showing *Xr
and df for Each Volume, for Length of O.S.
Grouped by Previous Number of S*'s .... 65
16. Means and Standard Deviations of Threshold
Bladder Pressure (mm. Hg) Values at
FR Bursts for Each Volume of Fluid .... 67
vii
LIST OF FIGURES
Figure Page
1. Diagram showing innervation of urinary
bladder.................. 10
2. Photograph of monkey seated in Foringer
Primate Chair • • . . • ................ . . 14
3. Schematic drawing of control room and
experimental chamber ....................... 16
4. Schematic drawing of equipment associated
with bladder stimulation .................. 17
5. Responses per minute during successive S^
and S" segments early (A) and late (B) in
discrimination training - Donna ...... 29
6. Responses per minute during successive
and S^ segments early (A) and late (B) in
discrimination training - Henrietta .... 30
7. Responses per minute during successive SD
and segments early (A) and late (B) in
discrimination training - Gigi .......... 31
8. Responses per minute during successive
and segments early (A) and late (B) in
discrimination training - Millie ..... 32
9. Respresentative cumulative record and
corresponding bladder pressure record -
Donna • • * . . . ............................ 35
10. Representative cumulative record and
corresponding bladder pressure record -
Henrietta..................................... 36
11. Representative cumulative record and
corresponding bladder pressure record -
Gigi ........................... 37
12. Representative cumulative record and
corresponding bladder pressure record -
M i l l i e ........................................ 38
viii
Figure Page
13. Data plotted from Table 4 - Donna...... 45
14. Data plotted from Table 5 - Henrietta .... 46
15. Data plotted from Table 6 - Gigi *..... 47
16. Data plotted from Table 7 - M i l l i e.... 48
17. Total S^/S^ ratios plotted for each
volume of fluid............ ................ 49
18. Mean changes in bladder pressure from to
plotted for each volume of fluid .... 50
19. Mean number of O.S. responses into plotted
for each volume of f l u i d .................. 51
20. Mean time in minutes to first response in
SD plotted for each volume of fluid .... 52
21. Means of totals for length of O.S., grouped
by previous number of SR ' s ........... 63
22. Photograph of the lower portion of Henrietta's
spinal cord ........... 68
23. Photograph of the lower portion of Gigi's
spinal cord . .. . . . . . 69
24. Photograph of the lower portion of Millie's
spinal cord ....... 70
ix
ABSTRACT
Interoceptive conditioning has been studied in the
Soviet Union and other Eastern European countries with the
use of Pavlovian techniques for forty years, but became of
interest in the United States only in the last decade,
with the application of operant techniques to intero
ceptive stimuli.
Studies in operant discrimination demonstrated that
the presence or absence of stimuli presented to the small
or large intestine were perceived by monkeys, whose
behavior was differentially reinforced.
The present experiment was designed to determine
whether the interoceptive stimulus, consisting of fluid
presented into the urinary bladder, could be manipulated
experimentally so as to function as a discriminative
stimulus in a normal subject, and to determine if the
interruption of the sensory nerves from the bladder to the
spinal cord would be sufficient to prevent the stimulus
from taking on discriminative functions. The independent
variables included stimulus volume and the degree of
intactness of the sensory nerves from the bladder; and the
dependent variables were lever pressing behavior and
bladder pressure.
Four experimentally naive female rhesus monkeys were
restrained in primate chairs, maintained on 22-hour
deprivation-feeding schedules, and trained to lever press
for sugar pellet reinforcement to a steady rate on an
FR 24 schedule. The dorsal roots of the spinal cord were
severed bilaterally in two £>s at S2-S4 (one of these
surgeries was incomplete) and in one S. at All Ss
were catheterized and fluid was presented into the bladder
and withdrawn according to a programmed design. Lever
presses emitted in the absence of the stimulus (SD) were
reinforced on FR 24, and presses emitted in the presence
of the stimulus (S^) were extinguished.
The normal S (non-operated), as well as the S with
surgery at Ti2“42» successfully discriminated between the
presence and absence of the stimulus beyond the criterion
of the S^/S^response rate ratio of 10tl. The other two
Ss did not. There x*as some evidence of discrimination in
the S, with the incomplete surgery at S2-S4, as reflected
by a decrease in the number of overshoot responses
(non-reinforced responses made after the switch to S^) to
less than 24 responses per S^, and a decrease in latency
until the first press after the switch to SD. The £ 3 with
the complete surgery at S2-S4 showed no evidence of
discrimination learning. All results were evaluated by
two non-parametric statistical testsi the Kruskal-Wallis
one-way analysis of variance by ranks and the Spearman
rank order correlation.
The overshoot and latency results were in agreement
with those of previous interoceptive studies using operant
techniques, in which stimulus onset was quickly discrimin
ated, but stimulus offset was not, suggesting that the
effect was due to characteristics of visceral stimulation.
That the adequate stimulus was pressure was
evidenced by the fact that the reflexive relaxation of the
full bladder decreased the pressure and produced responding
by the monkey, indicating that (absence of fluid in the
bladder) had been initiated. Thus, a more adequate
control of bladder discrimination would be effected by the
manipulation of pressure and not volume.
Implications of the study for psychological theory
and an understanding of psychophysiological mechanisms
underlying psychosomatic disorders were considered.
CHAPTER I
INTRODUCTION
Behavior, the subject matter of psychology, is a
function of stimuli: internal and external, in which
internal is arbitrarily defined as within the skin of the
organism, and external, outside of the skin. The commun
ity at large arranges reinforcement contingencies for
responses to external stimuli, but cannot consistently
reinforce responses to internal stimuli (Skinner, 1963).
For this reason, responses to external stimuli have been
reinforced more than responses to internal stimuli and,
therefore, are repeated. Also, because of the emphasis on
external stimuli, the internal have been largely neglected
in the study of behavior. Skinner (1963) stated that an
"adequate science of behavior must consider events taking
place within the skin of the organism . . . as part of
behavior itself" (p. 953). However, the individual will
be unable to perceive internal stimuli until reinforcement
contingencies for responses to these stimuli are estab
lished. The present behavioral approach to the internal
environment had its beginnings, not in psychology, but in
Russian physiology at the end of the last century, and was
not investigated in the United States until the last
decade (Razran, 1961), It is, therefore, important to
historically review the development of the approach.
Historical Development of the Study of Interoception
In 1866, Sechenov surmised the existence of "faint
sensations originating from the visceral organs" (Adam,
1967, p, 4), Sherrington, in 1911, (Ad/m, 1967) classi
fied three types of receptors: (a) exteroceptors, which
give the organism input about his environments (b) pro
prioceptors, which give information about the muscles and
body position; and (c) interoceptors, which initiate
impulses from the visceral field. According to Ad/m (1967)
interoceptors can be classified by function as mechano-
receptors, chemoreceptors, thermoreceptors, osmoreceptors,
and volume receptors (accumulation of water in the
tissues).
Interoceptive conditioning was first studied in the
Soviet Union with the use of Pavlovian techniques. In
respondent (Pavlovian or classical) conditioning, an
unconditioned stimulus (UGS) if first demonstrated to
normally elicit an unconditioned response (UCR), A
neutral stimulus is introduced, and after some number of
paired presentations with the UCS, the neutral stimulus
(then called the conditioned stimulus or CS) comes to
elicit the response (then called the conditioned response
or CS), For example, in Pavlov's research (1927), a
3
tone (CS) was paired with food (UCS), and the response
(UCR) was salivation. After several pairings, the subject
salivated at the sound of the tone. In reviewing the
Russian work on interoception, Razran (1961) stated,
" fi3nteroceptive conditioning may best be defined as
classical conditioning in which either the conditioned
stimulus (CS) or the unconditioned stimulus (US) or both
are delivered directly to the mucosa of some specific
viscus" (pp. 81-82). By Razran's definition, intero
ceptive conditioning is restricted to respondent. When
either both stimuli are interoceptive (interointero-
ceptive) or the CS is interoceptive and the UCS is
exteroceptive (interoexteroceptive), the cases of intero
ceptive signalization are more valid than the case when
the CS is exteroceptive and the UCS interoceptive (extero-
interoceptive). Interoexteroceptive conditioning is most
often studied.
The first interoceptive conditioning experiment was
reported in 1928 by Bykov and Ivanova (Bykov, 1959),
Saline was introduced into the stomach of a dog through a
fistula, causing diuresis (increased urination). The
presence of the saline in the tissues was the UCS and the
j wetting of the mucosa, the neutral stimulus. After 20-25
paired presentations, diuresis continued to occur when
saline was introduced even though the saline was immed-
| iately withdrawn. However, the results were confounded
since some absorption occurs with the wetting of the
mucosa. The Bykov and Ivanova study was an example of
interointeroceptive conditioning in Razran's framework
(1961), in which both stimuli were interoceptive. Since
this initial effort, numerous experimenters have utilized
a variety of interoceptive stimuli (in interoexteroceptive
conditioning) in many of the visceral organs (Bykov, 1959;
Razran, 1961), Some of the stimuli included the insertion
of a balloon into the organ and inflating it with air or
water, thermal stimuli, irrigation with chemical stimuli,
electrical and tactile stimuli. Visceral organs stimu
lated included the stomach, uterus, small and large
intestines, kidneys, blood vessels and urinary bladder,
Bykov (1959) reported work with dogs in which the
introduction of water into the stomach was paired with a
shock to the hind leg. Eventually, the wetting of the
gastric mucosa was sufficient to cause the S to lift his
hind leg, Bykov conditioned a salivary response to the
presentation of water into the stomach, Airapetyants,
working primarily in the 1940's (Bykov, 1959) showed that
a dog could discriminate water temperature from 26° C, to
36° C , in the stomach, when differentially reinforced by
food, A CR of salivation was elicited by stimulation of
the intestinal walls (CS) through a Thiry-Vella loop
preparation, in which an intact segment of intestine is
externalized, Airapetyants conditioned paw withdrawal to
the introduction of 6° C. water into a balloon in the
loops and Vasilevskaya (Bykov, 1959) conditioned saliva
tion to the presentation of 0,2% HC1 into the loop, Ss
were able to discriminate acid from water in leg with
drawal, Cocaine applied to the mucosa inhibited the
discrimination. Thermal stimuli were also discriminated,
Adam (1967) conditioned respondent salivation to the
rhythmic distension of the kidneys by introducing quan
tities of fluid, Ss discriminated one kidney from the
other and the position in the kidney or ureter where the
pressure was applied, Adam also conditioned EEG desynch
ronization to kidney distension. After habituation of the
EEG to the distension, the distension was paired with an
auditory or tactile stimulus for desynchronization of the
EEG, After this pairing, the kidney stimulation alone
again elicited the EEG desynchronization. However, it is
possible that the last presentation of the kidney
distension alone (after the pairing of the two stimuli)
could have acted as a novel stimulus and elicited the
desynchronization as a UCS, rather than a CS.
The urinary bladder has been stimulated as the
UCS part of exterointeroceptive conditioning, in which
interoceptively aroused reflexes were conditioned to
exteroceptive stimuli (Razran, 1961), which, according
to Razran, is not "true" interoceptive conditioning.
Human Ss with urinary bladder fistulas, after the pairing
6
of bladder distensions with physiological readings on
dials, conspicuous to the Ss, reported urination urges
when the dial readings were high, misrepresenting the
physiological condition.
Operant Methodology and Discrimination
In contrast to the elicited responses of respondent
conditioning paradigms, in the above discussion of
interoception, the emitted responses of operant behavior
are controlled by their environmental consequences
(Skinner, 1938), For example, the lever pressing behavior
of a food deprived rat is conditioned by making food
contingent upon it.
The stimulus control of behavior can be attained by
appropriate differential reinforcement contingencies:
emitted responses are reinforced only in the presence of
a given stimulus and extinguished in its absence (Frick,
1948; Sidman, 1960; Skinner, 1938). Subsequently, the
discriminative stimulus (S^), by virtue of its association
with reinforcement, sets the occasion for responding;
whereas, its absence (S^) reduces responses. Discrimina
tion is, therefore, any consistent difference in response
occasioned by a measurable change in the stimulus situa
tion (Frick, 1948),
Interoception and Operant Conditioning
Razran1s (1961) definition of interoceptive con
ditioning, given above, must be expanded to include the
operant, as well as the respondent. Thus, interoceptive
conditioning may best be defined as conditioning in which
the S*3 source, the reinforcement (S^) source, and/or the
response source are interoceptive. There is some
question as to whether some of the Soviet (Bykov, 1959)
work were respondent, or really operant. In the Bykov
leg withdrawal study reported above, the conditioning
was operant if, with the wetting of the gastric mucosa,
lifting of the leg avoided shock. The study was respon
dent if the delivery of shock was not contingent upon
leg lifting.
The first study to investigate interoception in the
United States utilized operant techniques. Cook,
Davidson, Davis and Kelleher (1960) conditioned a leg-
flexion avoidance response to shock in dogs, using, as
interoceptive stimuli, pressure in the jejunal portion of
the intestines and drugs (acetylcholine, norepinephrine,
and epinephrine) injected into the external saphenous vein.
Schuster and Brady (1964) conditioned a food reinforced
operant using, as stimuli for an interoceptive S*3, the
infusion of (a) epinephrine and (b) saline-destrose mix
ture directly into the superior vena cava of rhesus
monkeys•
Slucki, Adam and Porter (1965) surgically modified
the small intestines of rhesus monkeys to form a Thiry
fistula, in which one end is sealed off and the other used
for the insertion of a balloon to be rhythmically inflated
and deflated as the interoceptive stimulus. The response
rate of lever pressing for sugar pellets was stabilized at
FR 25 (a fixed ratio schedule of 25 responses per 1 rein
forcement). The first few presentations of the stimulus
caused EEG desynchronization, but no significant behavioral
alteration, suggesting that the stimulus was, therefore,
non-aversive. Ss were then conditioned to press during
the presence of the stimulus (S^), and were extinguished,
as no reinforcements would be available, during the
absence of the stimulus (S^). Ss were able to discrim
inate the onset of the stimulus well, but not the offset.
McCoy (196Ss Slucki, McCoy & Porter, 1969) prepared
a Thiry-Vella loop in the large intestine of a rhesus
monkey, which was trained to press for sugar pellets on
FR 24. Using the same techniques for administering the
interoceptive stimulus as in the Slucki et al, study
(1965), the experimenter confirmed the previous results
for the small intestine.
Urinary Bladder
No studies have been reported in which the intero
ceptive stimulation of the urinary bladder was used in
which bladder stimuli were experimentally manipulated to
determine the effects on behavior. The present research
was designed for this purpose.
Neuroanatomv. Figure 1 shows the probable innerva
tion of the bladder, since discrepancies exist as to the
pathways taken by visceral afferent nerve fibers from the
bladder to the spinal cord. Most neuroanatomists agree
that the fibers travel with the autonomic nervous system
(ANS) fibers which innervate the bladder. Some believe
the fibers travel with both the parasympathetic and the
sympathetic fibers, although mostly with the parasympa
thetic (Adam, 1967; Ruch, 1960; Truex & Carpenter, 1964),
Recent research by Edvardsen (1968), utilizing cats as
subjects, has strongly suggested that the visceral
afferent nerves from the bladder travel with the para
sympathetic fibers exclusively. Afferent nerves travel
ing with the parasympathetic fibers would be found in the j
pelvic nerve and would enter the cord at sacral levels
S2-S4. Afferent fibers traveling with the sympathetic . |
j
would be found in the hypogastric nerve and would enter the|
cord at thoracicolumbar level: Edvardsen (1968) j
i
documented the existence of a negative feedback loop
between the sensory fibers from the bladder, carrying
10
SPINAL
CORD
SYMPATHETIC CHAIN
INFERIOR MESENTERIC
X GANGLION
HYPOGASTRIC NERVE
X (SYMPATHETIC)
-VESICAL PLEXUS
URINARY
BLADDER
TRIGONE
PELVIC NERVE
(PARASYMPATHETIC)
INTERNAL
^ SPHINCTER
Fig. 1. Diagram showing the innervation of the
urinary bladder. Solid lines represent efferent pathways;
broken lines, afferent pathways, which are shown to
combine with the parasympathetic efferent fibers for
travel to the spinal cord, (Adapted from Edvardsen, 1968;
Truex & Carpenter, 1964.)
increased pressure information, and the sympathetic fibers,
which produce relaxation of the bladder. The severing of
the dorsal (sensory) roots bilaterally at the sacral level,
interrupted the feedback loop, so the bladder did not
relax? thus, bladder pressure was higher post-operatively
than pre-operatively, using a constant volume. However,
with a similar pre- and post-operative comparison, severing
the dorsal roots at level T12-L2 produced no effect on the
post-operative pressure and, thus, did not interrupt the
feedback loop. Edvardsen (1968) concluded that sensory
information from the bladder is not carried in the
hypogastric nerve.
Statement of the Problem
The first objective of the present experiment was to
determine whether the interoceptive stimulus of fluid
presented into the urinary bladder could be experimentally
manipulated so as to function as a discriminative stimulus;
that is, reinforcement would be available without fluid
present, and would not be available with fluid present.
The second objective was to determine whether the inter
ruption of the sensory nerves from the bladder would |
prevent the stimulus from taking on discriminative
functions. On the basis of Edvardsen's work (1968) it was j
predicted that severing the dorsal roots of S2-S4 would
eliminate the S's ability to learn the discrimination; j
i
whereas, severing the dorsal roots of T12* * 1*2 would have no j
effect on the abilities of the S to discriminate, as
compared to a non-operated control. The independent
variables of the study were the fluid volume introduced
into the bladder and the degree of intactness of the
sensory pathways from the bladder. The dependent
variables of the study were lever pressing behavior and
bladder pressure.
CHAPTER II
METHOD
Subjects
The subjects of the experiment were four female,
experimentally naive, adult rhesus monkeys about 4 years of
agei Donna, Henrietta, Gigi and Millie. Their weights
varied from 11-13 lbs. at the beginning of the experiment.
Each S was seated (Figure 2) in a Foringer Primate Chair
(Model 1101) and maintained on 22-hour food deprivation,
receiving a restricted diet of monkey cookies (Old Mother
Hubbard Chimp Crackers) and fruit a half hour after the
experimental session daily. Due to the food deprivation
schedule, the weights of Henrietta, Gigi and Millie dropped
to 75-77% of the original values, as measured post-
surgically. Each S had ad lib access to water, supple
mented with multi-vitamin drops (Vi-Syneral, USV Pharma
ceutical Corp.), except during the experimental sessions
when the bottle was removed. After discrimination
training began, access to water was ad lib up to 500 ml.
per day.
13
14
Fig. 2. Photograph of a monkey seated in a Foringer
Primate Chair, showing the location of the lever, the
pellet dispenser and the hopper, for an experimental
session.
Apparatus
Experimental chamber and control room. A schematic
drawing of the experimental chamber and the adjacent
control room is given in Figure 3. The experimental
chamber was sound-attenuated, we11-ventilated and moder
ately illuminated, with a one-way glass window for behav
ioral observation. The control room contained the behav
ioral apparatus and the visceral stimulation apparatus.
Behavioral apparatus. A lever (operandum) was
placed 20 cm. from the S at waist level, at the beginning
of the experimental session (Figure 2). Precision Food
Pellets (P0 J« Noyes Co.: dextrose 44,6 mg., starch 2.2
mg.) were delivered by a dispenser (Davis Scientific
Instruments, Model PD-105) into a hopper near the monkey's
chin. Responses were recorded on a cumulative recorder
(Davis Scientific Instruments, Model CRRC-133) and
reinforcement delivered through BRS-Foringer programming
equipment.
Visceral stimulation apparatus. A schematic drawing
of the bladder stimulation apparatus is presented in
Figure 4. Such equipment included an infusion and with
drawal pump and 50 cc. glass luer-lock syringe (B-D) for
the delivery and extraction of fluid to and from the
bladder, a condenser and water bath, maintained at
approximately 37° C. (98.6° F.), to heat the fluid to
body temperature, and a transducer (Statham, Model P23Dc)
Fig.
chamber.
■ CONTROL ROOM EXPERIMENTAL CHAMBER
WALL WITH ONE
S WAY MIRROR
BRS-FORINGER
CONTROL GEAR PRIMATE CHAIR-
CUMULATIVE
RECORDER"!
DYNOGRAPH PUMP
. Schematic drawing of the control room and the experimental
DYNOGRAPH
<
>
^ TO BRS-FORINGER CONTROL GEAR
PE 100 TUBING*
TRANSDUCER**
TOGGLE SWITCH
SYRINGE
PUMP
STOP COCKS
CABLE FROM TRANSDUCER TO DYNOGRAPH
WALL AND
INSULATED SLEEVE
CONDENSER
CATHETER
Fig. 4. Schematic drawing of the equipment associated with bladder
stimulation.
18
and recording apparatus with ink printout (Beckman-Offner
Type RS Dynograph) to measure bladder pressure. The
dynograph was calibrated with a column of mercury, since
the apparatus was not equipped with an internal calibrat
ing mechanism. The control panel (Type A 560) paper speed
was held constant at 1 mm./sec. The pre-amplifier (Type
461B), controlling the amount of deflection in the pen
writer, was set at 0.5 mV./cm.; and the amplifier (Type
462) setting was "XI pre-amp." Polyethylene tubing
(PE 100) connected the syringe with the transducer, and it,
in turn, with the condenser. Two 3-way stopcocks (B-D
Model MS10) functioned as flush points for the system.
The syringe and tubing system was flushed every other day
with fresh solution. The tubing, between the transducer
and the condenser, traveled through an insulated sleeve in
the wall from the control room to the experimental chamber.
The condenser was anchored to the wall in the experimental
chamber and covered with a towel to prevent the stimulus
from becoming visual and not interoceptive. The movement
of occasional bubbles in the tube, if visible, could have
become associated with the discrimination paradigm. The
volume of tubing plus the catheter, between the condenser
and the monkey, was approximately 7.5 cc. (which was not
being heated continually). The condenser held approxim
ately 25 cc. of fluid at about 37° C.
19
A toggle switch on the pump apparatus , when actuated
by the experimenter, initiated the pump and cued the
programmed BRS-Foringer equipment for the discrimination
paradigm.
Prior to catheterization, E wrapped a terry cloth
towel around the waist of the monkey and clamped it in
place, preventing the S from reaching the catheter with her
hands. The fluid used for the bladder stimulus was the
bottled sterile solution of sodium chloride injection
(Travenol normal saline). Double-tubed Eoley 8 catheters
(G. R. Bard, Cat. No. 165-P) were used. Air was injected
into one tube to inflate a small 3 cc. balloon in order to
anchor the catheter in place in the trigone of the bladder.
Two holes at the bladder end of the catheter minimized
squirting when the saline was presented into the bladder.
The Foley 8 proved to be too small for Donna, so a Foley 14
with a 5 cc. balloon was substituted, after one month of
discrimination training.
Procedure
The procedure consisted of three basic stepsj
initial training, surgery, and catheterization and
discrimination training.
Initial training. Ss were weighed, placed in chairs,
and allowed to adapt to the chairs for five days, during
which time they were introduced to the sugar pellets.
During the last two of these days the Ss were deprived of
20
food. Ss were trained to lick the pellets from the hopper
when the dispenser was activated. During this tray
approach training and all subsequent experimental sessions,
the hand holes in the neck plate of the chairs were
covered so the Ss could not tamper with the reinforcement
apparatus. Tray approach training took from 1-2 days.
Ss were placed individually in the experimental
chamber and shaped to respond to the lever, on a continuous
reinforcement (CRF) schedule. By an adjusting ratio, the
schedule was gradually incremented to FR 24, at which the
behavior was stabilized. This process took about 1 month
with daily experimental sessions.
Surgery. Donna received no surgery and served as
control. The other Ss were divided into two surgical
groups* S2-S4 and Ti2“J -2* F°r Gigi* and Millie, the dorsal
roots in the sacral area, including S2-S4, were severed
bilaterally at the point of entry to the cord. For
Henrietta, the dorsal roots from the lower thoracic area
to the upper lumbar, including Ti2”l*2» were severed
bilaterally.
A neurosurgeon, Martin Krell, M. D., performed the
surgeries, assisted by the experimenter. The surgical
procedure was as follows. Each S was removed from her
chair and restrained on her back on a monkey board.
Anesthetic (1.50 cc. Diabutal, Diamond Laboratories,
Sodium pentabarbital) was injected into the femoral vein.
21
An additional 1,00 cc. was required by Henrietta and Millie
to reach surgical level. When the S no longer responded
with leg withdrawal to pinching the foot, she was turned
over, and her arms and legs restrained. Surgical
preparation included shaving the fur from an approximately
10 cm. by 15 cm. area on the back and dousing with
merthiolate. All necessary instruments, gowns and towels
had been autoclaved, so that a standard sterile procedure
was followed for the surgery. Incisions were made along
the spine from about the bottom rib to the iliac crest.
Laminectomies (cutting away the bony matter) were performed
to expose the lower spinal cord and cauda equina. The
appropriate dorsal roots were located by counting up from
the cauda equina and severed. The dissecting microscope
was employed for Henrietta and Millie, but not for Gigi,
who underwent surgery first. Absorbable sutures (Chromic
C, G-123, Ethicon) were used for the muscles, and wound
clips for the skin. The surgical procedure took approx
imately 3 hours for each monkey. Surgery was performed on
a separate day for each monkey.
Immediately upon completion of surgery, each £ 3 was
injected with 2 cc, penicillin, placed in a monkey cage,
covered for warmth, and supplied with ad lib access to
water, but no food. Food was made available the following ,
day. For each of the 4 days immediately following surgery,;
22
each monkey ingested a capsule of terramycin (which E
opened and sprinkled into the food). Ss were weighed and
returned to their chairs 4-6 days after surgery.
Ss were tested for the Bulbocavarnosis reflex, which
if present indicated that the sacral roots were intact.
The process was as follows: the S, was catheterized and a
sharp tug was given to the catheter. If the sacral roots
were intact, the anal sphincter contracted. If they were
not intact, the sphincter was flacid and would not con
tract* After the reflex test, the S 3 was returned to a
regular 22-hour deprivation feeding schedule.
Catheterization and discrimination training. Each S.
was catheterized daily at the beginning of each experimen
tal session for the duration of the experiment. The used
catheters were individually labeled for each monkey and
stored in a sterilizing solution of zephrin chloride.
Prior to the beginning of discrimination training, but
during an experimental session, 10 cc. of fluid were
pumped into the bladder of each S, and behavioral observa
tions were made, in order to determine that the stimulus
was not aversive. Discrimination training began in the
very next session. Lever presses emitted during the
presence of the stimulus were extinguished (S^), while
those occurring in the absence of the stimulus (S^) were
reinforced on FR 24. The purpose for using S^3 and in
this fashion is explained in the Appendix.
23
The operational definition of the onset of was the
actuation of the toggle switch to the syringe to begin
feeding fluid into the bladder, and of S^, the actuation
of the toggle switch for the syringe to begin withdrawing
fluid from the bladder.
The rate of flow of the infusion withdrawal pump was
150 cc./ min. Table 1 lists a range of volumes and the
corresponding times necessary for delivery or withdrawal.
These durations, as indicated above, are definitionally
part of the period for delivery and the 5^ period for
withdrawal. In order for the time spent in to be at
least twice that spent in S*3, following convention, was
varied, according to a variable program, from approximately
2 minutes to 30 minutes, and SD from approximately 30
seconds to 6 minutes. Daily sessions were varied from
about 30 minutes to 2 hours per S. Discrimination
training continued until the criterion of 10:1, S^/S^
response rate ratio was exceeded, or until several sessions
(2-4) had passed and the S, had not learned the discrim
ination.
Since there were no data for monkeys as to the
sensation threshold of pressure in the bladder, the
threshold volume was to be established experimentally.
Micturition contractions begin to occur when the volume
reaches about 50 cc., although an anesthetized monkey j
bladder can hold 100 cc. (Porter, 1969). Edvardsen (1968) j
TABLE 1
VOLUMES TO BE INFUSED AND WITHDRAWN
AND CORRESPONDING DURATIONS
VOLUMES DURATION
10 cc • 4 secs.
20 cc ■ 8 secs.
30 cc ■ 12 secs.
40 cc < 16 secs,
50 cc. 20 secs •
60 cc. 24 secs,
70 cc. 28 secs.
80 cc • 32 secs.
25
suggested using no more than 2/3 of the micturition
threshold value, or 30-40 cc. in this case. The starting
infusion value was 10 cc., as suggested by Porter (1969),
which was incremented by 10 cc. in each instance that the
gD/sA ratio had not improved after several sessions (2-4)
at that volume.
Post-mortem examination. At the completion of the
experiment, the three Ss, having undergone surgery, were
sacrificed and their spinal cords excised and photographed
to determine the specific sites of the surgery. Judgment
was made by the neurosurgeon, Martin Krell, M. D.
CHAPTER III
RESULTS
Summary of experimental procedure
Subjects were trained to press a lever for sugar
pellet reinforcement (S^) at a fixed ratio of 24 presses
per S* (FR 24). The dorsal roots of the spinal cord were
severed bilaterally as follows: Gigi and Millie - S2-S4 ,
Henrietta - T12-^2 * and Donna - no surgery. The intero
ceptive stimulus of fluid pumped into the bladder was
introduced. Reinforcement was contingent upon behavior in
the absence of the stimulus (S*3), and was not available in
its presence (S^)» The volume of fluid presented, which
varied for each S, was increased sequentially until either
discrimination was established in excess of the set
criterion for the S^/S^ response rate ratio of 10:1, or
it was determined that discrimination could not be
established.
Discrimination learning
For Donna, the volume of fluid was increased from 20
to 30 to 40 cc., where the S^/S^ ratio of 10:1 was
exceeded. Subsequently, the volume was reduced to 10 cc.
to determine whether Donna could make the discrimination
at the lower volume after being successful at the larger
26........
27
one0 For Henrietta, the volume was increased from 10 to
40 cc,, in incremental steps of 10 cc., where the S^/S^
ratio of 10il was exceeded. For Gigi, the volume was
increased from 10 to 20 to 30 to 40 to 60 to 70 to 80; and
for Millie from 10 to 20 to 30 to 40 to 70 to 80; when it
was determined for both of the Ss that neither could
perform to the established discrimination criterion.
The 50 cc. value for Gigi and the 50 and 60 cc. values for
Millie were skipped to accelerate the experimental process.
To demonstrate the length of time required to exceed
criterion, Table 2 gives the daily SD/S^ ratios for each S,
showing the number of days at each volume. At no time did
Gigi or Millie reach criterion. Donna exceeded criterion
on 3 of the 5 days at 40 cc.; and Henrietta exceeded
criterion on 2 of the 6 days at 30 cc. and on 3 of the 4
days at 40 cc.
Figures 5 through 8 show, for each S, the number of
responses per minute in discrimination training during
successive and segments of the first session with
10 cc. fluid stimulus and a late session with 40 cc. The
mean S^/S^ ratios for these days are presented in Table 3.
The ratios of all four S. early in training were close to
1*1. Late in training, at 40 cc *, Donna and Henrietta
exceeded criterion (16.40:1 and 19.04:1, respectively),
Gigi and Millie did not reach criterion (3.21:1 and 4,44:1,
respectively).
28
TABLE 2
DAILY SD/S* RATIOS FOR EACH S, GROUPED
ACCORDING TO STIMULUS VOLUME
DONNA HENRIETTA GIGI MILLIE
DAY VOL. SD/SA VOL. sD/s^- VOL. SD/SA VOL. SD/SA
1 20 cc• 2.62T110 cc. 3.92 10 cc. 2.29 10 cc. 2.49
2 3.16 4.18 3,60 2.57
3 2.21 3.09 3,82 2.99
4 3.12 3,08 2.31 2.51
5 30 cc. 3.38 3.47 2.80 3.13
6 2.55 2.47 3.17 4.10
7 5.40 5,08 2.81 4.13
8 3.71 4.71 2.38 3.22
9 3.78 3.31 3.78 3.47
10 3.17 3.96 3,16 2.83
11 2.91 4.16 1.85 3.03
12 4.20 4.30 2.29 3.54
13 5.43 1.98 2.51 4.81
14 4.93 2.71 3.25 2.78
15 2.80 3.62 20 cc. 2.50 3.52
16 5.18 20 cc. 3.58 3.40 20 cc. 2.81
17 4.07 2.68 1.79 2.35
18 3.98 3.63 2.29 2.98
19 6.20 3.86 2.75 3.02
20 5.99 3.21 2.33 4.97
21 3.22 30 cc. 5.05 30 cc ■ 3.31 30 cc. 3.74
22 6.85 6.55 2,41 6.12
23 6.40 15.06* 2.92 4.21
24 5.00 6.12 2.36 3.94
25 6.36 25.64* 3.18 4.25
26 40 cc. 14,18^ 4.73 40 cc. 2.24 40 cc. 3.97
27 6.84 46 cc. 14.21^< 2.50 6.26
28 7.36 7.04 2.58 8.50
29 16,40* 19.04* 3.75 4.75
30 10.24* 17.92* 3.21 4.44
31 10 cc. 3.63 3.04 4.25
32 4.11 60 cc. 3.04 70 cc. 4.45
33 4.70 7.27 6.65
34 5.15 2.92 4.00
35 7.30 3.58 6.02
36 4.99 70 cc. 5.33 80 cc. 4.08
37 4.66 5.91 4.44
38 80 cc. 3.05
39 8.25
40 4.42
*sD/sA ratio exceeded criterion
J
29
w
H
W
n,
cn
w
C O
s
o
Hi
C O
w
a:
DONNA
B
RESPONSES
RESPONSES
Fig. 5. Number of responses per minute during
successive S^ and S^ segments of the first session (A)
early in discrimination training with 10 cc. of fluid
stimulus, and a session (B) late in discrimination training
with 40 cc. of fluid.
6
30
w
H
w
to
w
to
s
o
CM
to
1
HENRIETTA
B
SD RESPONSES
S*4 RESPONSES
Fig, 6, Number of responses per minute during
successive S^ and S^ segments of the first session (A)
early in discrimination training with 10 cc, of fluid
stimulus, and a session (B) late in discrimination training
with 40 cc. of fluid.
31
w
H
t —i
s
w
Ph
C O
w
C O
a
o
C O
GIGI
B
E531
SD RESPONSES
S^ RESPONSES
Fig. 7. Number of responses per minute during
successive S^ and S^ segments of the first session (A)
early in discrimination training with 10 cc. of fluid
stimulus, and a session (B) late in discrimination training
with 40 cc. of fluid.
4
32
w
H
g
1 —4
s
Pi
w
f i »
m
w
co
S
o
Ph
C O
w
Pi
250
230
210
190
170
150
130
110
90
70
50
30
10
I
I I I
MILLIE
B
■ I SD RESPONSES
E52I SA RESPONSES
Fig. 8. Number of responses per minute during
successive S* 5 and segments of the first session (A)
early in discrimination training with 10 cc. of fluid
stimulus, and a session (B) late in discrimination training
with 40 cc. of fluid.
TABLE 3
SD/S* RATIOS FOR THE DAYS ILLUSTRATED
IN FIGURES 5-8, EARLY AND LATE
IN DISCRIMINATION TRAINING
s EARLY LATE
DONNA 1.10 16.40
HENRIETTA 1.28 19.04
GIGI 0.99 3.21
MILLIE 0.96 4.44
34
Figures 9 through 12 show actual representative
cumulative records and corresponding bladder pressure
records for an S^-SA-S^ segment at 40 cc. fluid, for each
S.
Two of the four Ss learned the discrimination,
according to the 10:1 ratio criterion. Donna, the non
operated control, and Henrietta, with sensory roots at
T12-L2 severed, both exceeded criterion; whereas, Gigi
and Millie, with sacral roots severed, did not.
Basic Measures
Overall means for each S, given in Tables 4 through
7 were calculated for each volume of fluid for the
following four measures* S^/SA ratio, bladder pressure,
response overshoot into SA , and time latency into S^.
(a) SD/SA ratio = (RD/TD)/(RA/TA ) ,
where: = total responses in
T^ = total time in
RA s total responses in SA
TA - total time in SA .
Values are entered for each volume of fluid separately.
(b) Bladder pressure, given in mm. of mercury (Hg), j
i
was measured as the change in pressure from S^, which was j
i
taken as a baseline, to SA« The mean pressure of the first
SD of a daily session was subtracted from the mean j
A T\ !
pressure of each S*-* of that session. The first S1' mean
was used, rather than the mean of the session, since j
35
do/vna
CUMULATIVE RECORD
K a OEP (SD)
^ = O N ( S * )
* BLADDER PRESSURE RECORD
Fig. 9. Representative cumulative record and
corresponding bladder pressure record of an S*^-S^-S^
segment at 40 cc.
HENRIETTA
36
1M/AI,
CUMULATIVE RECORD
\ * OFF (S°)
(S&)
BLADDER PRESSURE RECORD
Fig. 10. Representative cumulative record and
corresponding bladder pressure record of an SD-S -S^
segment at 40 cc.
CUMULATIVE RECORD
BLADDER PRESSURE RECORD ' '
Fig, 11, Representative cumulative record and
corresponding bladder pressure record of an SD-S -SD
segment at 40 cc.
I 1QOmm,Us
38
MILLIE
CUM ULATIVE RECORD
BLADDER PRESSURE RECORD
Fig. 12. Representative cumulative record and — —a - — - — c---------—• — — —— — — - ■ — . * ~ . and
corresponding bladder pressure record of an S^-S^-S0
segment at 40 cc.
TABLE 4
MEANS PER VOLUME FOR SD/SA RATIO, BLADDER PRESSURE,
RESPONSE OVERSHOOT INTO S^ AND
TIME LATENCY INTO SD
DONNA
VOL. SD/S* RATIO
PRESSURE
(Mm. Hg)
O.S. (R’s) LAT. (MIN.)
20 cc. 2.68 no record 35.55 0.24
30 cc • 4.01 54.23 20.73 0.26
40 cc. 9.99 57.39 12.41 0.20
10 cc. 4.48 49.58 23.08 0,21
40
TABLE 5
MEANS PER VOLUME FOR SD/SA RATIO, BLADDER PRESSURE
RESPONSE OVERSHOOT INTO S^ AND
TIME LATENCY INTO SD
HENRIETTA
VOL. SD/SA RATIO
PRESSURE
(Mm. Hg)
O.S. (R's) LAT. (MIN.)
10 cc. 3.40 no record 18.72 0.18
20 cc. 3.16 24.95 25.37 0.15
30 cc. 6.53 31.09 14.94 0.44
40 cc. 8.68 36.31 9.02 0.43
41
TABLE 6
MEANS PER VOLUME FOR SD/SA RATIO, BLADDER PRESSURE,
RESPONSE OVERSHOOT INTO S* AND
TIME LATENCY INTO SD
GIGI
VOL.
sD/sA ratio
PRESSURE
(Mm. Hg)
O.S. (R's) LAT. (MIN.)
10 cc. 2.68 no record 30.63 0.56
20 cc. 2.25 24.87 37.90 0.36
30 cc. 2.72 33.55 13.12 0.21
40 cc • 2.71 33.68 7.77 0.16
60 cc. 3.32 41.72 5.80 0.15
70 cc. 5.66 48,15 2.60 0.15
80 cc. 3.50 48.03 9.58 0.13
42
TABLE 7
MEANS PER VOLUME FOR SD/SA RATIO, BLADDER PRESSURE
RESPONSE OVERSHOOT INTO SA AND
TIME LATENCY INTO SD
MILLIE
VOL. SD/S^ RATIO
PRESSURE
(Mm, Hg)
O.S. (R's) LAT. (MIN.)
10 cc. 2.99 no record 41.35 0.57
20 cc. 3.10 12.28 36.88 0.31
30 cc. 3.96 11.20 35.31 0.25
40 cc. 4.84 10.28 36.58 0.39
70 cc. 4.80 19.05 40.28 1.22
80 cc. 4.30 19.63 34.91 1.39
43
bladder contractions occasionally occurred in later S^'s,
spuriously elevating the session mean, and also since the
dynograph had occasional internal drift, causing the
baseline to shift. The mean changes in pressure were
weighted according to the time spent in each particular
S^, in order to obtain the overall mean for each volume.
The weight value in a particular S^was the number of
pressure readings (one every 20 seconds) in that (thus,
the longer the S^, the more readings taken). For example,
if the mean pressure of the first SD were 10 (mm. Hg) and
the means of three succeeding S^'s were 40, 25 and 35, then
the mean changes in pressure would be 30, 15 and 25,
respectively. If the weights for these S^'s were 10, 5
and 5, respectively, then using the formula! overall
mean = ( N, (where X is the S^mean; J-* is the weight
or frequency corresponding to the number of readings in
the S^j and N = the overall mean would be (10*30 +
5*15 + 5*25)/(10 + 5 + 5) or 25 (Hays, 1963, p. 162).
(c) Response overshoot into S^(0.S.) was measured
as the number of non-reinforcement contingent responses
made by the S. prior to the first pause, at least 2 seconds j
long, after the switch was made to S^. The mean for each j
|
volume was the total number of O.S. responses divided by j
|
the number of O.S.1 for each volume, j
(d) Time latency in S^ (IAT.) was the measured time
between the switch from to and the occurrence of the
first lever press in that S*\ The mean for each volume
was the total LAT. time divided by the number of IAT. 's
for each volume.
The data from Tables 4 through 7 are presented
graphically in Figures 13 through 16. For comparison
purposes across Ss, the same data have been arranged
according to the specific measure and are shown graphically
in Figures 17 through 20.
A Kruskal-Wallis one-way analysis of variance by
ranks was performed on each of the four measures presented
above for each S, using the following formula (Siegel,
1956, p. 185).
H = - 3 < m ) -
j=l
where.* k = number of samples (volumes, in this case),;
nj = number of cases in the j' * ' * 1 sample,
. N = £nj,
Rj = sum of the ranks in the sample.
H is distributed as a with k-1 degrees of freedom. The j
results of these analyses are presented in Table 8. Since j
calculations involve rank order rather than raw data, j
significant results merely indicate trends and the
consistency of behavior and do not represent the magnitudes
45
1.4
^1,2
51.0
5 .8
r! • 6
3 .4
.2
DONNA
60
M
I
cn
cn
w
at
pm
56
48
40
32
24
16
8
10
8
6
4
2
10 40 20
VOLUME IN CC. FLUID
Fig, 13. Data plotted from Table 4, showing SD/S^
ratio, bladder pressure, overshoot and latency for each
volume.
46
S3
t - J
a
§
1.4
1.2
1.0
.8
.6
.4
.2
48
40
32
24
16
8
60 56
a
48
I 40
32
: = > 24
W 16
i / )
o
Ph
o
M
H
8
HENRIETTA
10
8
6
4
2
20 40 10 30
VOLUME IN CC. FLUID
Fig. 14. Data plotted from Table 5f showing SD/S^
ratio, bladder pressure, overshoot and latency for each
volume.
RATIO PRESSURE (Mm. Hg) O.S. LAT. (MIN.
47
1.4
~ 1.2
1.0
.8
• 6
.4
.2
48
40
32
24
16
8
56
48
40
32
24
16
8
10
8
6
4
2
GIGI
VOLUME IN CC. FLUID
Fig. 15. Data plotted from Table 6, showing SD/S^
ratio, bladder pressure, overshoot and latency for each
volume.
.4
.2
.0
.8
• 6
.4
.2
48
40
32
24
16
8
56
48
40
32
24
16
8
10
8
6
4
2
MILLIE
10 20 30 40 50 60 70 80
VOLUME IN CC. FLUID
Fig. 16. Data plotted from Table 7, showing S^/S^
, bladder pressure, overshoot and latency for each
SD/SA RATIO
49
DONNA • ------ m
HENRIETTA O ---— o
GIGI
A — — A
MILLIE
4 — —
10
10 20 30 40 50 80
VOLUME IN CC. FLUID
*Measures at 10 cc. taken after 40 cc. for Donna
Fig. 17. Total S°/SA ratios plotted for each
volume of fluid.
56
52
48
44
40
36
32
28
24
20
16
12
8
4
50
*
DONNA •-
HENRIETTA O
GIGI-------A ---- A
MILLIE ▲---- A
10 20 30 40 50 60 70 80
VOLUME IN CC. FLUID
^Measures at 10 cc. taken after 40 cc. for Donna
Fig. 18. Mean changes in bladder pressure from SD
otted for each volume of fluid.
MEAN NUMBER OF O.S. RESPONSES
51
DONNA # -------#
HENRIETTA O
GIGX
44 MILLIE
40
36
32
28
24
* •
20
16
12
8
4
10 20 30 50 40 60 70 80
VOLUME IN CC. FLUID
*Measures at 10 cc. taken after 40 cc. for Donna
Fig. 19, Mean number of O.S. responses into
plotted for each volume of fluid. The horizontal line at
24 responses indicates 1 missed S .
MEAN LAT. TO FIRST RESPONSE I N Su (MIN.
52
DONNA
HENRIETTA
GIGI
MILLIE
1.0
.8
10 20 30 40 50 60 70 80
VOLUME IN CC. FLUID
*Measures at 10 cc. taken after 40 cc. for Donna
Fig. 20. Mean time in minutes to first response
in S*^ plotted for each volume of fluid.
53
TABLE 8
SUMMARY TABLE FOR KRUSKAL-WALLIS ONE-WAY ANALYSIS OF
VARIANCE BY RANKS, SHOWING df AND FOR
SD/S^ RATIO, PRESSURE, O.S. AND
LAT., FOR EACH S
a 2
S df SD/SA O.S. LAT. PRESSURE
DONNA 2 13.82**** 9.24** 1.98 6.27*
HENRIETTA 3 19,06**** 7.78a 10.87**b
(df=2)
4, 90c
GIGI 6 4.68 33.76**** 26.59****
(df=5)
13.10*
MILLIE 5 19.08*** 12.86* 12.33*b
(df=4)
14.01***
****p <.001
***p <.01
**p <.02
*p <.05
crit .05=7.82
bLAT. values for both Henrietta and Millie
were in the opposite direction from that predicted.
cJ^Tcrit .05=5.99
of difference. Donna, Henrietta (p^.001) and Millie
(p <.01) showed a significant trend for the increase in
ratio as the volume of fluid increased, Donna and
Henrietta both exceeded the criterion (ratio = 10:1); ]
I
whereas, Millie’s ratio, although increasing with the
increase in fluid, did not reach criterion. Gigi did not
show a significant trend for the increase in ratio,
although the ratio means' data in Table 6 suggest that
such a trend did exist, even though criterion performance
was not reached. However, the raw data were too variable
for the Kruskal-Wallis trend to appear. All S. displayed
a trend in the reduction of the O.S. with the increase in
volume: Donna, p^,02; Henrietta, just missing signifi
cance at .05 (X2 = 7.78;a2crit ,05 = 7.82); Gigi, p<.001;
and Millie, p ^.05. All Ss showed a trend for the
increase in pressure with the increase in volume: Donna,
p<^.05; Henrietta, just missing significance at .05
(X* = 4.90; X^crit ,05 = 5.99); Gigi, p^.05; and Millie,
p ^.01. Gigi was the only j3 showing a significant
decrease in IAT. with the increase in volume (p^.001).
Henrietta and Millie both showed significance in the j
|
opposite direction (p^.02 and p^.05, respectively).
In order to more fully analyze the relationship
between ratio and pressure, a Spearman rank order
I
correlation (rho or rs) was performed on the daily means j
of the S^/S^ ratio and the change in pressure for each S., j
using the following formula*
rs = H(^-iy
where dj_ was the difference between ranks on means
associated with a particular daily run and N was the
number of runs observed• The N associated with each S. and
the corresponding rs are presented in Table 9, Gigi
showed the only significant relationship between ratio
and pressure (p^.Ol), although both Donna and Henrietta
closely approached significance at the ,05 level (rs Donna
= ,381, rs crit ,05 = *388; rs Henrietta = *4-8, rs crit ,05
= .506).
Rate, Time and Pressure in
Spearman rank order correlations were performed for
each volume of fluid on the following three comparisons:
(a) time in each individual versus the corresponding
mean change in pressure for that (b) lever pressing
rate in each individual versus the corresponding mean
change in pressure; and (c) time in versus rate. The
same calculational formula for rs was used, as above,
where dj. was the difference between ranks on means
associated with a particular and N was the number of
S^'s observed. The resulting data are summarized for each
S, in Tables 10 through 13. Donna showed significance in
all three comparisons, such that an increase in time in
TABLE 9
rs SUMMARY TABLE FOR DAILY PRESSURE
VERSUS SD/SA RATIO FOR EACH S
s N
rs crit
DONNA 19 .381
.388
Cp=.05)
HENRIETTA 12 .480
.506
(p=,05)
GIGI 21 .700**
.521
(p=.01)
MILLIE 18 .080
.399
(pa.05)
W*P<.oi
57
TABLE 10
rs SUMMARY TABLE FOR TIME IN VERSUS PRESSURE, SA RATE
VERSUS PRESSURE, AND TIME VERSUS
RATE, FOR EACH VOLUME
DONNA
rs
VOL. N T VS P RT VS P T VS RT
rs crit
.05
rs crit
.01
10 cc. 28 -.35* -.02 .39* .317 .448
30 cc. 20 -.81** -.68** .76** .377 .534
40 cc. 21 -.68** -.47* . 54** .368 .521
**p<.01
*p ^,05
58
TABLE 11
rs SUMMARY TABLE FOR TIME IN SA VERSUS PRESSURE, SA RATE
VERSUS PRESSURE, AND TIME VERSUS
RATE, FOR EACH VOLUME
HENRIETTA
rs
VOL. N T VS P RT VS P T VS RT
rs crit
.05
rs crit
.01
20 cc. 5
o
•
i
-.30 .80 .900 1.000
30 cc. 26 .06 -.11 . 56** .329 .465
40 cc. 15 -.12 .05 -.18 .440 .623
**p^.01
59
TABLE 12
rs SUMMARY TABLE FOR TIME IN SA VERSUS PRESSURE, S^ RATE
VERSUS PRESSURE, AND TIME VERSUS
RATE, FOR EACH VOLUME
GIGI
rs
VOL. N T VS P RT VS P T VS RT
rs crit
.05
rs crit
.01
30 cc. 16 -.42 -.51* . 72** .425 .601
40 cc. 20 -.44* -.17 .71** .377 .534
60 cc • 15 -.43 .09 .57* ,440 .623
70 cc. 10 -.67* -.45 .62* .564 .746
80 cc • 12 -.47 -.06 .33 .506 .712
**p<.01
*p <.05
60
TABLE 13
rs SUMMARY TABLE FOR TIME IK S^ VERSUS PRESSURE, SA RATE
VERSUS PRESSURE, AND TIME VERSUS
RATE, FOR EACH VOLUME
MILLIE
rs
VOL. N T VS P RT VS P T VS RT
rs crit
.05
rs crit
.01
20 cc. 5 -.80 , 90a -.60 .900 1.000
30 cc. 11 .59* .76** • 64* .535 .729
40 cc. 21 -.39* -.01 .42* .368 .521
70 cc. 17 -.14 .45* .34 .412 .538
80 cc • 11 .11 .06 .61* .535 .729
**p< ,01
*p <.05
ap = .05
occurred with a decrease in pressure; an increase in rate
occurred with a decrease in pressure; and an increase in
time occurred with an increase in rate. Henrietta showed
significance in one comparison - an increase in time with
an increase in rate - and only at 30 cc. of fluid, Gigi
showed the strongest relationships between time and
pressure and between time and rate, Millie showed both a
significant positive and a significant negative relation
ship between time and pressure, at 30 and 40 cc., respec
tively, Her data also indicate a generally positive
relationship between rate and pressure and between time
and rate,
O.S. and Humber of S^*s
The O.S. data were grouped according to the number
of reinforcements (S^-) in the previous S^ for presentation
in Table 14. The number of O.S.' falling in each category
of S^'s and the mean O.S. value for each are given in
Table 14 for each S, For Donna, 88.7% of the reinforce
ments varied within the range of 4 to 11, for each S^»
or, only 11,3% of the number of reinforcements were either
below 4 or above 11, For Henrietta, 82.8% of the S^-'s were
between 4 and 10; for Gigi, 90.97. were between 4 and 8;
and for Millie, 74.8% were between 4 and 9. The mean O.S.
values given in Table 14 are graphically represented in
Figure 21.
62
TABLE 14
TOTAL MEANS AND N'S FOR LENGTH OF O.S. GROUPED
BY PREVIOUS NUMBER OF SR,S FOR EACH S*
DONNA HENRIETTA GIGI MILLIE
SR N MEAN N MEAN N MEAN N MEAN
3 1 57.0 1 17.0 4 35.0
4 11 39.4 6 21.7 21 33.8 12 39.3
5 24 26.3 13 29.1 21 28.6 21 43.8
6 16 28.1 14 21.1 21 18,9 16 39.5
7 16 25.4 15 23.6 23 15.6 14 41.8
8
17 21.9 12 26.8 14 18.6 8 39.6
9 10 17.3 4 30.8 3 8.7 6 44.8
10 2 24.5 8 12.2 2 2.5 3 27.0
11 6 12.3 3 12.3 2 26.5 4 41.2
12 1 17.0 2 11.5 5 36.8
13 2 9.0 2 14.0
1 40.0
14 1 17.0 4 15.0 5 34.8
15 1 0.0 2 13.0 1 2.0 2 30.0
16 1 23.0 1 5.0 1 36.0
17 2 9.0
18 1 5.0 1 38.0
19 1 3.0 1 18.0
20
21 1 23.0
22
23 1 2.0
24
25
26
27
28 1 3.0
*A11 first data eliminated from calculations
63
56 DONNA
HENRIETTA
52
GIGI
48 MILLIE
44
40
co
* 28
o
fn
O
W
24
CQ
S
20
16
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
NUMBER OF PRIOR SR,S
Fig. 21. Means of totals for length of O^S,,
grouped by previous number of SR 's. (All first data
eliminated from calculations.)
64
Kruskal-Wallis one-way analyses of variance by ranks
were performed on the O.S, data, grouped according to the
previous S^. The degrees of freedom and a 2 values are
presented in Table 15 by specific volume and for the total
O
O.S, data, for each S. Few of the individual % values
for each volume were significant. However, all of the
total values were significant.* Donna, p<.02; Gigi, p<.05;
and Millie, p^.05; except Henrietta's, which was almost
significant at the .05 lever (%2 = 22.12} eQ5 =
22.36).
Threshold
To examine more fully the process of discrimination,
it was necessary to consider the point at which discrim
ination was possible, or threshold. It has been suggested
(Table 2) that 40 cc. was the volume of fluid needed to
provide the necessary bladder pressure for discrimination
to occur. However, since the recordable pressure within
the bladder varied due to contraction and relaxation, the
pressure at the instance the subject began lever pressing
in may be a more accurate description of threshold,
suggesting that no discriminable difference existed
between that pressure in and the resting pressure
in SD .
Threshold values were calculated as the pressure at ;
the beginning of each FR burst in (24 or more presses,
such that one would be missed) minus the mean pressure
65
TABLE 15
SUMMARY TABLE FOR KRUSKAL-WALLIS ONE-WAY ANALYSIS OF
VARIANCE BY RANKS, SHOWING X 1 AND df FOR EACH
VOLUME FOR LENGTH OF O.S. GROUPED
BY PREVIOUS NUMBER OF SR,S
DONNA HENRIETTA GIGI MILLIE
VOL. df
X2
df df
X2
df
a2
10 cc • 7 9.41 7 2.33 6 21.14*** 8 4.27
20 cc. 5 8.70 6 7.88 5 7.68 5 4.92
30 cc. 15 31.70** 10 12.86 5 6.70 6 6.08
40 cc. 8 2.66 5 1.80 4 4.00 9 3.40
60 cc• 6 11.60
70 cc. 5 5.56 10 7.77
80 cc. 3 4.43 5 4.36
TOTAL 18 38.96** 13 22.12a 10 18.31* 14 24.67*
***p <.01
**P <«02
*P <» 05
aPC2crlt .05- 22.36
66
in the previous S^. Means and standard deviations of the
threshold data were calculated according to volume and
appear for each S in Table 16. For all Ss the means
increased with the increase in volume. The standard
deviations, however, were lowest at the higher volumes for
Donna, Henrietta, and Gigi; but lowest at the lower
volumes for Millie.
Summary of Results
Donna and Henrietta learned the discrimination in
excess of the criterion at 40 cc. Gigi and Millie did not
reach criterion after 80 cc. of fluid. Generally, all Ss
showed an increase in ratio, increase in bladder pressure,
and decrease in O.S. with the increase in volume. The LAT.
data were inconsistent. Donna (the normal, non-operated S)
showed the greatest and most consistent relationships
between rate, time and pressure in S^, although some
significance appeared for the other Ss, especially
between time and rate. It was shown for all Ss that the
number of O.S. responses decreased as the number of S^-’s
in the previous S^ increased. Threshold bladder pressure
means and standard deviations were presented for each S.
Post-mortem examination
Photographs of the lower portions of the spinal
cords, showing the dorsal aspect, of Henrietta, Gigi and
Millie are presented in figures 22, 23 and 24, respectively.
For Henrietta, dorsal roots above T12 were intact
67
TABLE 16
MEANS AND STANDARD DEVIATIONS OF THRESHOLD BLADDER
PRESSURE (Mm. Hg) VALUES AT SA FR BURSTS
FOR EACH VOLUME OF FLUID
DONNA HENRIETTA GIGI MILLIE
VOL. MEAN S.D. MEAN S.D. MEAN S.D. MEAN S.D.
10 cc. 14.42 10.92
20 cc. 27.44 8.11 13.21 8.35 10.67 3.37
30 cc. 23.66 5.05 18.55 11.98 25.34 16.58 10,28 3.56
40 cc. 30.85 7.62 25.35 5.89 28.97 18.00 10.71 5.67
60 cc. 32.07 4.35
70 cc. 43.65 6.18 17.99 10.41
80 cc. 36.84 7.06 18.51 6.32
68
Fig. 22. Photograph of the lower portion of
Henrietta's spinal cord, showing the dorsal aspect, with
transected fibers numbered.
69
Fig. 23. Photograph of the lower portion of Gigi's
spinal cord, showing the dorsal aspect, with transected
fibers numbered.
70
Fig. 24. Photograph of the lower portion of Millie's
spinal cord, showing the dorsal aspect, with transected
fibers numbered.
71
bilaterally; and dorsal roots below L2 on the left side
and below L3 on the right side were intact# Thus, the
surgery for Henrietta was complete, with dorsal roots
severed including T12-L2, and with the sacral roots
intact.
For Gigi, all dorsal roots down to and including S2
remained intact. Those severed bilaterally were S3-S5,
The surgery, therefore, was not complete, as S2 remained
bilaterally intact. Roots to have been included were
.
For Millie, all dorsal roots down to and including
L5 remained intact, as well as S5. Thus, dorsal roots
which were bilaterally severed were S2-S4, and the surgery
was complete.
CHAPTER IV
DISCUSSION
Procedural Review and Experimental Objectives
After the Ss were trained to lever press for sugar
pellets on FR 24, dorsal roots were severed bilaterally as
follows: Henrietta - T12"L2' Gigi and Millie - S2-S4.
Donna received no surgery. After surgery, discrimination
training was begun, with S^* the absence of the interocep
tive stimulus, and SA , its presence, and was continued
until the Ss exceeded the S^/S^ response rate ratio
criterion of 10:1, or until it was determined that an S
could not learn the discrimination. The independent
variables were the stimulus volume and the degree of
intactness of the sensory pathways from the bladder; and
the dependent variables were lever pressing behavior and
bladder pressure.
The first objective of the experiment was to deter
mine whether the interoceptive stimulus of fluid presented
into the urinary bladder could be manipulated experimen
tally so as to function as a discriminative stimulus in a
normal subject. The results for Donna, the non-operated
control of the experiment, clearly indicate that operant
behavior may be brought under the control of such an
72
73
interoceptive stimulus. The present study has further
substantiated and extended earlier work on visceral
stimuli, using operant techniques, by Slucki et al, (1965)
and McCoy (1968;Slucki et al., 1969), in which the small
and large intestines, respectively, were stimulated.
The second objective of the experiment was to
determine whether the interruption of the sensory nerves
from the bladder to the spinal cord would be sufficient to
prevent the stimulus from taking on discriminative
functions. Relative to this, it was predicted that the
severing of the dorsal roots of S2-S4 , believed to contain
the majority, if not all (Edvardsen, 1968), or the sensory
nerves from the bladder, would eliminate the Ss ability to
learn the discrimination; whereas, severing the dorsal
roots of T22-L2 , believed by some (Adam, 1967; Ruch, 1960;
Truex & Carpenter, 1964) to carry only some of the sensory
nerves from the bladder, would not interfere with the S/s
ability to learn the discrimination. It will be recalled
that Henrietta was given the T12"^2 surgery, and Gigi and
Millie, the sacral surgery. However, the surgery for Gigi
was incomplete (Figure 23), leaving S2 bilaterally intact.
The bulbocavarnosis reflex was absent in both Gigi and
Millie, indicating that at least some sacral afferent
nerves were not intact in either S, Gigi's behavior, since
her surgery was incomplete, should have shown, under
74
experimental conditions, some evidence of conditioning,
and should not have resembled the behavior of Millie,
whose sacral surgery was complete,
Basic Measures
From the S^/S^ response rate ratio data, Henrietta
(T];2“L2) reached the criterion of 10:1, as did Donna, the
non-operated control. Neither Gigi nor Millie reached this
criterion, and thus, their behavior on this variable
resembled each other (see Table 2), It is of interest to
compare the data on each subject (Figures 5-8) showing the
progress from the beginning of discrimination training to
late in training, at 40 cc,, when Donna and Henrietta were
discriminating. All Ss showed erratic behavior during the
"A" segments, in response to the lack of reinforcement
during the periods. Later in discrimination training,
behavior was more consistent. The stippled areas during
the "B" segments of the graphs of Donna and Henrietta
(Figures 5 and 6, respectively) are very small; whereas,
those areas for Gigi and Millie (Figures 7 and 8, respec
tively), although smaller than at the beginning of
discrimination training, are still much larger than those
of Donna and Henrietta, The fact that the rates during
for Gigi and Millie were smaller late in discrimination
training than at the beginning can be explained without
bladder conditioning having occurred. What, in effect,
happened was the differential reinforcement of pauses
75
during (the longer the pause, the sooner the switch to
S^), The switch rarely was made until after one minute
had passed; so the probability was good that the £ would
get reinforced after a long pause which occurred after the
one minute mark. Thus, lever pressing during S^was
commonly characterized by long pauses and bursts of lever
pressing behavior (see Figures 11 and 12), and therefore,
the reduction in the rate. The rate, however, did not
drop as low as those of Donna and Henrietta, such that the
10s1 criterion was not reached.
To reiterate, it is thus far apparent that the second
objective of the experiment was accomplished. The subjects
with sacral surgery could not learn the discrimination,
according to the data considered to this point; whereas,
the S, with the T12-^2 surgery could. However, more
clarification is necessary to determine any behavioral
differences between Gigi and Millie, since Gigi’s surgery
was incomplete.
If discrimination learning were to occur, it would
be predicted that changes in the basic measures other than
the sP/S^ ratio, already discussed, (i.e., bladder pres
sure, O.S., and LAT.) would be as follows.
(a) Bladder pressure would increase with the in
crease in volume. This bladder pressure increase should
occur whether conditioning is established or not, since
this variable is primarily physiological, although there
may be a behavioral component. The subject could learn to
control the pressure within the bladder, if there are
reinforcement contingencies associated with it. This
concept will be discussed later. According to Edvardsen
(1968), severing of the dorsal roots at the sacral level
has the same effect on subsequent bladder pressure as
severing the sympathetic efferents, which normally serve to
relax the bladder. The effect in the normal subject after
the infusion of fluid into the bladder would be an initial
large rise in pressure followed by a slow decrease in
pressure over about 2 minutes to a relaxed level, which is
somewhat higher than the resting pressure (with no fluid
present in the bladder). After sacral surgery, there
should be little if any of the relaxation, due to the
interruption of the sympathetic feedback loop. This lack
of relaxation would be an indication of the completeness
of surgery.
(b) It would be predicted that the O.S. would drop
as training progressed, and that specifically, if training
were complete, the O.S, values would fall below 24 presses,
suggesting that the S, stopped responding due to the pre
sence of the stimulus, rather than because she missed a
reinf orcement.
(c) It would be predicted that the LAT. would drop
as training progressed, since the off-set of the stimulus
should become associated with available reinforcement.
Donna, the non-operated control S, performed much as
predicted. Her ratio and pressure increased with the
increase in volume; and the O.S. dropped with the increase
in volume, so that it was less than 24 responses (see
Tables 4 and 8 and Figure 13). However, the 1AT. remained
fairly constant across volumes at about 15-20 seconds, not
dropping, as predicted; although it was short throughout
the experiment. The trends in the ratio, the pressure and
the O.S., as well as the lack of variation in the LAT.,
are supported by the Kruslcal-Wallis analyses in Table 8,
It should be recalled that Donna's fluid volume was
reduced to 10 cc. after successful conditioning at 40 cc.
The resulting ratio at 10 cc. was higher than previously
at 30 cc. Similarly the O.S. and LAT. values at 10 cc.
were lower than at 20 cc. and 30 cc., respectively. Thus,
it can be suggested that, although Donna originally could
not learn the discrimination at lower volumes (20 and 30
cc.), after conditioning at a higher volume, she became
more sensitive to the stimulus at 10 cc. Atypical
sequence of behavior for Donna is shown in the cumulative
record in Figure 9. During S^, lever pressing occurred at
a steady rate. At the onset of S^, there was a small O.S.
of under 24 responses. Pressing did not resume until
shortly after the switch to S^, which was again at a steady
rate. Occasionally, during S^, there were FR bursts (24
or more presses), but such bursts are not shown in Figure
78
9. The corresponding bladder pressure record shows the
increase in pressure at the beginning of S^, and the
gradual tapering off of this pressure, due to the
sympathetic feedback system. The large spikes which are
riding on the basic bladder pressure curve are presumed to
be skeletal muscle movement artifacts (Porter, 1969).
Henrietta, with surgery at T12-L2, behaved much like
the normal subject, and her results were as predicted,
except for the IAT. Generally, with the increase in
volume, there were an increase in ratio, an increase in
pressure, and a decrease in O.S. to less than 24 responses
(see Table 5 and Figure 14). Although only the ratio
analysis was significant on the Kruskal-Wallis (Table 8),
the trend for both of the others of these two measures
approached significance in the appropriate direction. The
LAT. was significant in the opposite direction to that
predicted; that is, with the increase in volume, the LAT.
increased, rather than decreased; however, the mean values
even at the larger volumes were less than 30 seconds. A
possible explanation for this finding may be due to com
plications from the surgery. Toward the termination of the
experiment, Henrietta became somewhat listless, and did not
always consume all food given to her. The wound on her
back from the surgery opened sLightly and was oozing fluid.
Medication was applied, and the conditioned improved, but
her physical condition may have had its effect on her
subsequent behavior in the experimental session#
Henrietta's cumulative record (Figure 10) showed a steady
rate during a short overshoot of under 24 responses
(not in the example shown), very little responding during
s A and not a long latency into s A Henrietta's
behavior differed somewhat from Donna's, in that Henrietta
did not respond with FR bursts, but rather with an
occasional press or two, as in Figure 10, such that the
rate during responding differed greatly from to sA
The corresponding bladder pressure record was similar to
Donna's with an initial increase in pressure in s A
tapering off to a lower level. Overall, however, her
behavior was enough like that of Donna, the normal S, that
it is concluded that the surgery did not alter her ability
to learn the discrimination, and thus, did not interrupt
nerve pathways carrying sensory information from the
bladder#
The behavior of Gigi, with incomplete surgery at
S2-S4, although not reaching criterion on the SD/S^ ratio,
resembled that of the normal S (Donna) on the other
measures. Her results for pressure, O.S. and LAT. were as
predicted for a subject completing training, and these
three trends were all significant on the Kruskal-Wallis
analyses (Table 8), With the increase in volume, there
were an increase in pressure, a decrease in O.S. to below
24 responses, and a decrease in IAT. (see Table 6 and
80
Figure 15). The fact that the O.S. values dropped and
remained far below 24 responses suggests strongly that
some amount of discrimination of the interoceptive
stimulus was occurring. However, Gigi continued to show
frequent FR bursts during S^, so the S^rate was high,
thereby deflating the S^/S^ ratio. The representative
cumulative record in Figure 11 exemplifies Gigi’s
behavior. She stopped responding almost immediately after
the onset of S^, and began almost immediately after the
offset of S^. Also similar to the normal S, after the
initial rise, the bladder pressure tapered off to a low
level. To recapitulate, both Gigi's cumulative record and
bladder pressure record are much like those of the normal
S,, and not as would have been predicted for one with the
dorsal roots at S2-S4 severed.
The behavior of Millie, with complete surgery at
S2-S4, although similar in some respects to that of the
normal S (Donna), differed sufficiently from that pre
dicted to suggest the conclusion that she did not and
could not learn the discrimination. The Kruskal-Wallis
analyses (Table 8) on all the basic four measures for
Millie showed significance. However (Table 7 and Figure
16), the ratio increased with the increase in volume, but
did not reach criterion; and the O.S. did drop with the
increase in volume, but did not fall below the crucial 24
responses (mean O.S. varied between 41.35 and 34.91
81
responses), suggesting that she stopped responding, not
because of the presence of the stimulus as with the other
Ss, but rather because she missed a reinforcement. The
bladder pressure increased with the increase in volume,
as was predicted for all Ss, regardless of the surgery.
Millie's bladder pressure record (Figure 12) also differed
from those of the other Ss, in that there was always only
a small difference between SD and pressure. (The mean
change in pressure varied from 10.28 to 19.63 mm. Hg
throughout the experiment.) The comparison across Ss is
shown in B’igure 18. Also, after the initial increase in
pressure with the introduction of fluid into the bladder
(S^) , there was no succeeding lowering in pressure due to
relaxation of the bladder, from the sympathetic feedback
loop. This lack of feedback was as predicted, since the
severing of the dorsal roots at levels S2-S4 presumably
interrupted that loop. Table 8 shows that the LAT, for
Millie significantly increased (.05 level) with the
increase in volume to a high of 1,39 minutes (Table 7).
As can be seen in Figure 20, this value was much higher
than any of the values for the other Ss, It will be
recalled that Henrietta's LAT. also increased with the
increase in volume, but not with this magnitude. It is
probable that as training progressed for Millie the
probability of obtaining reinforcement became less and less
predictable, as she was not learning the discrimination.
82
Her pauses during S^ became longer between bursts, and
thus, the pauses at the beginning of were longer. A
typical example of Millie's behavior is presented in the
cumulative record of Figure 12. Her responding during S*5
was at a steady rate, which continued with a long overshoot
into S^. There was very little responding during sA
followed by a long LAT. into S^.
When comparing the data of the four Ss (Figures
17-20), it becomes obvious that although Gigi and Millie
were similar for the S^/S^ ratio (Figure 17), and therby
different from Donna and Henrietta who exceeded criterion,
they differed highly on their data for the other measures.
Gigi's bladder pressure (Figure 18) behaved much like that
of Donna and Henrietta, and not at all like Millie's. The
O.S, (Figure 19) dropped below 24 responses for all Ss
except Millie, so again Gigi resembled Donna and Henrietta,
and not Millie. The LAT, (Figure 20) data differed among
all of the Ss, so it is difficult to draw a comparason.
However, Gigi was the only S showing significance in the
direction predicted for an S learning the discrimination.
A further comparison can be made across Ss on the
rank order correlation data presented in Table 9, between
the change in pressure and the corresponding S^/S^ ratio
for each day. The only significant relationship between
these two variables appeared for Gigi (p<1.01), although
the comparison for both Donna and Henrietta closely
83
approached significance. The correlation for Millie was
practically zero (.08), showing no relationship between
pressure and ratio.
Therefore, it can be concluded that enough
behavioral evidence exists to support the observed
incompleteness of Gigi's surgery (i.e., Gigi's O.S.
dropped to below 24 responses, and her IAT. dropped with
the increase in volume - both predicted for an S with
sensory fibers from the bladder intact). Some physiolo
gical evidence also exists to support the observed
differences between Gigi and Millie, Millie exhibited
bowel incontinence for a few weeks post-operatively;
whereas, Gigi showed bowel control. Also, Millie was the
only £3 to develop a decubitus ulcer on the seat pad. These
two effects, which occurred exclusively with Millie,
provide physiological support for the fact that her sacral
surgery was complete and that Gigi's was incomplete.
Since physiological and behavioral differences existed
between Gigi and Millie, and since Millie could not learn
the discrimination, then severing the sensory roots of
S2-S4 completely eliminated the ability of the to learn
the discrimination. The second objective of the experiment
was thus accomplished, since the interruption of the
sensory nerves from the bladder to the spinal cord was
sufficient to prevent the stimulus from taking on
discriminative functions.
84
O.S. and LAT.
The results of this study regarding the O.S. and LAT.
are consistent with those of the two previous operant
interoceptive studies conducted (Slucki et al. , 1965?
McCoy, 1968), stimulating the small and large intestines,
respectively. Initially, both of these earlier studies
presented the S^ and the S^ in the more conventional
manner? that is, with the presence of the stimulus, and
S^ the absence, which is the reverse of the presentation
in the present study. In the prior studies, the experi
menters found that the O.S. 's were long and the LAT. 's were
very short. The previous studies did not quantify the
amount of overshoot in the same manner as in the present
study. In the Slucki et al. study (1965), was ter
minated at any time in the FR sequence, so the amount of
O.S. in terms of a missed could not be quantified and
compared to the present study. In the present study and
the McCoy study (9168), S*3 was normally terminated
immediately following a reinforcement. The amount of
O.S. was defined as time in the McCoy study, but as the
number of responses in the present study, although McCoy
did record the number of responses in the O.S. That
number remained above 24 throughout the McCoy study, so it
is possible that the S stopped responding because he
p
missed an S . This notion was considered by McCoy who
defined the O.S. as lasting no longer than 30 seconds.
It is important to note that the LAT.'s of those Ss
learning the discrimination in the present study did not
exceed 30 seconds. This 30 seconds refers to the time of
stimulus offset in both stiuations. In the Slucki et aL.
(1965) study, the S^ and S^ conditions were reversed after
original training, and the respective lengths of the O.S.
and the IAT. also reversed, such that the O.S. was short,
as in the present study. The authors concluded, "the
onset of stimulation is immediately discriminated but its
termination is not so readily perceived" (p. 413). They
suggested that the effect was due to some characteristic
of the stimulation itself. The present author agrees with
this conclusion. As discussed earlier, the bladder
accomodates to the presence of the stimulus through the
sympathetic feedback loop. It is probable, therefore,
that the removal of the stimulus when the bladder is
already in a somewhat relaxed state is not as readily
perceived as the introduction of the stimulus into a
bladder which is empty.
McCoy (1968) found that the characteristics of
interoceptive and exteroceptive stimuli differed when
either a light or the visceral stimulus functioned as S^'s
within the same experiment. There were 30 or more
responses into (O.S.) upon the termination of the
visceral stimulus, but few, if any, upon the termination
of the exteroceptive stimulus. Thus, differences existed
86
between the effects of the two stimuli, one interoceptive
and one exteroceptive, Bykov (1959) and Adam (1967)
stated that learning rates with internal stimuli are
slower than with external stimuli. However, there was no
way to match the characteristics of the stimuli, DiCara
and Miller (1969) found that the visceral learning of
noncurarized rats was much poorer than of curarized Ss,
suggesting that the curare eliminated sources of variation
and distraction coming from the skeletal muscles,
A further result involving the O.S., which is
consistent with that of McCoy (9168) is the trend which
existed for each S for the number of O.S. responses to
decrease as the number of previous reinforcements increased
reflecting the greater probability that reinforcement
would be available after a few reinforcements than after
many, McCoy found the effect to an even greater extent
than in the present study. It is possible that the effect
is a characteristic of the FR schedule, and would not
occur with variable ratio (VR) or variable interval (VI)
schedules, in the same way, since the probabilities of
reinforcement are not so readily discernable in these
latter schedules.
By using the and in a manner, as in the
present study, information has been gained with respect to
the O.S., since discriminating Ss stopped responding due
to the presence of the stimulus, rather than a missed
reinforcement. However, some information may have been
lost due to the relaxation of the bladder during S^. By
using SD as the presence of the stimulus, in an intero
ceptive study of the bladder, the experimenter could more
readily utilize short stimulus-on periods, thereby not
including segments with the stimulus present during which
the bladder might be in a non-contracting or a relaxed
state. Thus, the and S^ would be not only operationally
defined, but also would be defined physiologically. In
the present study, physiological definitions were not
discrete because of the variation in bladder pressure.
The problem was that was too long and the bladder had
too much time to vary in pressure and relax. Ss were able
to discriminate gross changes in pressure (S^ to S^),
since conditioning was successful, so it could be deduced
that they were also discriminating, or at least sensitive
to, variations in pressure within the period.
Rate. Time and Pressure in
Since so much variability did exist in the
pressure, the analyses of the interrelationships between
the rate, the time in S^, and the pressure in S^ were
performed in order to determine if the Ss were responding
to changes in pressure within S^. If so, an increase in
rate is predicted to occur with a decrease in pressure.
However, one cannot merely draw a relationship between the
pressure in any individual S^ and the rate of lever
88
pressing during the same without considering time.
According to the sympathetic feedback loop, a lower over
all pressure in any specific is predicted with the
increase in time spent in that S^. Also, an increase in
time spent in the would suggest an increase in lever
pressing, since the probability that the behavior will
get reinforced increases as time passes in S^, Thus, the
relationship between rate and pressure would be greatly
influenced by time. It is predicted for the normal £3,
having learned the discrimination, that all three of these
relationships would exist. For Donna (Table 10), the
normal S, the relationships were all significant in the
appropriate direction at both 30 and 40 cc. of fluid
stimulus. However, at 10 cc. the correlation between rate
and pressure was practically zero, even though the other
two comparisons were still significant, suggesting that
the S could not perceive changes in pressure at the low
volume. Since the 10 cc. were presented to Donna after
training was complete at the larger volume of 40 cc., any
perceived changes in pressure would have been discrimi
nated. Thus, at least for the normal S., there is
evidence to suggest that she was discriminating changes
in pressure within the period at 30 and 40 cc.
Henrietta's data (Table 11) showed very little
relationship in any of the comparisons, except between
time and rate, and then only at 30 cc. These results are
89
surprising in that Henrietta otherwise responded in very
much a normal manner, as the surgery did not appear to
interfere with her ability to learn the gross discrimi
nation. A.s mentioned earlier, however, there were
complications from the surgery, possibly accounting for
abnormalities in her behavior.
The relationship between rate and pressure for Gigi
was only significant at 30 cc., although there was more of
a correlation between both of the other two measures. It
is probable that the lack of relationship between rate and
pressure was-’ due to the partially complete surgery which
did eliminate some of Gigi's ability to discriminate.
Millie's data are very difficult to interpret (Table
13). First of all, it is important to note that there was
very little variation in pressure within any particular
SA , suggesting that the results of either of the analyses
involving the change in pressure may be spurious, espec
ially since rs does not take magnitudes of difference into
consideration. The results, in fact, do show both a
positive and negative correlation between time and
pressure, and a positive correlation between rate and
pressure. This latter relationship is opposite to that
being conditioned, so it is unlikely that it was a
learned relationship. The relationship between time and
rate, which should be a function of the training procedure
and not be related to surgery or what is happening in the
90
bladder, was as would have been predicted for Millie;
that is, with the increase in time in S^, there was a
corresponding increase in rate.
The fact that the normal S showed signs of a rela
tionship between rate and pressure within an individual
is further evidence that conditioning did, in fact, occur.
This ability for fine discrimination suggests possible
future research in which reinforcement would be contingent
upon the pressure in the bladder; that is, when the
pressure dropped to a certain point in S^, would become
effective. In this way the S could learn to control his
own reinforcement.
Threshold
Further support for the notion that the Ss may have
been discriminating varying pressures within can be
gained from the threshold data in Table 16. As indicated
in the results* chapter, the means' data varied similarly
from subject to subject. However, since the standard
deviations (S.D, 's) were lowest at the higher volumes for
all Ss except Millie, it appears that the smaller S.D.*s
for these three jSs are indicative of the presence of
discrimination learning. It seems reasonable that less
variability would exist for the point at which the S began
responding after discrimination training was accomplished.
For Millie, who never learned the discrimination, the S.D.
increased as the pressure withing the bladder increased.
A psychophysical type of threshold study is necessary,
using a series of volumes. The above proposed study in
which would be contingent upon the pressure in the
bladder could function in this respect. The pressure
would then be the independent variable which could be
manipulated by the E as the onset of until the S began
responding 50% of the time when the pressure dropped to
that level.
Experimental Problems
As was mentioned in the procedural section, the
dissecting microscope was not employed in the surgery for
Gigi. Since the possibility exists that her subsequent
incompleteness of surgery was due to this fact, it seems
imperative that the dissecting microscope be used for such
surgery in future research.
When tested, the pump, syringe and tubing system
appeared airtight when withdrawing fluid from a flask.
When attached to the monkey, via the catheter, air
occasionally was forced into the system during the with
drawal of the stimulus. A negative pressure or vacuum was
created, as evidenced by the downward sweep of the pen in
the bladder pressure records (Figures 9-12), forcing air
into the system either at weak connection points or from
around the catheter, or both. If air were entering at
connection points, all would need to be thoroughly sealed.
92
If air were entering from around the catheter, the air
could have introduced an exteroceptive stimulus. If
possible a larger catheter should be used.
As mentioned in the results' chapter, some internal
drift was inherent in the functioning of the dynograph,
making the measurement of, or reliance upon, absolute
pressure values (based on zero) difficult and inadvisable.
For this reason the change in pressure from to was
employed. It was unfortunate to lose such data infor
mation, so more stable recording apparatus would be
necessary for future research.
Conelusions
Both objectives of the experiment were accomplished.
Operant behavior was brought under the control of an
interoceptive stimulus, reflecting both the initiation and
the termination of the visceral stimulation, as in pre
vious studies using operant techniques (McCoy, 1968;
Slucki et al,, 1965; Slucki et al,, 1969), and in contrast
with respondent conditioning methods (^dam, 1967) which
reflect the initiation of the stimulus, but not the
termination. The interruption of the sensory nerves from
the bladder to the spinal cord was sufficient to prevent
the stimulus from taking on discriminative functions.
Since the complete surgery at spinal level S2-S4 elimi
nated the ability to learn the discrimination, and since
the surgery at lever T12-L2 had no apparent effect on the
93
ability to learn the discrimination, credence is lent to
the hypothesis of Edvardsen (1968) that all of the sensory
nerves from the bladder to the spinal cord travel with the
pelvic nerve and enter the cord at levels S2-S4.
Since behavior is a function of both internal and
external stimuli, the study of the viscera as a stimulus
source for operant conditioning is vital. Internal stimuli
can be governed by reinforcement contingencies, just as
external stimuli, even though stimuli emenating from the
internal environment may be more subtle.
The results which have recently been obtained in the
operant conditioning of visceral responses (Miller, 1969)
have great significance in the understanding, cure and
possible prevention of psychosomatic disorders, A
visceral organ may be weak and may respond under some
stimulus situation with a specific response, as an
increase in peripheral blood pressure. This response may
have been socially reinforced, as by the withdrawal of
some noxious external stimulus. Later, the response may
be repeated often, gradually reducing the resilience of
the organ and leading to a psychosomatic disorder. Using
the blood pressure example, the individual develops a
condition known as hypertension, in which the blood
vessels lack necessary elasticity. DiCara and Weiss (1969)
found that rats that had been rewarded for decreasing
their heart rate learned well in a shuttle box, but that
those rewarded for increasing their heart rate did not
learn well. Miller (1969) and his associates found that
none of 40 rats rewarded for increasing their heart rates
died during training under curarization; whereas, 7 of 40
rewarded for decreasing their heart rates did die during
training. Thus, due to prior conditioning, Ss responded
differentially to later stimulus conditions. Such
characteristic patterns of visceral response could lead
to breakdown in specific organs, and eventually death.
In the context of psychosomatic medicine, indivi
duals can learn to attend to internal stimuli as cues to
possible abnormalities, opening the avenue to the
prevention of such disorders. As a point of interest, it
is possible that the hypochondriac is too discriminative,
that is, responding to internal stimuli which most
individuals do not discriminate.
95
CHAPTER V
SUMMARY
Interoceptive conditioning has been studied in the
Soviet Union and other Eastern European countries with the
use of Pavlovian techniques for forty years, but became of
interest in the United States only in the last decade,
with the application of operant techniques to intero
ceptive stimuli.
Studies in operant discrimination demonstrated that
the presence or absence of stimuli presented to the small
or large intestine were perceived by monkeys, whose
behavior was differentially reinforced.
The present experiment was designed to determine
whether the interoceptive stimulus, consisting of fluid
presented into the urinary bladder, could be manipulated
experimentally so as to function as a discriminative
stimulus in a normal subject, and to determine if the
interruption of the sensory nerves from the bladder to the
spinal cord would be sufficient to prevent the stimulus
from taking on discriminative functions. The independent
variables included stimulus volume and the degree of
96
intactness of the sensory nerves from the bladder} and the
dependent variables were lever pressing behavior and
bladder pressure.
Four experimentally naive female rhesus monkeys were
restrained in primate chairs, maintained on 22-hour
deprivation-feeding schedules, and trained to lever press
for sugar pellet reinforcement to a steady rate on an
FR 24 schedule. The dorsal roots of the spinal cord were
severed bilaterally in two Ss at S2-S4 (one of these
surgeries was incomplete) and in one S_ at T\2-l>2* All Ss
were catheterized and fluid was presented into the bladder
and withdrawn according to a programmed design. Lever
presses emitted in the absence of the stimulus (S^) were
reinforced on FR 24, and presses emitted in the presence
of the stimulus (S^) were extinguished.
The normal S > (non-operated) , as well as the S. with
surgery at ^12-^2* successfully discriminated between the
presence and absence of the stimulus beyond the criterion
of the S^/S^ response rate ratio of 10:1, The other two
Ss did not. There was some evidence of discrimination in
the S with the incomplete surgery at S2-S4, as reflected
by a decrease in the number of overshoot responses
(non-reinforced responses made after the switch to S^) to
less than 24 responses per S^, and a decrease in latency
until the first press after the switch to S^. The S, with
the complete surgery at S2-S4 showed no evidence of
97
discrimination Learning, ALL resuLts were evaLuated by
two non-parametric statisticaL tests: the KruskaL-WaLLis
one-way anaLysis of variance by ranks and the Spearman
rank order correlation.
The overshoot and Latency results were in agreement
with those of previous interoceptive studies using operant
techniques, in which stimulus onset was quickly discrimin
ated, but stimulus offset was not, suggesting that the
effect was due to characteristics of visceral stimulation.
That the adequate stimulus was pressure was
evidenced by the fact that the reflexive relaxation of the
full bladder decreased the pressure and produced responding
by the monkey, indicating that (absence of fluid in the
bladder) had been initiated. Thus, a more adequate
control of bladder discrimination would be effected by the
manipulation of pressure and not volume.
Implications of the study for psychological theory
and an understanding of psychophysio logical mechanisms
underlying psychosomatic disorders were considered.
REFERENCES
Adam, G. Interoception and Behavior, Budapest:
Akademiai Kiad<5, 1967.
Bykov, K. M. The Cerebral Cortex and the Internal Organs, i
(Tr. by R. Hodes)• Moscow: Foreign Languages
Publishing House, 1959,
Cook, L., Davidson, A., Davis, D. J. & Kelleher, R. T.
Epinephrine, norepinephrine, and acetylcholine as
conditioned stimuli for avoidance behavior.
Science. 1960, 131, 990-991.
DiCara, L. V. & Miller, N. E. Heart-rate learning in the
noncurarized state, transfer to the curarized state,
and subsequent retraining in the noncurarized state.:
Physiology and Behavior. 1969, 4, 621-624.
DiCara, L. V. 6c Weiss, J. M. Effect of heart-rate
learning under curare on subsequent non-curarized
avoidance learning. Journal of Comparative and
Physiological Psychology, 1969, 69, 368-374,
Edvardsen, P. Nervous control of urinary bladder in cats
I. The collecting phase. Acta Physiologica
Scandanavia. 1968, 72, 157-171.
Frick, F, C. An analysis of operant discrimination.
Journal of Psychology. 1948, 26, 93-123.
Hays, W. L. Statistics for Psychologists. New York:
Holt, Rinehart 6c Winston, 1963. i
McCoy, F. B. Operant discrimination of a mechanical
stimulus in the colon. Masters Thesis, University
of Southern California, 1968.
Miller, N. E. Learning of visceral and glandular
responses. Science. 1969, 163, 434-445.
Pavlov, I. P. Conditioned Reflexes. (Tr. by G. V.
Anrep.) London: Oxford University, 1927.
Porter, R. W. Personal communication, 1969.
98
99
Razran, G. The observable unconscious and the inferable
conscious in current Soviet psychophysiologyi
Interoceptive conditioning, semanitc conditioning,
and the orienting reflex. Psychological Review.
1961, 68, 81-147.
Ruch, T. C. The urinary bladder. In T. C. Ruch 6c J. F.
Fulton (Eds,). Medical Physiology and Biophysics.
Philadelphia* Saunders, 1960. Pp. 955-962.
Schuster, C. R. 6c Brady, J. V. The discriminative control
of a food reinforced operant by interoceptive
stimulation. I. P. Pavlov Journal of Higher
Nervous Activity. 1964, 14, 448-458.
Sidman, M. Tactics of Scientific Research. New York:
Basic Books, 1960,
Siegel, S, Nonparametric Statistics for the Behavioral
Sciences. New York: NcGraw-Hill, 1956.
Skinner, B. F. Behavior of Organisms. New York:
Appleton-Century-Crofts, 1938.
Skinner, B. F. Behaviorism at fifty. Science. 1963, 140,
951-958.
Slucki, H. Unpublished research, 1969.
Slucki, H ., Adam, G 6c Porter, R. W, Operant discrimination
of an interoceptive stimulus in rhesus monkeys.
Journal of the Experimental Analysis of Behavior.
1965, 8, 405-414.
Slucki, H ,, McGoy, F . B. 6c Porter, R. W, Interoceptive
S^ of the large intestine established by
mechanical stimulation. Psychological Reports.
1969, 24, 35-42.
Truex, R. C. 6c Carpenter, M. B. Strong and Elwvn's Human
Neuroanatomv. Baltimore: VJi 1 liams 6c Wilkins , 1964.
APPENDIX
Statement of the Problem
The first objective of the original experiment was
to determine whether the interoceptive stimulus of fluid
presented into the urinary bladder could be manipulated
so as to function as a discriminative stimulus. The
second objective was to determine whether the interruption
of the sensory nerves from the urinary bladder would be
sufficient to prevent the subject from performing the
learned discrimination. On the basis of Edvardsen's (1968)
work (see Chapter I of this paper), it was predicted that
severing the dorsal roots of S2-S4 would eliminate
discrimination; whereas, severing the dorsal roots of
^12“^2 would have no effect on the learned discrimination.
The independent and dependent variables were as defined in
Chapter I,
Method
Subjects. See Chapter II.
Apparatus. The apparatus were the same as those
described in Chapter II, with the following addition. The
original apparatus used to record bladder pressure was the
Sanborn physiological pressure transducer and Carrier
pre-amplifier with photographic printout, which ceased to
100
101
function upon completion of initial training. The
replacement (Beckman-Offner Type RS Dynograph) was not
obtained until one month after surgery.
Procedure. The procedure consisted of four basic
steps, taken in this order: (a) initial training,
(b) catheterization and discrimination training,
(c) surgery, and (d) testing.
(a) Initial training. See explanation in
Chapter II.
(b) Catheterization and discrimination training.
See explanation in Chapter II, making the following
additions and changes. The purpose for using the S^ as
the absence of the stimulus and the as the presence is
explained as follows. It might have been necessary to
distinguish between behavior effected by cutaneous
anesthesia (Slucki, 1969) and that effected by the S's
inability to discriminate, if sensory input from the
viscera had been eliminated. If an anesthetic effect
were present after surgery, there might have occurred a
suppression of all lever-pressing behavior. If all
sensory information from the bladder were interrupted,
then the S would continue to lever press into S^ until at
least one reinforcement were missed. If the S^ were the
presence of the interoceptive stimulus, then a suppression
of behavior would have occurred in either case and the
results would have been confounded.
102;
Discrimination training continued for Henrietta,
Gigi and Millie with 10 cc. of interoceptive stimulus,
until the criterion of lOtl, S^/S^ response rate ratio
was exceeded. This procedure took two months of daily
experimental sessions. Discrimination training at 10 cc.
continued for Donna until there had been no improvement
in the S^/S^ ratio (remaining below criterion) for 10
days. After nine weeks of discrimination training, the
volume for Donna was incremented to 20 cc.
(c) Surgery. See explanation in Chapter II,
making the following change, Donna received no surgery,
since she had not yet reached the S^/S^ ratio criterion.
(d) Testing. The Ss having had the surgery were
catheterized and the regular procedure for discrimination
training was followed to test whether the Ss were able to
make the discrimination.
Results
When tested after surgery, all three Ss were able
to discriminate. This result was not at all as predicted.
i
Possible explanations were either that the surgeries were
j
incomplete, that there were other nervous pathways '
mediating sensory information from the bladder, or that
an exteroceptive stimulus, associated or correlated with
the interoceptive stimulus, was present and being
discriminated. The last of these posibilities was
considered.
103
E observed the behavior of the Ss during a
subsequent experimental session and found that on each
instance the toggle switch was actuated, each S moved her
ears and looked toward the speaker delivering white noise
for sound attenuation. She immediately began pressing, if
it were onset, or stopped pressing, if it were onset■
In the experimental chamber E could hear a faintly audible
* static click from the speaker, whenever the toggle switch
was actuated. E unhooked the syringe from the pump, so
that the actuation of the toggle switch did not infuse or
wi’ -'.idraw fluid, and the Ss discrimated immediately S^ or
S^. Additionally, after Gigi looked at the speaker, she
turned and looked at the tubing, carrying the fluid to
and from the bladder. E could see bubbles in the tubing,
which traveled one way at S^ onset and the other way at
S^ onset. Thus, the Ss appeared to be discriminating
some combination of exteroceptive stimuli.
To combat the click problem, three changes were i
instituted in the following order.
(a) Since the source of the clicks was believed to j
■ be 60 hertz static interference, and since the noise
i
| generator and pump apparatus received power from the same
i A.C. outlet, a transformer was added to the power line
I
j between the relays of the pump mechanism and the outlet,
| to prevent any electrical discharge from the toggle switch
i ;
! and the pump relays from reaching the noise generator and
104 ;
i
i
the speaker. This seemed to eliminate some of the inter- I
ference, but occasionally E could hear a click from the |
i
speaker. Ss still appeared to be responding to clicks,
(b) In order to mask the static clicks, a random j
click generator (BRS PPl) was programmed into the BRS-
Foringer equipment. Ss responded to most clicks, both
those from the click generator, and those from the toggle
switch and relays.
(c) The noise generator was removed from the
equipment rack and was plugged into another A.G. circuit
different from that used for the pump, so that the power
source of the noise generator and that of the pump could i
be separated. E could no longer hear clicks from the
speaker when the toggle switch was actuated, and Ss no
i
longer responded at the actuation of the toggle switch.
The random click generator was eliminated.
i
When tested again, none of the Ss was able to j
discriminate the presence and absence of the stimulus, ,
Experimental predictions were revised? the two S^s with \
sacral surgery would be unable to learn the discrimination;!
j
whereas, the S, with thoracicolumbar surgery and the non- '
operated S > would not differ in ability and would learn the ;
discrimination.
105
Discrimination training began again after elimin
ating exteroceptive sources of stimulation. The three
operated S began with 10 cc. of interoceptive stimulus,
and Donna (non-operated) began with 20 cc. The data from
the remainder of the study are reported in Chapter III.
Discussion
Some Ss may have behaved differently in the final
results if this study had been carried out as it was
intended. The time necessary to establish the discrim
ination may have been prolonged, since responses to the
external stimuli may have to have been extinguished before
Ss could learn to discriminate the internal stimulus.
However, the posibility exists that Donna never responded
to the exteroceptive stimuli, since she never reached
criterion prior to the changes made in the apparatus, so
that the time for her eventual discrimination of the
interoceptive stimulus may not have been prolonged, Donna
and Henrietta exceeded the criterion at approximately the
same time. It is important that the research outlined in
this appendix be done, so there will be no question about
the influence of prior conditioning upon the results
obtained in the experiment and presented in Chapter III.
The apparatus controls which existed for the remainder of
the experiment still allow the conclusions drawn in
Chapter IV to be reached.
106 |
Experimental control is more critical in unknown j
or pioneering areas of research, as in this study, and
must be more extreme than in areas which are established.
It is vital that changes in interoceptive stimuli be
uncorrelated with exteroceptive changes.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Visual Evoked Potentials, Laterality Of Eye Movements, And The Asymmetry Of Brain Functions

PDF

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

PDF

Electrophysiological Properties Of Reaction Time And Aging

PDF

Human Operant Eye Blink Conditioning, Awareness, And The Extraversion-Introversion Dimension Of Personality

PDF

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

PDF

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

PDF

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

PDF

The Conditioning Of A Discriminative Stimulus Measured As An Orienting Reaction In Profoundly Retarded Blind Children

PDF

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

PDF

Biofeedback Control Of The Eeg Alpha Rhythm And Its Effect On Reaction Time In The Young And Old

PDF

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

PDF

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

PDF

Simple Auditory Reaction Time Of Young Adult And Elderly Subjects In Relation To Perceptual Deprivation And Signal-On Versus Signal-Off Conditions

PDF

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

PDF

Cue-Dependent Amnesia

PDF

Ucr Diminution As A Function Of Isi

PDF

The Effects Of Two Types Of Experimenter Intervention And Schedules Of Reinforcement On Verbal Operant Conditioning Of Affective Self-References

PDF

The Influence Of The Achievement Motive, The Affiliation Motive, And Incentive Conditions On Roleplaying Ability In Children

PDF

Recognition Accuracy In The Method Of Single Stimuli: A Test Of An Operational Definition Of The Distinctiveness Of Stimuli

PDF

The Effects Of Anxiety And Threat On Self-Disclosure

Asset Metadata

Creator Soldoff, Susan (author)

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

Degree Doctor of Philosophy

Degree Program Psychology

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

Tag OAI-PMH Harvest,psychology, experimental

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Galbraith, Gary C. (committee chair), Dunn, Arnold S. (committee member), Slucki, Henry (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-533034

Unique identifier UC11362265

Identifier 7127958.pdf (filename),usctheses-c18-533034 (legacy record id)

Legacy Identifier 7127958.pdf

Dmrecord 533034

Document Type Dissertation

Rights Soldoff, Susan

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