University of Southern California Dissertations and Theses

Hedonic aspects of conditioned taste aversion in rats and humans

(USC Thesis Other)

Hedonic aspects of conditioned taste aversion in rats and humans

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content HEDONIC ASPECTS OF CONDITIONED TASTE AVERSION IN RATS AND
HUMANS
Copyright 2001
by
Richard Daniel Kunz
A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTERS OF ARTS
PSYCHOLOGY
May 2001
Richard Daniql Kunz
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 1406455
___ ®
UMI
UMI Microform 1406455
Copyright 2001 by Bell & Howell Information and Learning Company.
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
Bell & Howell Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, Ml 48106-1346
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY OF SOUTHERN CALIFORNIA
The G raduate School
University Park
LOS ANGELES, CALIFORNIA 900894695
This thesis, w ritten b y
R l C.HA$S> pAU iZL- K U M2______........
Under the direction o f h..IS.. Thesis
C om m ittee, and approved b y a ll its m em bers,
has been presented to and accepted b y The
Graduate School, in p artial fu lfillm en t o f
requirem ents fo r th e degree o f
MAST&d O f AfcTS (p^VctJOLOUy)_____
Dean o f Graduate Studies
M av 1 1 . 2001
THESIS CO M M ITTEE
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
List of tables iv
List of figures v
Abstract viii
1. Introduction 1
2. Estrogen and the conditioned taste aversion paradigm 2
2.1. Conditioned taste aversion 2
2.1.1. The taste reactivity test 4
2.1.2. Neural substrates of conditioned taste aversion 6
2.1.3. Aversion versus avoidance 9
2.2. Estrogen as a conditioned Illness agent 13
2.3. Estrogen as a conditioned satiety agent 17
2.3.1. Estrogen and cholecystokinin 19
2.3.2. Estrogen and other hormones that suppress or stimulate food
intake ■ 24
2.3.3. Estrogen and the opioid system 26
2.3.4. Estrogen and the reward system 27
3. The effects of estradiol implants and injections on orofacial responses in the
rat 29
3.1. General methods 29
3.1.1. Subjects 29
3.1.2. Drugs 30
3.1.3. Intraoral cannulae 31
3.1.4. Surgery 31
3.1.5. Testing apparatus 33
3.1.6. Intraoral infusion procedure 33
3.1.7. Conditioned taste reactivity and taste avoidance procedure 35
3.1.8. Scoring of taste reactivity data 37
3.1.9. Statistical analyses 39
3.2. Experiment 1 41
3.2.1. Hypotheses 41
3.2.2. Methods 41
3.2.3. Results 42
3.2.4. Discussion 47
3.3. Experiment 2 49
ii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
HEDONIC ASPECTS OF CONDITIONED TASTE AVERSION IN RATS AND
HUMANS
Copyright 2001
by
Richard Daniel Kunz
A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTERS OF ARTS
PSYCHOLOGY
May 2001
Richard Daniel Kunz
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
List of tables iv
List of figures v
Abstract viii
1. Introduction 1
2. Estrogen and the conditioned taste aversion paradigm 2
2.1. Conditioned taste aversion 2
2.1.1. The taste reactivity test 4
2.1.2. Neural substrates of conditioned taste aversion 6
2.1.3. Aversion versus avoidance 9
2.2. Estrogen as a conditioned illness agent 13
2.3. Estrogen as a conditioned satiety agent 17
2.3.1. Estrogen and cholecystokinin 19
2.3.2. Estrogen and other hormones that suppress or stimulate food
intake 24
2.3.3. Estrogen and the opioid system 26
2.3.4. Estrogen and the reward system 27
3. The effects of estradiol implants and injections on orofacial responses in the
rat 29
3.1. General methods 29
3.1.1. Subjects 29
3.1.2. Drugs 30
3.1.3. Intraoral cannulae 31
3.1.4. Surgery 31
3.1.5. Testing apparatus 33
3.1.6. Intraoral infusion procedure 33
3.1.7. Conditioned taste reactivity and taste avoidance procedure 35
3.1.8. Scoring of taste reactivity data 3 7
3.1.9. Statistical analyses 39
3.2. Experiment 1 41
3.2.1. Hypotheses 41
3.2.2. Methods 41
3.2.3. Results 42
3.2.4. Discussion 47
3.3. Experiment 2 49
ii
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3.3.1. Hypotheses 49
3.3.2. Methods 50
3.3.3. Results 49
3.3.4. Discussion 57
3.4. General discussion 59
4. Human food aversions and the articulated thoughts in simulated situations
paradigm 61
4.1. Human food aversions 61
4.1.1. Taste hedonics 61
4.1.2. Articulated thoughts in simulated situations 66
4.1.3. Hypotheses 69
4.2. Methods 70
4.2.1. Subjects 70
4.2.2. Materials 70
4.2.3. Procedure 72
4.2.4. Scoring of ATSS Data 75
4.2.5. Statistical Analyses 78
4.3. Results 79
4.3.1. Reliability 79
4.3.2. Articulated thoughts 79
4.4. Discussion 86
5. Directions for future research 88
5.1. Animal studies 88
5.2. Human studies 91
References 93
Appendices 105
Appendix A: Food aversions and food preferences questionnaire 105
Appendix B: Post-experimental questionnaire 107
Appendix C: Informed consent form 109
Appendix D: CTA-ATSS instructions and scenario transcripts 111
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF TABLES
Table 1. Orofacial response types, scoring criterion, and duration.
Table 2. Results of 2-factor (groups x tests) ANOVAs with repeated
measures on tests for ME vs. FE, MC vs. FC, and E vs. C group
mean orofacial responses across the four taste reactivity tests in
experiment 1.
Table 3. Results of independent t-tests for E and C group mean orofacial
responses on ACQ1 and PA1 in Experiment 1.
Table 4. Results of 2-factor (groups X tests) ANOVAs with repeated
measures on tests for ME vs. FE, MC vs. FC, and E vs. C group
mean orofacial responses across the four taste reactivity tests in
experiment 2.
Table 5. Results of independent t-tests for E and C group mean orofacial
responses on ACQ1 and PA1 in experiment 2.
Table 6. Results of 1-Factor (Tests) ANOVA group by group comparisons
for mean consumption of 10% sucrose solution on PA2, PA3 and
PA4 in experiment 2.
Table 7. Means and standard errors of articulated thought categories for each
condition.
Table 8. Results of 2-factor- and 1-factor-ANOVAs of articulated thought
categories across conditions.
Table 9. Results of Newman-Keuls pairwise comparisons of articulated
thought categories by condition.
40
43
43
52
52
55
80
80
81
iv
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
LIST OF FIGURES
Figure 1.
Figure 2.
Figure 3.
Figure 4.
Figure 5.
Figure 6.
Figure 7.
Mean (+/- SE) GAPE responses among the E and C group animals
in response to intraoral infusions of 10% sucrose solution across
the four tests. The frequency of GAPEs changed differently across
the 4 tests for E and C group animals (p=.002). The E group
animals displayed more GAPEs than the C group animals on PA1
test (p=.017).
Mean (+/- SE) ingestive responses among the E and C group
animals in response to intraoral infusions of 10% sucrose solution
across the four tests. The frequency of ingestive responses did not
change differently across the 4 tests for E and C group animals
(p=.17). However, the E group animals displayed fewer ingestive
responses on PA1 than the C group animals (p=.007).
Mean (+/- SE) aversive responses among the E and C group
animals in response to intraoral infusions of 10% sucrose solution
across the four tests. No significant differences were found.
Mean (+/- SE) non-ingestive responses among the E and C group
animals in response to intraoral infusions of 10% sucrose solution
across the four tests. No significant differences were found.
Mean (+/- SE) MM responses among the E and C group animals in
response to intraoral infusions of 10% sucrose solution across the
four tests. No significant differences were found.
Mean (+/- SE) ingestive responses among the E and C group animals
in response to intraoral infusions of 10% sucrose solution across the
four tests. The frequency of ingestive responses changed differently
across the 4 tests for E and C group animals (p=.002). The E group
animals displayed fewer responses than C group animals on ACQ2,
ACQ3 andPA1 (p=.00,p=.02andp=00,respectively).
Mean (+/- SE) aversive responses among the E and C group animals
in response to intraoral infusions of 10% sucrose solution across the
four tests. No significant differences were found.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 8. Mean (+/- SE) non-ingestive responses among the E and C group
animals in response to intraoral infusions of 10% sucrose solution
across the four tests. No significant differences were found. 54
Figure 9. Mean (+/- SE) MM responses among the E and C group animals in
response to intraoral infusions of 10% sucrose solution across the
four tests. The frequency of mouth movements changed differently
across the 4 tests for E and C group animals (p=.002). The E group
animals displayed more mouth movements than C group animals on
ACQ2, ACQ3 and PA1 (p=.00, p=00 and p=.04, respectively). 54
Figure 10, Mean (+/- SE) sucrose consumption for ME, FE, MB and FB
groups on PA2, PA3, and PA4. The FE group consumed less of
the sucrose solution than either the ME or control groups on all
tests. 55
Figure 11. Mean (+/- SE) ESC responses by condition. A, D, and P indicate
a significance difference from the aversive, disliked or preferred
conditions, respectively. There were virtually no ESC responses
during the preferred condition. 82
Figure 12. Mean (+/- SE) NE responses by condition. A, D, and P indicate
a significance difference from the aversive, disliked or preferred
conditions, respectively. 82
Figure 13. Mean (+/- SE) DE responses by condition. A, D, and P indicate
a significance difference from the aversive, disliked or preferred
conditions, respectively. There were virtually no DE responses
during the preferred conditions. 83
Figure 14. Mean (+/- SE) VD responses by condition. A, D, and P indicate
a significance difference from the aversive, disliked or preferred
conditions, respectively. The disliked condition did not differ
significantly from the other conditions. 83
Figure 15. Mean (+/- SE) PE responses by condition. A, D, and P indicate
a significance difference from the aversive, disliked or preferred
conditions, respectively. There were virtually no PE responses
during the aversive and disliked conditions. 84
vi
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Figure 16. Mean (+/- SE) PLE responses by condition. A, D, and P indicate
a significance difference from the aversive, disliked or preferred
conditions, respectively. There were virtually no PLE responses
during the aversive and disliked conditions.
Figure 17. Mean (+/- SE) NAU responses by condition. A, D, and P indicate
a significance difference from the aversive, disliked or preferred
conditions, respectively. There were virtually no NAU responses
during the preferred condition.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
INTRODUCTION
In recent years, the ability to distinguish experimentally between aversion and
avoidance has become increasingly important in the field of conditioned taste
aversion research. While conditioned taste aversion involves, by definition, a
fundamental negative shift in an organism’s hedonic evaluation of an illness-paired
food, much of the animal research in this field relies on food avoidance to infer this
hedonic shift. However, the existence of drugs that produce conditioned avoidance
without accompanying aversion has brought the appropriateness of such assumptions
into question and suggests that behavioral avoidance may not be sufficient to infer a
conditioned aversion. Many researchers currently believe that behavioral indices of
an organism’s hedonic evaluation of foods need to be employed in order to report
conditioned aversions convincingly.
Human studies of conditioned taste aversion face the opposite challenge. It is quite
simple to assess a human subject’s hedonic evaluation of foods that have previously
caused them to become ill through self-report (usually nausea). However, ethical
considerations make it exceedingly difficult to study the behavioral avoidance of
illness-paired foods among humans in a controlled experimental environment.
Modem brain imaging techniques, such as fMRI technology, provide tremendous
potential for increased understanding of the neural basis of conditioned taste aversion
and avoidance among humans if a suitable method of stimulus presentation can be
developed. The task remains for the field of human taste aversion research to devise
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
an ethical, reliable and controlled experimental situation in which to employ brain
imaging tools.
This thesis describes three experiments that address the issues described above.
Experiments 1 and 2 are described in chapters 2 and 3 and concern the role of
estrogen in conditioned reductions in food intake. While estrogen’s anorexic effects
are well known, it is unclear whether it produces reductions in food intake through
aversion or avoidance mechanisms. Experiment 3 is described in chapter 4 and
examines a novel application of the Articulated Thoughts in Simulated Situations
cognitive assessment technique to the study of human food aversions. Chapter 5
discusses the implications of the experiments described in this thesis for future
inquiries.
2. ESTROGEN AND THE CONDITIONED FOOD AVERSION
PARADIGM
2.1. Conditioned taste aversion
Conditioned food aversions (or conditioned taste aversions, CTA) are aversions to
specific foods or taste stimuli that result from the association of these tastes with
aversive internal states such as gastrointestinal malaise or nausea (Garcia, 1955).
CTAs are acquired through a process similar to that of the classical conditioning
paradigm with the major distinctions being that CTA can require as few as one
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
pairing of the taste stimulus and illness (the CS and US, respectively) to produce a
strong and enduring avoidance of the taste stimulus (CR) and that the CS and US
may be separated for up to several hours (Chambers, 1990; Garcia, Hankins and
Rusiniak, 1974). CTAs thus represent a powerful and effective form of learning,
which are common to many animals including humans (Bernstein, 1985; Gustavson
and Gustavson, 1985; Gustavson, Gustavson, Young, Pumariega and Nicolaus,
1989).
The efficiency of the CTA learning system can be best understood in light of its
adaptive significance for the organism. Gustavson et al. (1989) point out that once
an animal has ingested a toxic substance there are few behavioral responses available
to the animal for alleviating the physiological effects of the substance. Regurgitation
and diarrhea are the most common mechanisms employed by animals to achieve the
expulsion of an ingested toxin, but an illness-producing or even fatal amount of the
substance may inevitably be absorbed. When an animal survives a food-induced
illness it behooves the animal to learn quickly and unequivocally that the food
produces illness and must be avoided.
The above hypothesis accounts for the ability of animals to associate the CS and
US across relatively large temporal spaces. Time is required for the gastrointestinal
system to absorb food and for homeostatic feedback mechanisms in the gut to relay
information about the results of ingestion to the appropriate neural areas. In order
for a CTA to develop, the gastrointestinal illness may be a result of the ingested food
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
or merely coincidental, as when one blames an ingested food for flu symptoms
(Chambers and Bernstein, 1995). Garb and Stunkard (1974) report that humans and
many animals show a strong tendency to associate any illness with an ingested food.
2.1.1. The taste reactivity test
Grill and Berridge (1985) define the palatability decision as an animal’s decision to
eat an encountered food. They propose three types of information that are integrated
by the central nervous system (CNS) as the principle determinants in this decision:
taste signals, internal-state signals, and cues from previous associations with the
taste. CTA involves primarily this third piece of information. If an animal has
previously become ill following ingestion of a particular food it will avoid that food
in the future and its palatability evaluation of the food will reflect this change. The
authors point out that palatability, by its definition, is a response measure rather than
a stimulus measure. A food that may be innately palatable to an animal, such as
sucrose, may cease to be palatable after it has been associated with gastrointestinal
illness (Chambers and Bernstein, 1995). The qualities of the sucrose stimulus that
cause it to be innately palatable have not changed but the animal’s evaluation of its
palatability has. CTA thus involves a hedonic shift in the animal’s palatability
evaluation of the taste (Grill and Berridge, 1985).
As a tool for the measurement of the rat’s hedonic evaluation of taste stimuli, Grill
and Norgren (1978) developed the taste reactivity test. They found that rats respond
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to intraoral infusions of taste stimuli with characteristic or stereotyped fixed action
patterns (FAPs) and that these responses directly reflect the CNS evaluation of taste
stimuli. They identified several distinct FAPs that may be grouped into three types
of responses, ingestive, aversive, and non-ingestive. Ingestive responses consist of
rhythmic mouth movements, tongue protrusions, lateral tongue protrusions and paw
licking. Aversive responses are composed of gapes, head shakes, chin rubs, forelimb
flailing and face washing. The non-ingestive response is passive dripping of fluid
from the mouth (Grill and Berridge, 1985). Ingestive FAPs are believed to facilitate
movement of a substance into the oral cavity while aversive FAPS are believed to
dispel a substance (Chambers and Bernstein, in press).
Grill and Norgren (1978) found that rats produce characteristic response profiles to
intraoral infusions of innately satisfying or aversive tastes. For example, intraoral
infusion of a sucrose solution (which is innately palatable) increases the number of
tongue protrusions and paw licks displayed while an intraoral infusion of quinine
(which is innately unpalatable) increases the number of head shakes and chin rubs
displayed (Grill and Norgren, 1978; Grill and Berridge, 1985). Furthermore, these
innate responses are altered following pairing of a taste with some internal state.
Following pairing of a sucrose solution with LiCl-induced illness, rats shift from an
ingestive response profile to an aversive response profile (Spector, Breslin and Grill,
1988). The opposite shift in palatability is also possible. Berridge, Flynn, Schulkin
and Grill (1984) demonstrated that the palatability evaluation of rats in response to
5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
intraoral infusions of hypertonic salt solutions (which are innately unpalatable)
shifted from aversive to ingestive during sodium depletion.
2.1.2. Neural substrates o f conditioned taste aversion
The majority of investigations into the neural substrates of conditioned taste
aversion have employed LiCl as the US and, because of its various physiological and
behavioral effects, it is regarded as the putative illness-inducing agent (Chambers
and Bernstein, in press). Consequently, a discussion of the neural mechanism by
which LiCl exerts its illness effects is useful, as other drugs are generally compared
to LiCl in an effort to establish them as illness agents (Chambers and Bernstein, in
press).
Among the neural substrates of LiCl-induced food avoidance in rats, the
parabrachial nucleus in the caudal pons, appears to be a key site (Yamamoto,
Shimura, Sako, Yasoshimaand Sakai, 1994a). First order visceral and gustatory
afferents project to separate areas of the nucleus of the solitary tract (NST), which in
turn project efferent axons to a confined region of the PBN (Hermann, Kohlerman
and Rogers, 1983). The convergence of gustatory and visceral information in the
PBN makes it a prime candidate for the CS/US association involved in the
development of conditioned taste avoidance. It has been found that ibotenic acid
lesions of the PBN, made before the pairing of saccharin and LiCl inhibit formation
of conditioned food avoidance (Yamamoto et al., 1994a). In addition, Wang,
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Lavond and Chambers (1996) have found that reversible cooling of the lateral PBN
only during administration of the US prevents acquisition of LiCl-induced food
avoidance in rats.
Yamamoto, Shimura, Sakai and Ozaki (1994b) have used c-fos
immunohistochemical methods to identify two specific subnuclei of the PBN that are
associated with positive and negative hedonic reactions to intraorally infused taste
stimuli in the rat. The dorsal lateral subnucleus of the PBN appears to be associated
with ingestive behavior while the external lateral subnucleus of the PBN appears to
be associated with aversive behavior. C-fos immunoreactivity has been induced in
the dorsal lateral and external lateral subnuclei in response to administrations of
sucrose and LiCl, respectively (Ogawa, Hayama and Ito, 1987; Swank and Bernstein,
1994; Yamamoto et al., 1994ab). Following a sucrose-illness pairing, the sucrose
stimulus will induce c-fos expression in the external lateral subnucleus (Yamamoto
et al., 1993). These data strongly suggest that the LiCl-flavor pairing may take place
in the PBN, at least for sucrose. Furthermore, Grill and Berridge (1985) have shown
the PBN to be the level at which innate responses to taste stimuli are controlled.
They found that supracollicular decerebrate preparations (leaving only the NST and
PBN) were capable of the same innate orofacial responses to taste stimuli as
controls. However it is unclear whether the PBN generates or maintains orofacial
responses to taste (Travers and Norgren, 1983).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In addition to its connections with the NST, the PBN also receives projections from
the area postrema (Shapiro and Miselis, 1985). The area postrema (AP) is a
circumventricular organ positioned on the dorsal surface of the medulla oblongata on
the floor of the fourth ventricle. Due to its lack of a blood-brain barrier and cerebral
spinal fluid (CSF)-brain barrier the AP operates as a chemoreceptor in the detection
of blood-borne and CSF-bome chemicals (Borison, 1989). The AP plays an
important role in the acquisition and retention of food avoidance induced by LiCl
administration. Adachi, Kobashi, Miyoshi and Tsukamoto (1991) report that the AP
contains neurons that respond to LiCl administration. C-fos immunoreactivity is
evident in the AP following paired or unpaired administrations of LiCl (Chambers,
Hayes, Kunz and Wang, 1999; Swank, Schafe and Bernstein, 1995).
It appears that LiCl effects are mediated by information arriving at the AP through
the blood stream and at the NST through visceral afferents, though only the AP is
necessary for development of conditioned food avoidance (Chambers and Bernstein,
in press). Intraperitoneal administration of LiCl stimulates both the splanchnic and
vagus nerves (Niijima and Yamamoto, 1994). Martin, Cheng and Novin (1978)
report that LiCl can produce CTA in the rat following subdiaphragmatic vagotomy
but reversible lesions of the AP (which do not include the NST) inhibit the formation
of a LiCl-induced food avoidance in rats (Wang, Lavond and Chambers, 1997a,b).
Permanent lesions of the AP prevent acquisition of a conditioned taste avoidance and
prevent the negative hedonic shift, as measured by taste reactivity, that are associated
8
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
with LiCl administration (Eckel and Ossenkopp, 1996; Ritter, McGlone and'Kelley,
1980). So for LiCl, the AP appears to be a critical part of the illness pathway.
The LiCl pathway is less well characterized above the level of the brain stem.
Projections from the PBN are believed to extend to the basolateral amygdala through
the gustatory insular cortex or the zona incerta and intralaminar thalamic complex
(Sakai and Yamamoto, 1999). Discrete lesions of each of these areas do not affect
LiCl-induced taste avoidance, but combined lesions do prevent LiCl-induced taste
avoidance (Sakai and Yamamoto, 1999). Evidence concerning the role of the
basolateral amygdala in LiCl-induced effects has been contradictory. Gu, Gonzalez,
Chen and Deutsch (1993) report that administration of LiCl induces c-fos
immunoreactivity in the basolateral amygdala. Dunn and Everitt (1988) and Morris,
Frey, Kasambir and Petrides (1999) have both subjected the basolateral amygdala to
excitotoxic lesions with opposite results. Dunn and Everitt (1988) found that
excitoxic lesion of the basolateral amygdala had no effect on conditioned taste
avoidance induced by LiCl while Morris et al. (1999) found that it did. However, as
Chambers and Bernstein (in press) point out, the Morris et al. (1999) study was
further supported with neural tract tracing and is thus more convincing.
2.1.3. Aversion versus avoidance
Among the more common measures used to evaluate conditioned taste aversion in
rats have been the one-bottle consumption test and the two-bottle preference test
9
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
(Grill and Norgren, 1978). A potential confound with these measures is that they do
not actually address the aversive internal states associated with illness that were
described by Garcia (1955). That is, we cannot know from a bottle test what the
animal’s hedonic evaluation of a tastant is (Grill and Berridge, 1985). Aversion is
inferred from the animal’s failure to consume a solution that has been previously
paired with illness.
Although avoidance is often inferred to be a negative hedonic reaction, Parker
(1995) has shown that rewarding drugs (e.g., amphetamine and morphine) produce
an avoidance of a paired sucrose solution in the bottle test but no aversion, as
measured by the taste reactivity test. Using this paradoxical finding that behaviorally
reinforcing drugs also induce conditioned avoidance as an example, Grigson (1997)
suggests that it may be that the CS is less salient than, or outweighed by, the
reinforcing properties of the drug. Hunt and Amit (1987) have come to the same
conclusion based on their review of the literature.
The ideas of these authors are interesting because they suggest the existence of a
positive continuum where the hedonic issue is one of relative positive hedonic
evaluation rather than the standard conceptualization of the food avoidance process
as exclusively positive or negative. An animal may rate one effect of a substance as
less positive than another without involvement of the negative hedonic axis.
A caveat is in order regarding the combined use of taste reactivity and bottle tests
to examine the issue of avoidance vs. aversion. One measure may confound the
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
other. That is, the animal’s behavior in the bottle test may be influenced by the fact
that it has been forcefully exposed (through intraoral application) to the tastant prior
to the bottle test. This is a serious issue as the animal’s natural motivational systems
are being usurped and we are unaware of the subsequent behavioral consequences of
this. Moreover, it has been suggested that bottle and taste reactivity tests may be
“apples and oranges,” as they are dealing with appetitive (or approach) and
consummatory behaviors, which reflect instrumental and Pavlovian conditioning,
respectively, and may proceed by different neural mechanisms (Chambers and
Bernstein, in press; Spector et al., 1988).
Booth (1977) has offered another hypothesis to account for food avoidance. He
has suggested that food avoidance may be the CR to internal states other than illness,
and has shown that satiety can serve as the US in a conditioned food avoidance
paradigm (Booth, 1985). Moreover, he has shown that under conditions of satiety,
rats will even display many of the behavioral reactions that have traditionally been
associated with illness, such as “lying-on-belly” (Booth, 1985). Grill and Berridge
(1985) also found that ingestive orofacial responses decreased and aversive orofacial
responses increased as animals became sated during a meal. It should be noted that
based on these findings it is clear that aversive orofacial reactions alone are not
sufficient to infer aversion.
Another issue surrounding the practice of inferring aversion from avoidance is that
avoidance behavior is the output of lattice hierarchies in the CNS (Chambers and
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Bernstein, 2001). A lattice hierarchy is defined as a behavioral organization in
which lower components can be invoked by different higher-level systems at
different times, depending on the organism’s goals (Mook, 1996). This concept is
consistent with current evolutionary theory in which it would be advantageous to
have a flexibility of behavior that allows a specific behavior to serve several
functions (Mook, 1996). That is to say, an animal might avoid consuming a solution
for many different reasons. For example, the animal may associate the solution with
gastrointestinal discomfort, relative reward, or may simply not be thirsty.
Possibly the most serious issue in the current conceptualization of CTA, at least as
it is operationally defined, has been our inability to definitively identify the
occurrence of nausea in an experimental animal (Grant, 1987). Garcia’s (1955)
definition of CTA (described in 2.1. Conditioned food aversion, above) requires the
incidence of nausea for proper identification as such. However, as Grant (1987)
notes, Garcia relies heavily on the Borison and Wang (1953) model of emesis which
describes an “emetic syndrome” consisting of nausea, retching and emetic
symptoms. The problem is that while the AP has been demonstrated, behaviorally,
to be the emetic trigger zone (Wang and Chinn, 1954; Brizzee, Ordy and Mehler,
1980), virtually nothing is known about the neural substrates of nausea or retching.
To date, we believe that no one has definitively demonstrated nausea in an
experimental animal. Nausea is simply assumed to accompany emesis and an
instance of vomiting thus behaviorally indicates nausea and aversion. However, it is
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
also known that nausea and emesis can occur independently of each other (Grant,
1987). This poses an especially intractable problem for research in the CTA field
since this research is done almost exclusively on the rat, which does not vomit
(Grant, 1987).
In many cases it may be reasonable to assume that a toxin produces the same
aversive internal states in the rat as it does in humans (which can be measured by
self-report). Indeed, CTA in humans shares the same essential features that have
been characterized for non-human species (Chambers and Bernstein, 1995; Garb and
Stunkard, 1974; Logue, Logue and Strauss, 1983; Logue, Ophir and Strauss, 1981).
It has been shown, for example, that LiCl produces nausea in humans (Boland,
Mellor and Revusky, 1978). Parker and Macleod (1991) have suggested that the
chin rub is a specific symptom of nausea in the rat. These authors found that an
antiemetic agent administered prior to the LiCl-sucrose pairing markedly reduces the
incidence of chin rubs during taste reactivity testing. These results are not
conclusive, however, since even if chin rubs are a nausea-induced response it is
unlikely that they serve only the nausea circuitry of the brain (Chambers and
Bernstein, in press).
2.2. Estrogen as a conditioned illness agent
The ability of estrogen to produce conditioned taste avoidance in rats has been
well-documented (Chambers and Bernstein, 1995). Bernstein et al. (1986) found
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
that chronic infusions of B-estradiol produced conditioned taste avoidance in rats.
Subcutaneous implantation of B-estradiol pellets and injections of E2-CDS (estradiol
chemical delivery system) have been shown to produce conditioned taste avoidance
in rats (Mordes, Longscope, Flatt, Maclean and Rossini, 1984; Ganesan and
Simpkins, 1990; Ganesan and Simpkins, 1991). Subcutaneous implantation of
30mm B-estradiol implants for one hour produces conditioned taste avoidance in rats
(Chambers, Hayes, Kunz and Wang, 1998; Kunz, 1997). All of these experiments
employed bottle tests and thus measured only the effects of estrogen on avoidance
(as described in 2.1.3. Aversion versus avoidance above).
As described above (2.1.2. Neural substrates o f conditioned food aversions) many
investigations have compared the effects of target stimuli to LiCl (as the putative
illness agent) in order to establish them as illness agents. The illness-producing
properties of estradiol are supported by work, which has shown that there is
crossfamiliarization between estradiol and LiCl. Preexposure to estradiol before
acquisition of a LiCl-induced avoidance attenuates acquisition and preexposure to
LiCl before acquisition of an estradiol-induced avoidance has a similar effect
(Chambers et al., 1998,1999; Chambers and Yuan, 1999a,b).
While it is clear that estrogen produces conditioned taste avoidance and interferes
with the effects of LiCl in rats, the evidence concerning its mechanism of action is
confusing and contradictory. First, there is evidence that estradiol acts on the AP.
The AP is known to contain chemoreceptors specific to estradiol (Simerly, Chang,
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Muramatsu and Swanson, 1990; Adachi et al., 1991) and that the AP acts as a
chemoreceptor trigger site for emesis (Borison, 1989). Bernstein et al. (1986) report
that permanent lesions of the AP result in attenuation of taste avoidance induced by
Leydig LTW(m) cell tumor syndrome, and Mordes et al. (1984) report that Ley dig
LTW(m) tumor-induced taste avoidance is associated with increased levels of
plasma estradiol. However, it has also been demonstrated that reversible cooling
lesions of the AP fail to attenuate estradiol-induced taste avoidance and that estradiol
does not increase c-fos expression in the AP (Chambers, Makena, Kunz and Wang,
2000). The conflicting evidence surrounding the AP may be explained by the fact
that mass lesions of the AP frequently include part of the underlying NST while
cooling lesions do not (Chambers and Bernstein, in press). This suggests that it may
actually be the NST that mediates the effects of estradiol on conditioned taste
avoidance. Referring back to Neural substrates o f conditioned taste aversion (2.1.2),
cooling of the AP alone has opposite effects on the acquisition of LiCl- and
estradiol-induced avoidance.
Taste reactivity evaluations of the behavioral effects of estradiol in rats have
produced contradictory results. Ossenkopp, Rabi and Eckel (1996) reported that the
pairing of a novel sucrose solution and estradiol produced a subsequent avoidance of
the sucrose solution (in a two-bottle preference test) and a robust negative shift in
palatability as measured by the taste reactivity test. King, Grigson, Grill and
Flanagan-Cato (1997), on the other hand, paired a novel sucrose solution with an
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
injection of estradiol benzoate and found that rats subsequently avoided the sucrose
solution in a one-bottle test. However, these animals displayed ingestive rather than
aversive responses during taste reactivity testing, suggesting that estradiol produces
conditioned taste avoidance as opposed to a conditioned taste aversion.
Similar estradiol effects on taste reactivity have been found in human populations.
Weizenbaum, Benson, Solomon and Brehony (1980) have reported that the phase of
the menstrual cycle does not affect pleasantness ratings of a sucrose solution, but that
meal size is increased during the luteal phase, at least for short-menses females.
These results are compelling since they involve normal physiologic levels of
estradiol, and they seem to support the hypothesis that it may be acting through a
satiety, rather than an aversion, mechanism.
Estradiol is known to produce nausea in humans, such as male cancer patients
undergoing pre-surgical hormone therapy (Goodman and Gilman, 1975). As we
have seen however, the evidence for this same estrogen-induced internal state in rats
is less clear. While the Ossenkopp et al. (1996) findings suggest that estrogen exerts
its effects through a nausea mechanism, the King et al. (1997) finding requires an
alternative explanation. Their finding that the animals avoided an estrogen-paired
solution but did not show aversive taste reactivity may be accounted for by a couple
of different hypotheses. First, it could be that estrogen does not induce an internal
state strong enough to elicit the aversive taste reactivity seen with other toxins.
Work in our lab has shown that estrogen produces a mild avoidance relative to LiCl,
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
as measured with the bottle test. Second, estrogen may be acting through a pathway
tangentially related to the illness pathway characterized for LiCl and thus produces
only a subset of the behavioral constellation invoked by LiCl.
The second hypothesis seems more likely than the first. If estrogen is producing a
mild aversion via the illness pathway, then we should expect mildly aversive taste
reactivity responses. We also already have evidence that estrogen is probably not
using the same illness pathway as LiCl, at least in part, as it does not seem to require
the AP to exert its effects (Chambers et al., 2000).
2.3. Estrogen as a conditioned satiety agent
As we have seen, there is evidence that estrogen exerts its effects through a
pathway other than that characterized for LiCl. It has been suggested that estrogen
operates through a conditioned satiety mechanism to reduce food intake rather than a
conditioned aversion mechanism (Chambers and Bernstein, in press). Conditioned
satiety occurs when gustatory, olfactory, visual, tactile, etc. qualities of a food
become the CS for the unconditioned satiety effects of that food (Smith and Gibbs,
1979). Booth (1985) has conceptualized satiety as a conditioned aversion that is
distinguished by its dependence on the internal state effects of the ingested food.
Estrogen has been found to transiently decrease the intake of food and permanently
lower body weight by a long-term satiety mechanism. That is, it reduces intake by
causing early meal termination. This, combined with changes in body metabolism
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
contribute to a reduction in body weight and the maintenance of a lower body weight
set point (Blaustein and Wade, 1976; Drewettt, 1974; Wade, 1972; Wade and Gray,
1979; Wade and Schneider, 1992). Wade (1972) reports that ovariectomy in rats
causes a transient increase in meal size and weight gain, with the meal sizes
returning to pre-surgical levels but the gained weight remaining. The opposite is
also true. Administration of estradiol to ovariectomized rats causes a transient
decrease in meal size and body weight. With continued estradiol treatment, meal
size eventually returns to the pre-treatment level and the lower body weight set point
is maintained (Wade, 1972). Wade and Gray (1979) have suggested that the
permanent alterations in body weight may be due to regulation of enzymatic
mechanisms controlling the storage and breakdown of fats by estradiol.
If estrogen causes conditioned food avoidance through a satiety mechanism, it
must operate through the normal feeding system (as opposed to illness mechanisms).
The role of estrogen in this system is exceedingly complex and only poorly
understood at this time. In the following subsections (2.3.1 through 2.3.4), we
provide a broad examination of the possible neural areas and chemicals that estrogen
may interact with to reduce food consumption. We focus heavily on the interaction
of estrogen with cholecystokinin (CCK) as a putative short-term satiety agent. We
also explore the role of estrogen in the opiate and reward system as pertains to
feeding.
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.3.1. Estrogen and cholecystokinin
Cholecystokinin (CCK) is the most studied of the peripheral satiety factors and acts
both peripherally and centrally in a complicated manner (Fink, Rex, Voits and Voigt,
1998). Although the experimental effects of CCK on food consumption appear to be
heavily dependent on research methodology, it is generally accepted that a major role
of CCK is the inhibition of feeding (Fink et al., 1998). Peripheral injections of CCK
suppress food intake while injections of CCK antagonists increase food intake (Qian,
Johnson, Kallstrom, Carrer and Sodersten, 1997).
In order to be considered a satiety agent, CCK must not produce its inhibitory
effects on food intake through sickness (Deutsch and Hardy, 1977). However, there
is evidence that it produces inhibition of feeding through its illness-associated
properties (Deutsch and Hardy, 1977). It has been shown that CCK, administered in
high doses, produces abdominal cramps, nausea and emesis in humans and rats
(Strieker and Verbalis, 1991). Moore and Deutsch (1985) showed that
Trimethobenzamide (an antiemetic that acts specifically on the AP) attenuates the
inhibition of food intake caused by CCK. Bowers, Herzog, Stone and Dionne (1992)
report a defensive burying response among rats (considered to be a sign of aversion)
following pairing of a sweetened milk solution with exogenous CCK administration.
Melton and Riley (1994) have also reported conditioned taste avoidance in rats
following exogenous administration of CCK. Further evidence that CCK acts
through an illness mechanism comes from an experiment reported by Pelchat, Grill,
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Rozin and Jacobs (1983). The authors compared the taste reactivity of rats exposed
to LiCl (upper gastrointestinal tract discomfort), high doses of lactose (lower
gastrointestinal tract discomfort), and shock (peripheral pain). They found that only
the LiCl-treated group displayed aversive taste reactivity responses on subsequent
testing and suggested that this is due to the nausea-inducing effects of LiCl. Since
we know that CCK exerts its effects on the upper gastrointestinal tract, we might
expect to see aversive taste reactivity in animals following pairing of CCK with a
sucrose solution if it is operating through the same nausea mechanism.
However, the results of Cross-Mellor, Kent, Gssenkopp and Kavaliers (1999) do
not support this hypothesis. They found that CCK-treated animals significantly
reduced their ingestive responses, but did not significantly increase their aversive
responses, in the taste reactivity test. They suggest that the animals’ taste reactivity
responses to CCK are consistent with satiety. It should be noted that all of the
experiments showing conditioned taste avoidance involved exogenous administration
of CCK at doses that may not mimic its endogenous physiological effects. Smith
and Gibbs (1979) report that a dose of CCK capable of producing a 40-50%
suppression of food intake does not produce any of the behavioral indications of
toxicity. This provides additional evidence to support the hypothesis that much of
the data indicating that CCK causes CTA may not be indicative of its actual
physiological effects.
20
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
It is known that CCK receptors exist along the entire gut-brain pathway (Linden
Hirschberg, 1998) and this includes several nuclei previously identified as part of the
illness pathway. There are two types of CCK receptors: CCK-A and CCK-B. CCK-
A receptors are found in both the periphery and centrally, and mediate the satiety
effects of CCK (Melton and Riley, 1994). CCK-B receptors are found only in the
central nervous system and are involved in the production of anxiety-type behaviors
(Fink et al., 1998).
In response to a meal, CCK inhibits food intake through its action on the vagus
nerve by two mechanisms. Directly, through CCK-A receptors of the vagus and
indirectly, by causing gastric distention (which also activates the vagus). The vagus
fibers terminate in the NST, and the information is further processed by the PBN,
and ventromedial hypothalamus (VMH) or paraventricular nucleus (PVN)
(Geiselman, 1996; Linden Hirschberg, 1998). CCK receptors have been isolated in
the AP, mediocaudal level of the NST, VMH, PVN and medial preoptic area
(MPOA) of the rat (Holland, Norby and Micevych, 1998; Micevych, Eckersell,
Brecha and Holland, 1997; Qian et al., 1997; Tetel, Getzinger and Blaustein, 1994).
The VMH and PVN are known to be involved in satiety (Leibowitz, Weiss and Suh,
1990; Linden Hirschberg, 1998).
Estrogen has also been shown to be active in the VMH, (Tetel et al., 1994) and
MPOA (Conde, Herbison, Femandez-Galaz and Bicknell, 1996; Pfaus, Marcangione,
Smith, Manitt and Abillamaa, 1996), and is known to modulate the expression of
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CCK mRNA in the MPOA (Holland et al., 1998). Additionally, the number of
CCK-immunoreactive cells in the MPOA has been shown to fluctuate as a function
of estrous cycle variations, with the lowest levels of staining (comparable to that in
males) correlating with the highest levels of circulating estrogen (Oro, Simerly and
Swanson, 1988).
However, the more likely site for an interaction between estrogen and CCK is the
PVN (Butera and Beikirch, 1989). The authors conducted an experiment in which
estradiol was injected directly into the PVN, MPOA, VMH or posterior
hypothalamus (PH) and they found that increased estradiol levels in the PVN, but not
in the MPOA, VMH or PH, were associated with significantly lowered food intake.
Butera, Campbell and Bradway (1993) also showed that implants of Anisomycin
(which inhibits the protein synthesis believed to be activated by estrogen) in the PVN
of rats disrupted estrogen’s anorectic effects and caused the animals to gain weight.
Butera, Willard and Raymond (1992) found that lesions of the PVN prevented
estradiol from suppressing food intake in rats. However, Dagnault and Richard
(1997) found that injections of estradiol into the MPOA reduced food intake in a
dose-dependent manner and have concluded that it may be the MPOA which
mediates the anorectic effects of estradiol. One possible source of the contradiction
on this point is that Butera et al. (1989,1992,1993) and Dagnault and Richard
(1997) employ differing methodologies involving acute and chronic estradiol
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
treatments, respectively. As mentioned above, the effect of estradiol on eating is
transient. After chronic treatment, estradiol no longer has a suppressive effect.
Despite disagreement over the site of interaction between estradiol and CCK, it is
clear that the two interact and have an effect on food consumption. It has been
shown that cyclic administration of estradiol to ovariectomized rats enhances the
satiating effect of exogenously administered CCK (Geary, Trace, McEwen and
Smith, 1994). Dulawa and Vanderweele (1994) performed a similar experiment,
though with smaller doses of CCK. They found that estrogen levels characteristic of
late diestrous (when circulating estrogen levels are relatively high) significantly
suppressed food intake compared to the levels characteristic of early metestrous
(when circulating estrogen levels are low). In the same experiment, they also found
that estrogen cycling in intact female rats also affected food consumption, with
bilaterally intact females consuming less than unilaterally intact females and males in
response to low-dose CCK administration. Dulawa et al. (1994) point out that even
the dose they use is probably well above the normal physiological range for CCK.
Again, the issue of CCK dosage is important in light of the contradictory findings
regarding its effects on food intake. The effects of CCK on feeding are known to be
sensitive to experimental methodology (Deutsch et al., 1977), which may be due, in
part, to the fact that CCK probably arrives at the CNS through either peripheral
afferents or through the blood (Fink et al., 1998).
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.3.2. Estrogen and other hormones that suppress or stimulate food intake
Bombesin is known to have strong suppressive effects on food intake in several
species and has been shown to cause a defensive burying reaction in rats when paired
with a sweetened milk solution (Bowers et al., 1992; Geiselman, 1996). However, it
is also known that while bombesin decreases meal size, it does not affect the
initiation of feeding (Geiselman, 1996). It is presently unclear what, if any,
interaction bombesin has with estrogen.
Corticotropin-releasing factor (CRF) is a powerful anorectic and is widely
distributed in the brain, but the highest concentration of CRF-containing cells is in
the PVN and it is believed that this is its major site of production (Dagnault and
Richard, 1994; Linden Hirschberg, 1998). CRF is also known to promote expression
of proopiomelanocortin (POMC), a molecule which gives rise to p-endorphin
(Levine and Billington, 1997) although the exact nature of this relationship is
unclear. While the ability of CRF to suppress food intake has been established
(Linden Hirschberg, 1998) and there is known co-localization of estrogen and CRF
receptors in both the PVN and MPOA (Dagnault and Richard, 1994; 1997), how
these two hormones interact is presently unclear. It has been demonstrated that
administration of a CRF antagonist blocks the anorectic effects of estrogen and the
promoter region of the CRF gene contains an estrogen response segment (Dagnault,
1997). However, as we discussed previously (2.3.1. Estrogen and cholecystokinin),
evidence concerning the role of the PVN (and thus, the estrogen and CRF receptors
24
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
in this area) is conflicting (Butera et al., 1989; 1992; 1993; Dagnault and Richard,
1994; 1997). There is evidence that the MPOA is the site of interaction between
estrogen and CRF (Dagnault and Richard, 1997). However, these authors report that
injections of CRF and estrogen into this area produce different effects. Estrogen
decreases food consumption while CRF does not when injected into the MPOA.
They conclude that CRF exerts its anorectic effect elsewhere in the brain.
Leptin is a relatively newly discovered hormone and acts on various central nuclei
to reduce food intake, particularly fats (Linden Hirschberg, 1998). It is presently
unknown whether leptin has any interactions with estrogen, but it does interact with
systems that are known to involve estrogen. Linden Hirschberg (1998) reports that
leptin acts on neuropeptide-Y (NPY) cells in the arcuate nucleus to decrease the
production of NPY mRNA, and that it also increases the levels of CRF mRNA and
release in the hypothalamus. It seems likely, considering these common sites of
action, that estrogen and leptin might interact to reduce food intake.
Neuropeptide-Y is produced by cells in the arcuate nucleus and is the most
powerful orexigenic signal known (Bonavera, Dube, Kalra and Kalra, 1994). Levine
and Billington (1997) have suggested that NPY is involved in reward-driven feeding
and this may be the reason for its relatively powerful effects. Infusions of anti-NPY
serum into the PVN have been shown to prevent the feeding response induced by
fasting, and CRF is also known to block the feeding effects of NPY (Linden
Hirschberg, 1998). In addition to its activity in the PVN with regard to CCK and
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CRF, estrogen is known to affect the release of NPY in this area as well (Bonavera et
al., 1994). These authors report that non-cyclic physiologic levels of estradiol
reduced food intake in rats and selectively decreased NPY levels in the PVN.
2.3.3. Estrogen and the opioid system
The effect of estrogen on the opioid system of the brain is markedly complex and
only poorly understood at this time (Morley, 1981). However, it is generally
believed that opioids are involved in the intake of fatty and highly palatable foods
(Linden Hirschberg, 1998). Among the endogenous opioids, dynorphin and, to a
lesser extent, P-endorphin appear to play a key role in the initiation of feeding
behavior in the rat (Bonavera et al., 1994; Linden Hirschberg, 1998; Morley, Levine,
Grace, Kneip and Gosnell, 1984).
Morley et al. (1984) report that estrogen-treated rats are significantly more resistant
to the suppressive effects of naloxone (an opioid antagonist) on spontaneous feeding
behavior. Wagner, Manzanares, Moore and Lookingland (1994) reported that
estrogen suppresses the K-receptor-mediated inhibition of tuberoinfundibular
dopaminergic (TIDA) neurons, presumably by decreasing the release of dynorphin.
This is consistent with estrogen’s suspected suppressive effects on feeding through
interactions with CCK in the brain (described in 2.3.1. Estrogen and cholecystokinin
above) as dopamine (DA) is known to suppress food intake through its interactions
with CCK in the brainstem (Qian et al., 1997).
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The sexual behavior literature provides evidence that in addition to estrogen
modulating opioid function (Carter, Williams and Lightman, 1986) the reverse is
also true. Eckersell and Misevych (1997) report that opiates suppress the estrogen-
induced expression of CCK in the hypothalamus. Micevych et al. (1997) also found
that opiates suppressed the expression of CCK in the MPOA. Piva et al. (1995) have
shown that administration of estrogen increases the number of p-opioid receptors
(associated with (3-endorphin) in the thalamus. While these latter results seem to
contradict those of the previous paragraph, it should be noted that this experiment
addressed sexual behavior rather than eating. It is possible that the interaction of
estrogen with the opioid system with regard to eating and sexual behavior, although
occurring in the same central nuclei, proceeds by separate mechanisms (Butera and
Beikirch, 1989).
2.3.4. Estrogen and the Reward System
Dopamine is known to be involved in the feeding system at least at the level of the
NST where it interacts with CCK to produce a decrease in food consumption (Qian
et al., 1997). Exogenous administration of CCK increases DA levels in the NST and
administration of a DA receptor antagonist partially reverses the inhibitory effects of
CCK on ingestive behavior in the rat (Qian et al., 1997).
While it is unknown if estrogen has any interaction with the DA system at the level
of the brain stem, it is known to modulate both the nigrostriatal and mesolimbic DA
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
systems. In the nigrostriatal system, estrogen has been shown to both up- and down-
regulate dopaminergic function, depending on the time course of administration and
on the sex of the animal (Dorce and Palmero-Neto, 1992). hammers et al. (1999)
report an experiment in which chronic high-dose continuous estrogen treatment
produced down-regulation of the DA D2 receptor in the dorsal and ventral striatum.
Disshon and Dluzen (1997) have shown that estradiol enhances the increase of DA
released by cells of the striatum in response to MPP+ (a neurotoxin which decreases
DA levels) in a dose-dependent manner when administered before MMP+, but less
so when administered concurrently with MPP+. In the mesolimbic system, estrogen
alone has been found to have no effect on the rewarding qualities of brain
stimulation reward (BSR) targeted at the medial forebrain bundle, but estrogen and
progesterone increase the rewarding qualities of BSR in this area (Bless, McGinnis,
Mitchell, Hartwell and Mitchell, 1997).
Estrogen is also known to influence the DA system at higher brain levels such as
the prefrontal cortex (Kritzer and Kohama, 1999). While it is clear that estrogen has
important interactions with this system, it is much less clear what its effects on the
eating system above the brain stem are. Certainly, the mesolimbic system is a likely
candidate for an estrogen-DA interaction that is relevant to CTA since this system
contains abundant connections to the limbic regions that are important in the
motivational aspects of eating behavior (Mook, 1996).
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
3. THE EFFECT OF ESTRADIOL IMPLANTS AND INJECTIONS ON
OROFACIAL REPONSES IN THE MAT
The first two experiments in this thesis provide information that can contribute to
our understanding of whether estradiol produces conditioned taste avoidance via an
illness mechanism or a satiety mechanism. They were designed specifically to
determine whether estradiol produces a negative palatability shift after it has been
s »
paired with a sucrose solution.
3.1. General Methods
3.1.1. Subjects
The subjects were adult male and female Sprague-Dawley rats (Simonsen
Laboratories, Gilroy, CA) that were 90 days old and weighed approximately 300 and
200 g, respectively, upon arrival at the laboratory. They were housed in pairs in
solid bottom cages (58 x 38 cm), separated by stainless steel dividers, that had wood
chips as bedding material. The animals had access to food and water at all times in
the home cage and were kept on a 12:12 hour light/dark cycle throughout the
experiments.
In experiment 1, animals were provided with a “mashed and wetted” preparation of
their normal solid diet from surgery day through the remainder ofthe study. Each
animal was given approximately 200 g of this preparation each day. In experiment
2, the animals were provided with a custom-made, non-sweetened liquid diet (PMI
29
Reproduced with permission o fth e copyright owner. Further reproduction prohibited without permission.
Lab Diet, Richmond, IN) from two weeks prior to surgery day through the remainder
of the study. Males were given a 150ml/day ration while females were given
lOOml/day. The special preparations in experiments 1 and 2 were made daily and no
animals were observed to exceed their daily rations. These preparations were
provided in addition to the animals’ solid food and water in all experiments and
facilitated eating following placement of the intraoral cannulae.
3.1.2. Drugs
The estradiol implants used in experiment 1 were constructed of 34 mm lengths of
silastic tubing (1.6 mm inner diameter, 3.1 mm outer diameter, Dow Coming,
Auburn, MI) filled with 30 mm of crystalline estradiol (l,3,5(10)-estratrien-3,17beta-
diol, Steraloids, Wilton, NH), and sealed on either end with 2 mm of Silicone Type
A medical adhesive (Dow Coming, Auburn, MI). “Blank” implants were
constructed in the same manner but were empty.
The estradiol injections used in experiment 2 contained a 100 pg/0.05 ml
suspension of crystalline estradiol (1,3,5 (10)-estratrien-3,17 beta-diol, Steraloids,
Wilton, NH) in a sesame seed oil (Inspired Natural Foods, San Leandro, CA)
suspension. The dose administered was 500 pg/kg. “Blank” injections contained
only sesame seed oil.
30
Reproduced with permission o fth e copyright owner. Further reproduction prohibited without permission.
3.1.3. Intraoral cannulae
Intraoral cannulae were constructed of 3.8 cm lengths of PE 100 tubing
(polyethylene tubing, Clay Adams, Franklin Lakes, NJ). The tubing was heat-flared
at one end, using a high temperature cautery blade (Aaron Medical Industries, St.
Petersburg, FL), and fitted with a Teflon washer (Small Parts, Inc., Logansport, IN).
The other end of the tubing was fitted with a 19G x 2.5 cm needle from which the
hub had been removed. The needle was used to “guide” the implant during the
implantation procedure. Following placement of the cannula, the needle was
removed and replaced with a 0.7 cm long, 19G stainless steel, thin-walled tube
(Small Parts, Inc., Logansport, IN).
3.1.4. Surgery
In experiment 1, the surgeries performed included gonadectomy, intraoral cannula
implantation and silastic capsule implantation. A 5:1 ketamine (100 mg/ml, Phoenix
Pharmaceuticals, Inc., Mountain View, CA) to xylazine (20 mg/ml, Lloyd
Laboratories, Shenandoah, LA ) mixture was used to produce anesthesia (15 ml/kg
i.m.) during gonadectomy and intraoral cannula placement procedures. Both of these
procedures were carried out under a single administration of the anesthetic, back-to-
back, with gonadectomies followed immediately by intraoral cannula placement.
Gonadectomy in females consisted of bilateral incisions of the skin and muscle (1
to 2 cm from the midline and anterior to the hip) and the removal of the ovaries. In
31
Reproduced with permission o fth e copyright owner. Further reproduction prohibited without permission.
male animals, it consisted of an incision along the midline of the scrotum and the
removal of the testes. All wounds were closed with silk suture.
Placement of the intraoral cannulae in experiment 1 proceeded as follows. Each
cannula was inserted through the mouth slightly anterior to the first maxillary molar,
brought up subcutaneously along the skull, anterior to the ear and behind the
zygomatic arch, and exited at the top of the skull. Cannulae were then anchored to
the skull using 2 #1 - 72 x 0.32cm machine screws (Small Parts, Inc., Logansport,
IN) and Bosworth Fastray, self-curing dental plastic (Harry J. Bosworth Company,
Skokie, IL). Two cannulae were implanted in each animal, bilaterally.
Subcutaneous placement of estradiol and blank implants in experiment 1 proceeded
under halothane gas anesthesia. The halothane (Halocarbon Laboratories, River
Edge, New Jersey) was vaporized (Fluotec 3) and mixed with oxygen using a Fraser
Harlake VMC gas-driven small animal inhalation anesthesia system (Datex-Ohmeda,
Madison, Wisconsin). The halothane-oxygen mixture was maintained between 2.0
and 3.0 while the drive gas (oxygen) rate was maintained between 3.5 and 4.0 L/min
and delivered to the animals through a nose cone. The implantation procedure
consisted of a 1 cm incision at the nape ofthe neck, separation ofthe epidermis and
underlying muscle with a blunt surgical probe, insertion of the capsule, and wound
closure with silk suture. The same procedure was used for removal of the implants
one hour later.
32
Reproduced with permission o fth e copyright owner. Further reproduction prohibited without permission.
Intraoral cannula placement in experiment 2 proceeded as above, however only 1
cannula was implanted. There were no subcutaneous estradiol implantation
procedures in experiment 2.
3.1.5. Testing apparatus
Taste reactivity trials for experiments 1 and 2 were conducted using two 40 x 40 x
12.5 cm acrylic (Plexiglas) cages, each modified with a stainless steel divider to
measure 40 x 20 x 12.5 cm. Trials for individual animals could occur in either side
of the cages. The cages were mounted on a wall with a table below supporting a
mirror that enabled observation of the rat’s mouth through the bottom. Six 2.5 cm
diameter holes in the tops of the cages facilitated airflow and the passage of tubing to
the animals’ cannulae.
3.1.6. Intraoral infusion procedure
One-ml injections of tap water or test solution were dispensed from 12 ml syringes
fitted with 3.8 cm lengths of 27G stainless steel tubing. Two syringes were used
during testing; one contained 12 ml of the test solution and the other contained 12 ml
of tap water, which was used to flush the cannulae following application of the test
solution. Fluid injections were delivered to the oral cavity through a 45 cm length of
silastic tubing (1.6 mm inner diameter, 0.86 mm outer diameter, Dow Coming,
Auburn, MI) which was fitted onto the ends of the 27G stainless steel tubes on the
33
Reproduced with permission o fth e copyright owner. Further reproduction prohibited without permission.
syringes. The silastic tubing was connected to the 19G stainless steel portion of the
cannulae using a 3.8 cm length of 21G stainless steel tubing. One end ofthe 21G
tube was fitted into the lumen of the silastic tubing while the other end was inserted
into the lumen of the 19G stainless steel tube portion of the animals’ cannulae,
producing a 21G stainless steel “bridge” between the silastic tubing and the
cannulae. Test solutions were delivered to the oral cavity at a rate of 1.0 ml per
minute using a CMA/100 Microinjection Pump (Carnegie Medicin AB, Stockholm,
Sweden).
In experiment 1, video imaging of the animals’ orofacial responses to taste stimuli
was achieved using a “surveillance-style,” Panasonic WV-CL350 video camera with
a Panasonic LZ81/8 85 mm zoom lens (Panasonic Broadcast & Television Systems
Company, Los Angeles, CA) positioned approximately 60 cm from the testing cage.
The animals were free to move within the 40 x 20 cm area of the cage during testing
and the camera was moved so that the head of the animal remained dominant on the
monitor screen (WV-5490, Panasonic Broadcast & Television Systems Company,
Los Angeles, CA) during fluid injection sequences. In general, the animals do not
move once acclimated to the cage, and if they do, they usually pause during an
injection sequence. Orofacial responses to fluid injections were recorded with a
Panasonic NV-9240XD Video Cassette Recorder (Panasonic Broadcast & Television
Systems Company, Los Angeles, CA) at 30 frames per second onto Sony KCA-
60BRS Video Cassette Tapes (Sony Broadcast and Professional Company, Park
34
Reproduced with permission o fth e copyright owner. Further reproduction prohibited without permission.
Ridge, New Jersey) for later frame-by-frame analysis using a Panasonic AG-A650
Editing Controller (Panasonic Broadcast & Television Systems Company, Los
Angeles, CA).
In experiment 2, video images of the animals’ orofacial responses to taste stimuli
were captured with the same camera as above and presented on a Panasonic CT-
2087Y SVHS monitor/receiver (Panasonic Broadcast & Television Systems
Company, Los Angeles, CA). Images were recorded at 30 frames per second and
analyzed frame-by-frame using a Panasonic AG-DV2000 Digital Video Cassette
Recorder with “on-board” editing controller (Panasonic Broadcast & Television
Systems Company, Los Angeles, CA). Images were stored on Panasonic AY-
DV123EB digital video cassettes (Panasonic Broadcast & Television Systems
Company, Los Angeles, CA).
3.1.7. Conditioned taste reactivity and taste avoidance procedure
The conditioned taste aversion procedure was divided into three periods:
preconditioning, acquisition, and post-acquisition. All solutions were prepared on
the day of testing and delivered to the animals at room temperature. Although the
length of the preconditioning period depended on the experiment, the procedure was
universal and consisted of the following. The animals were placed in the testing
apparatus, had their cannulae connected to a syringe containing tap water and were
allowed to acclimate to the environment for five minutes. After the acclimation
35
Reproduced with permission o fth e copyright owner. Further reproduction prohibited without permission.
period, a 1.0 ml injection of tap water was administered to the oral cavity over a one-
minute period. The animals remained in the apparatus for another minute and were
then returned to their home cages.
The testing procedure for acquisition days was similar to that of preconditioning
except that infusion of a test solution (the exact composition of which depended on
the experiment) was followed by drug implantation for one hour (experiment 1) or
drug injection (experiment 2) and return to the home cage. After the five-minute
acclimation period, an infusion ofthe test solution was given, followed by an
infusion of tap water to flush the cannulae. Alternation between test solutions and
tap water consisted of, 1) manually immobilizing the animal, 2) disconnecting the
21G tube (and associated silastic tubing) from the animal’s cannula, and 3)
connecting the next 21G needle (and tubing) to the cannula. Specifically, each rat
was handled three times during taste reactivity testing. First, to be placed in the
testing apparatus and have the test solution tubing connected to the cannula; second,
to remove the test solution tubing and connect the tap water tubing to the cannula;
and third, to disconnect the tap water tubing and remove the rat from the testing
apparatus. Each acquisition trial was followed by a “recovery” day during which the
animal was handled only for weight measurement and not otherwise removed from
the home cage.
The procedure for post-acquisition day 1 was similar to that described for
acquisition but no drug implantation or injection followed taste reactivity testing.
36
Reproduced with permission o fth e copyright owner. Further reproduction prohibited without permission.
The animals’ orofacial responses to test solution injections were recorded on all
occasions. Although the procedure for taste reactivity testing was universal, the total
number of acquisition and post-acquisition tests depended on the experiment. In
experiment 1, the animals received 1 post acquisition taste reactivity test (PA1). In
experiment 2, the animals received 1 post-acquisition taste reactivity test (PA1)
followed by 3 post-acquisition bottle tests (PA 2, 3 and 4), during which the animals
were given access to a graduated cylinder containing test solution for one hour in
their home cages. The PA2 test was given immediately after PAL The PA3 and
PA4 tests were given on the following two days. There were no recovery days
during the post-acquisition period and no taste reactivity tests after PA1. Animals
were undisturbed during the bottle tests. After one hour, the cylinders containing test
solution were removed and the amount consumed was recorded.
3.1.8. Scoring o f taste reactivity data
For the scoring of the animals’ orofacial responses to intraoral infusions of the test
solutions during all acquisition and post acquisition tests, we used the 10 FAPs
described in the Taste Reactivity Test section (2.1.2) of this thesis. They are: mouth
movements (MM), tongue protrusions (TP), lateral tongue protrusions (LTP), paw
licks (PL), passive drip (PD), gapes (G), head shakes (HS), chin rubs (CR), forelimb
flailing (FF), and face washing (FW). Based on the work of Spector et al. (1987) and
A. C. Spector (personal communication, January, 1998), we have added two
37
Reproduced with permission o fth e copyright owner. Further reproduction prohibited without permission.
additional scoring categories. These categories are no response (NR) and no
data/movement (ND). NR indicates the absence of oral or other behavioral
responses for at least one second and ND indicates the absence of the animal from
the viewing screen for at least one second, as the result of locomotion or rearing.
Based on pilot studies done in this laboratory and the work of Grill et al. (1985),
Spector et al. (1987; A. C. Spector, personal communication, January, 1998), we
divided these individual FAPs into four general response profiles: digestive,
aversive, non-ingestive and ambiguous. The digestive response profile consists of
TP, LTP and PL; the aversive response profile consists of G, HS, CR and FF
(although FW is an aversive response, it is analyzed separately because it is scored
by duration rather than as a discrete event); the non-ingestive response profile
contains PD and NR; and the ambiguous response profile is composed of MM. Pilot
studies conducted in this lab have shown that MMs generally precede both digestive
and aversive response sequences and thus may not be definitively associated with
either positive or negative hedonic reactions. ND, though scored and analyzed, was
not included in the profiles since we have found, in pilot studies, that the animal may
be engaged in digestive or aversive response sequences during a bout of locomotion
or rearing.
The 12 FAPs listed above typically are scored either as discrete events or by the
duration (in seconds) of their appearance as relatively continuous behaviors (Spector
et al., 1987). The standard procedures were followed when scoring FAPs in
38
Reproduced with permission o fth e copyright owner. Further reproduction prohibited without permission.
experiments 1 and 2. For example, the PL was a discrete, cyclical event, occurring
with a frequency of approximately 6 per second. Such behaviors were scored by
simply tallying them as they occurred. If the view of the animal’s mouth was
obscured by paws, shadows, etc., during a PL sequence, and PLs were noted on
either end of the obstruction (i.e., before and after), it was assumed that they were
occurring at the 6 per second rate and scored accordingly. FW or NR sequences, on
the other hand, occurred relatively continuously and were scored as duration (in
seconds) from the appearance of the behavior (or a lack of behavior in the case of
NR) until it ceased or the animal engaged in another behavior. Table 1 describes the
12 FAPs, the morphological phenomena that define them and the scoring procedure
for each.
3.1.9. Statistical analyses
All ingestive, aversive, non-ingestive and ambiguous responses displayed during
acquisition and post-acquisition tests were analyzed as mean total responses, both
individually and as category profiles. For both experiments 1 and 2, these data were
analyzed with a two-factor (groups X tests) ANOVA with repeated measures on tests
for each experiment. A Newman-Keuls test also was used for paired comparisons in
each experiment. A t-test for independent samples also was used to compare the
mean orofacial responses for all FAPs on ACQ 1 and PA1 in both experiments. For
the bottle tests in experiment 2, sucrose consumption on post-acquisition test 2 was
39
Reproduced with permission o fth e copyright owner. Further reproduction prohibited without permission.
Table 1.
Orofacial response types, scoring criterion and duration.
Response Type
° ----- * -----------------......................... ...............
Scoring Criterion
Response
Duration
Ingestive
Tongue Protrusion (TP) Rhythmic protrusions of the tongue along the
midline, obscuring view of upper incisors.
discrete event
(appx. 6/s)
Lateral Tongue Protrusion
(LTP)
Non-rhythmic extension of the tongue from
either side of the mouth, resulting in an
asymmetrical lip flare.
discrete event
Paw Lick (PL)
Aversive
Large amplitude rhythmic protrusions of the
tongue along the midline directed at the
forepaws or substrate.
discrete event
(appx. 6/s)
Gape (G) Large amplitude opening of the mandible and
retraction of the comers of the mouth resulting
in a clear view of both the upper and lower
incisors.
discrete event
Head Shake (HS) High frequency burst of side-to-side head
movements resulting in expulsion of fluid from
the oral cavity.
discrete event
Chin Rub (CR) The mouth is brought into direct contact with the
substrate (or forelimb laid flat on the substrate)
and moved forward along the surface, dispelling
fluid from the oral cavity.
discrete event
Forelimb Flail (FF) High frequency side-to-side shaking of one or
both paws.
discrete event
Face Washing (FW) Similar to paw lick except the forelimbs are continuous
brought into direct contact with the face and
head.
(duration in s)
Non-Ingestive or Ambiguous
Mouth Movement (MM) Low amplitude, rhythmic movement of the
mandible that resembles chewing behavior.
discrete event
(appx. 6/s)
Passive Drip (PD) Build up of fluid at midline of lower mandible
and subsequent dripping of fluid onto substrate.
continuous
(duration in s)
No Response (NR) Absence of oral or other behavior for at least
one second. Scored from end of last response to
beginning of next response.
continuous
(duration in s)
No Data (ND) Absence of animal from viewing screen for at
least one second.
continuous
(duration in s)
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
analyzed with a one-factor analysis of variance (ANOVA) and sucrose consumption
across post-acquisition tests 2, 3 and 4 was analyzed with a two-factor (groups X
tests) ANOVA with repeated measures on tests.
3.2. Experiment 1
3.2.1. Hypotheses
In this experiment, gonadectomized male and female Sprague-Dawley rats were
implanted on acquisition days with either a 30 mm estradiol or 30 mm blank implant
for one hour to determine whether there is a hedonic shift in the animals’ evaluation
of a sucrose solution that has been previously paired with estradiol, as measured by
the taste reactivity test. We also sought to determine whether there is a sex
difference in the effect of estradiol on the animals’ hedonic evaluation of an
estradiol-paired sucrose solution.
3.2.2. Methods
Ten male and 10 female Sprague-Dawley rats that were 97-100 days old were each
gonadectomized and then implanted bilaterally with intraoral cannulae. They were
given one week to recover from the surgery and preconditioning occurred during this
period. Following 7 days of preconditioning, they were given 3 acquisition tests and
1 post-acquisition test. All testing was done at the beginning of the dark phase of the
light/dark cycle. Two animals died prior to the first acquisition test.
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
On all three acquisition tests, the remaining animals were given an intraoral
infusion of 10% (weight/volume) sucrose solution followed immediately by
implantation of either one 30 mm estradiol implant (n=9) or one 30 mm blank
implant (n=9) for one hour. Animals were allowed to recover for one day before the
next acquisition trial began.
Following the third recovery day, a post-acquisition test was conducted. Animals
received an intraoral infusion of the 10% sucrose solution but did not receive an
estradiol or blank implant. Four animals (1 estradiol and 3 blank) experienced a total
failure of both cannulae and did not receive the post acquisition trial.
3.2.3. Results
Tables 2 and 3 show the results of the analyses of variance and independent t-tests
for the taste reactivity data. Figures 1 through 5 show the mean (± SE) frequency of
GAPE, ingestive, aversive, non-ingestive, and MM responses across the 4 tests,
respectively. Estrogen-treated males (ME [n=4]) and females (FE [n=4]) did not
differ significantly on any measure of orofacial reactivity. Similarly, control males
(MC [n=4]) and females (FC [n=4]) did not differ significantly on any measure of
orofacial reactivity (Table 2). ME and FE animals and MC and FC animals were
pooled for further analysis (E [n=8] and C [n=8] groups, respectively).
The E and C group animals differed significantly in their display of GAPE
responses (F(3,42) = 5.985, p = 0.002; also see Figure 1). Two-factor ANOVA for
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 2.
Results of 2-factor (groups x tests) ANOVAs with repeated measures on tests for ME vs. FE, MC vs.
FC, and E vs. C group mean orofacial responses across the four taste reactivity tests in experiment 1.
Response Category
ME vs. FE MC vs. FC E vs. C
F(3,18)
P
F(3,18)
P
F(3,42) P
Ingestive Responses 1.316 0.300 0.570 0.642 1.774 0.167
Aversive Responses 1.040 0.399 0.568 0.643 1.502 0.228
Mouth Movement 1.810 0.181 0.385 0.765 0.456 0.714
Tongue Protrusion 1.829 0.178 0.992 0.419 1.042 0.384
Lateral Tongue Protrusion 0.974 0.427 1.149 0.356 0.205 0.892
Paw Lick 2.950 0.061 0.118 0.948 1.488 0.232
Gapes 0.153 0.927 2.876 0.065 5.985 0.002**
Forelimb Flail 1.286 0.546 0.925 0.449 0.574 0.635
Head Shake — — 1.000 0.415 1.000 0.402
Face Washing 0.600 0.623 1.157 0.353 0.175 0.913
No Reaction 0.228 0.875 1.390 0.341 1.825 0.157
Note: Dashes indicate that there were no statistics generated for the HS category in the ME vs. FE
comparison as these groups did not display these responses. Since no animals displayed the passive
drip response in this experiment, the mean non-ingestive category is reflected in the no response
score. **Significance at the alpha = .01 level.
Table 3.
Results of independent t-tests for E and C group mean orofacial responses on ACQ1 and PA1 in
experiment 1. ■ ____
ACQ 1 ____________ PA 1
Response Category
..... t(16)
P
t(14)
P
Ingestive Responses -0.971 0.346 -3.173 0.007**
Aversive Responses -0.266 0.794 1.462 0.166
Mouth Movement 1.656 0.117 2.694 0.017*
Tongue Protrusion 0.295 0.772 -2.933 0.011*
Lateral Tongue Protrusion -0.864 0.401 -2.073 0.057
Paw Lick -1.798 0.091 -1.606 0.131
Gape -0.406 0.690 2.708 0.017*
Forelimb Flail -1.000 0.332 -.0.822 0.425
Head Shake — — _ _ —
Face Washing - -- 0.632 0.537
No Reaction -1.491 0.155 1.142 0.273
Note: Dashes indicate that there were no statistics generated for these categories as the animals did
not display such responses during these tests. Since no animals displayed the passive drip response in
this experiment, the mean non-ingestive category is reflected in the no response scores. *Significance
at the alpha = .05 level. ** Significance at the alpha = .01 level.
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Gape Responses
o
+
ACQ 1 ACQ 2 ACQ 3 PA 1
Tests
Figure 1. Mean (+/- SE) GAPE responses among the E and C group
animals in response to intraoral infusions of 10% sucrose solution across
the four tests. The frequency of GAPEs changed differently across the 4
tests for E and C group animals (p=.Q02). The E group animals displayed
more GAPEs than the C group animals on PA1 test (p=.017).
Ingestive Responses
3001
1:
o *
< L >
S'
200 -
100 -
1 1 1 -----
ACQ 1 ACQ 2 ACQ 3
Tests
PA 1
Figure 2. Mean (+/- SE) ingestive responses among the E and C group
animals in response to intraoral infusions of 10% sucrose solution across
the four tests. The frequency of ingestive responses did not change
differently across the 4 tests for E and C group animals (p=.17).
However, the E group animals displayed fewer ingestive responses on
PA1 than the C group animals (p=.007).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Aversive Responses
o
th
S'
£Z3
15-|
10 -
1r---
— ■ — E
- ■ * - C
ACQ 1 ACQ 2 ACQ 3 PA 1
Tests
Figure 3. Mean (+/- SE) aversive responses among the E and C group
animals in response to intraoral infusions of 10% sucrose solution across
the four tests. No significant differences were found.
Non-ingestive Responses
c
< L >
3
O r *
< D
£
2 T
0 0
7.5-1
5.0
0.0
ACQ 1 ACQ 2 ACQ 3 PA 1
Tests
-®— E
--A— Q
Figure 4. Mean (+/- SE) non-ingestive responses among the E and C
group animals in response to intraoral infusions of 10% sucrose solution
across the four tests. No significant differences were found.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Mouth Movements
O
§
£
S'
m
+
200 n
100 -
- * - E
- Q
ACQ 1 ACQ 2 ACQ 3 PA 1
Tests
Figure 5. Mean (+/- SE) MM responses among the E and C group
animals in response to intraoral infusions of 10% sucrose solution across
the four tests. No significant differences were found.
GAPE responses across the 4 taste reactivity tests indicated that E group animals
displayed an increase in GAPE responses whereas C group animals did not
(F(3,42)-5.985, p=0.0Q2). No other significant interaction effects between E and C
group animals were found (Table 2). Compared to C group animals, the E group
showed a non-significant decrease (F(3,42) = 1.774, p = 0.167) in ingestive
responses (Figure 2) and a non-significant increase (F(3,42) = 1.502, p = 0.228) in
aversive responses (Figure 3) across tests. E and C groups did not differ in their
display of non-ingestive responses and mouth movements (F(3,42) = 1.825, p =
0.157 and F(3,42) = 0.456, p = 0.714, respectively) nor did they exhibit any non
significant trends (Figures 4 and 5). Since no PD responses were observed, the non-
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
ingestive category is composed of only NR. The only HS observed was displayed by
an animal from the C group.
Independent t-test for ACQ1 and PA1 data indicated that E and C group animals
did not differ significantly on any measure of orofacial reactivity on ACQ1 but E
group animals showed significantly fewer TP and ingestive responses (t(14)=-2.933,
p=0.011 and t(14)=-3.173, p-0.007, respectively) and significantly more GAPE and
MM responses (t(14)=2.708, p=0.017 and t(14)=2.694, p=0.017, respectively) than C
group animals on PA1 (Table 3).
3.2.4. Discussion
These results indicate that there is no sex difference in the effect of 30mm estradiol
implants on taste reactivity among Sprague-Dawley rats. The effect of estradiol
implants on taste reactivity irrespective of sex is less clear. While the estradiol-
treated animals showed a significant decrease in ingestive responses relative to
control animals on PA1, they failed to show a significant increase in aversive or non-
ingestive responses. This suggests that an overall shift (from ingestive to aversive)
in the animals’ palatability evaluation did not occur, as has been reported for LiCl-
paired sucrose solutions (Parker and MacLeod, 1991).
The significant decrease in TP responses and significant increase in GAPE and
MM responses, individually, is difficult to interpret as they are individual response
measures and, as Spector, et al. (1988) have suggested, the composite scores may be
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
more meaningful. It may be that these individual responses reflect elements of an
incomplete palatability shift or they may just reflect variability in the samples. We
believe this second hypothesis is more likely considering the small sample size.
Moreover, since there were no bottle tests in this experiment, we were unable to
evaluate the animals’ appetitive behavior with respect to the sucrose stimulus. It
would have been informative to know how the animals reacted in a bottle test,
compared to the intraoral infusions. Hence, we cannot determine from the present
results whether or not the observed taste reactivity measurements were associated
with a concomitant conditioned avoidance.
There are several possible sources of variability in this experiment. First, our
failure to find a sex difference could be due to the small sample size (n=4) of the
groups. This is especially true with regard to the analyses of individual orofacial
response measures. Pilot studies in this lab have indicated that there is often a high
degree of variability among the individual orofacial response measures across tests.
Second, the repeated exposure to halothane gas anesthesia during the implantation
procedures may have interfered with the acquisition process. Pilot work in our lab
suggests that repeated exposure to halothane attenuates LiCl-induced conditioned
taste avoidance (Chambers and Hayes, unpublished data). Third, the use of bilateral
intraoral implants in this experiment may have contributed to the aversive responses
observed. We have found, during pilot studies in this lab, that the after implantation
of the cannulae there is a marked decrease in the animals’ consumption of the normal
48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
solid food diet, which we assume to be the result of pain associated with eating. This
may be exacerbated by the use of bilateral cannula implants.
All of the taste reactivity experiments described in chapter 2 (Estrogen and the
Conditioned Food Aversion Paradigm) of this thesis, with the exception of the King
et al. (1997) study, utilized male rats as they are larger and thus better able to tolerate
the cannulae. Taste reactivity studies often employ chronic preparations as well,
which serve in several studies over an extended period of time, and the use of
bilateral implants minimizes the possibility of losing animals due to cannula failure
(A.C. Spector, personal communication, January, 1997).
In order to address the issues cited above, we designed experiment 2 in an effort to
minimize the influence of these variables. Since we employ acute preparations in
our studies, we decided that it would be more appropriate to use a combination of
unilateral intraoral cannulae and larger sample sizes. We felt that unilateral implants
would reduce feeding complications and larger sample sizes would off-set the
increased potential for cannula failures. Also, the employment of an estradiol
injection removed any possible confound caused by the use of halothane.
3.3. Experiment 2
3.3.1. Hypotheses
In this experiment, gonadectomized male and female Sprague-Dawley rats were
injected on acquisition days with either an estradiol suspension or a control vehicle
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
to determine whether there is a hedonic shift in the animals’ evaluation of a sucrose
solution that has been previously paired with estradiol, as measured by the taste
reactivity test. We also sought to determine whether there is a sex difference in the
effect of estradiol on the animals’ hedonic evaluation of an estradiol-paired sucrose
solution.
3.3.2. Methods
Twenty male and 20 female Sprague-Dawley rats that were 104-107 days old were
each gonadectomized and then implanted unilaterally with intraoral cannulae. They
were given one week to recover from the surgery and preconditioning occurred
during the final 4 days of this period. Following the 4 preconditioning tests, they
were given 3 acquisition tests and 4 post-acquisition tests. All testing was done at
the end of the light phase of the light/dark cycle. Six animals experienced a total
failure of their cannulae prior to the first post-acquisition test and were dropped from
the study.
On all three acquisition tests, the remaining animals were given an intraoral
infusion of 10% (weight/volume) sucrose solution followed immediately by an
injection of either estradiol (n=17) or a control vehicle (n=17). Animals were
allowed to recover for one day before the next acquisition test.
Following the third recovery day, the first post-acquisition test (PA1) was
conducted. Animals received an intraoral infusion of the 10% sucrose solution but
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
did not receive an estradiol or control injection. Immediately following the intraoral
infusion, the animals were returned to their home cages and given a graduated
cylinder containing a 10% (weight/volume) sucrose solution for one hour (PA2).
The PAS and PA4 tests occurred on the two following days, with no intervening
recovery days. All animals had access to their normal food and water during PA2, 3,
and 4.
3.3.3. Results
Tables 4 and 5 show the results of the analyses of variance and independent t-tests
for the taste reactivity data. Figures 6 through 9 show the mean (± SE) frequency of
ingestive, aversive, non-ingestive, and MM responses across the 4 tests, respectively.
Table 6 shows the results of 1-factor ANOVA, group by group comparisons of
sucrose consumption across the 3 bottle tests. Figure 10 shows the mean (± SE) ml
of sucrose consumed across the 3 bottle tests. ME (n=8) and FE (n=9) group animals
did not differ significantly on any measure of orofacial reactivity. Similarly, MC
(n=8) andFC (n=9) group animals did not differ significantly on any measure of
orofacial reactivity (Table 4). ME and FE animals and MC and FC animals were
pooled for further analysis (E [n-17] and C [n=T7] groups, respectively).
E and C group animals differed significantly in their display of ingestive responses
and mouth movements (F(3,96) = 5.223. p = 0.002 and F(3,96) = 5.401, p = 0.002,
respectively). One-factor ANOVA for each of the 4 taste reactivity tests indicated
51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 4.
Results of 2-factor (groups X tests) ANOVAs with repeated measures on tests for ME vs. FE, MC vs.
FC, and E vs. C group mean orofacial responses across the four teste reactivity tests in experiment 2.
ME vs. FE MC vs. FC E vs. C
Response Category F(3,45) p F(3,45) p F(3,96) P
Ingestive Responses 0.668 0.576 2.204 0.101 5.223 0.0024*
Aversive Responses 0.446 0.721 0.908 0.445 0.995 0.399
Non-ingestive Responses 2.012 0.126 0.850 0.474 2.675 0.052
Mouth Movement 0.804 0.498 2.457 0.075 5.401 0.002^
Tongue Protrusion 0.701 0.556 1.191 0.324 2.334 0.079
Lateral Tongue Protrusion 1.062 0.375 1.527 0.221 0.565 0.639
Paw Lick 0.068 0.977 0.095 0.962 1.206 0.312
Passive Drip 2.378 0.082 1.134 0.345 1.796 0.153
Gape 0.222 0.881 1.037 0.386 1.418 0.242
Forelimb Flail 0.718 0.546 1.341 0.273 0.735 0.534
Head Shake 0.882 0.457 -- — 1.000 0.396
Face Washing 0.639 0.594 - -- 0.735 0.534
No Reaction 1.166 0.333 1.099 0.360 1.230 0.303
Note: Dashes indicate that there were no statistics generated for these categories in the MC vs. FC
comparison as these groups did not display such responses. ♦♦Significance at the alpha = .01 level.
Table 5.
Results of independent t-tests for E and C group mean orofacial responses on ACQ1 and PA1 in
experiment 2. _________ ___________________
ACQ 1 PA 1
Response Category t(32)
P
t(32)
P
Ingestive Responses -1.112 0.275 -3.135 Q.004M
Aversive Responses -1.074 0.291 1.705 0.098
Non-ingestive Responses -0.780 0.441 1.451 0.157
Mouth Movement 0.993 0.328 2.119 0.042^
Tongue Protrusion -0.875 0.388 -1.858 0.072
Lateral Tongue Protrusion 1.378 0.178 0.596 0.556
Paw Lick -0.551 0.585 -1.978 0.057
Passive Drip - - 1.416 0.166
Gapes -1.074 0.291 1.191 0.242
Forelimb Flail - -- 1.376 0.178
Head Shake - - - -
Face Washing - - 1.000 0.325
No Reaction -0.780 0.441 0.688 0.497
Note: Dashes indicate that there were no statistics generated for these categories as the animals did
not display such responses during the tests. ♦Significance at the alpha = .05 level. ♦♦Significance at
the alpha = .01 level.
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Ingestive Responses
300
0
S
§< 200
£
S'
m
^ 100
1
0-1------------------- 1 ------------------- 1 ----------------—i--------------------- 1 ------------------
ACQ 1 ACQ 2 ACQ 3 PA 1
Tests
Figure 6. Mean (+/- SE) ingestive responses among the E and C group
animals in response to intraoral infusions of 10% sucrose solution across
the four tests. The frequency of ingestive responses changed differently
across the 4 tests for E and C group animals (p=.002). The E group
animals displayed fewer responses than C group animals on ACQ2,
ACQ3 and PA1 (p=.00, p=.02 and p=.00, respectively).
Aversive Responses
o
: c
a >
§ *
a >
£
S'
G O
2
1
0
ACQ 1 ACQ 2 ACQ 3 PA 1
■ E
■ C
Tests
Figure 7 . Mean (+/- SE) aversive responses among the E and C group
animals in response to intraoral infusions of 10% sucrose solution across
the four tests. No significant differences were found.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Non-ingestive Responses
10.0-
o
S3
8“
£
S'
c w
7.5-
5.0-
2.5-
0 . 0 -
_ * _ E
— C
ACQ 1 ACQ 2 ACQ 3
Tests
PA 1
Figure 8. Mean (+/- SE) nOn-ingestive responses among the E and C
group animals in response to intraoral infusions of 10% sucrose solution
across the four tests. No significant differences were found.
Mouth Movements
150-1
O
S
s
t Z )
100-
50-
ACQ 1 ACQ 2 ACQ 3
Tests
PA 1
Figure 9. Mean (+/- SE) MM responses among the E and C group
animals in response to intraoral infusions of 10% sucrose solution across
the four tests. The frequency of mouth movements changed differently
across the 4 tests for E and C group animals (p=.Q02). The E group
animals displayed more mouth movements than C group animals on
ACQ2, ACQ3 and PA1 (p=.00, p=.00 and p=.04, respectively).
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 6.
Results of 1-Factor (Tests) ANOVA group by group comparisons for mean consumption of 10%
sucrose solution on PA2, PA3 and PA4 in experiment 2.
Groups PA 2 PA 3 PA 4
ME vs. FE F(l,15) = 1.207
p =0.289
F(l,15) = 5.589
p = 0.032**
F(l,15)= 15.781
p = 0.001**
ME x FC F(l,15) = 3.400
p = 0.085
F(l,15) = 4.177
p = 0.060'
F(l,15) = 0.155
p = 0.700
ME x MC F(l,14) = 24.973
p = 0.000**
F(l,14) = 2.843
p = 0.114
F(l,14)= 1.250
p = 0.282
FE x MC F(l,15) = 67.880
p = 0.000**
F(l,15) = 19.784
p = o.ooo**
F(l,15) = 31.684
p = 0.000**
FE x FC F(l,16)= 10.637
p = 0.005**
F(l,16) = 48.832
p = 0.000**
F(l,16) = 20.600
P = 0.000**
FC x MC F(l,15) = 6.687
P = 0.021*
F(l,15) = 0.049
P = 0.828
F(l,15) = 3.069
P = 0.100
* Significance at the alpha = .05 level **Significance at the alpha = .01 level
Sucrose Consumption
-m - ME
— *r-MB
FE
—o— FB
v ---------------- 1 ----------------1 — : ---- : ------- r ---------------
PA 2 PA 3 PA 4
Tests
Figure 10. Mean (+/- SE) sucrose consumption for ME, FE, MB and FB
groups on PA2, PA3, and PA4. The FE group consumed less of the sucrose
solution than either the ME or control groups on all tests.
30-.
§
a,
§
U
S'
m
20 -
10 -
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
that E group animals displayed significantly fewer ingestive responses on ACQ2
(F(l,32)=46.674, p=0.000), ACQ3 (F(l,32)=5.990, p=0.020), and PA1
(F(l,32)=9.826, p-0.004) compared to C group animals (Fig. 6). One-factor
ANOVA for each of the 4 taste reactivity tests indicated that E group animals
displayed significantly more mouth movements on ACQ2 (F(l ,32)=37.669,
p=0.000), ACQ3 (F(l,32)=9.439, p=0.004), and PA1 (F(l,32)=4.485, p=0.042)
compared to C group animals (Fig. 9).
Two-factor ANOVA did not indicate any other significant interaction effects
between E and C group animals (Table 4). E and C group animals showed a similar
decrease and increase in aversive and non-ingestive responses, respectively, across
the 4 tests (Figures 7 and 8). Independent t-tests for ACQ1 and PA1 data also
indicated that E and C group animals did not differ significantly on any other
measure of orofacial reactivity (Table 5). The only HS and FW observed in this
experiment were displayed by FE group animals. One FE animal displayed 3 HS
and 5 seconds of FW while another FE animal displayed 2 seconds of FW.
ME, FE, MC, and FC groups consumed significantly different amounts of the 10%
sucrose solution across the 3 PA bottle tests (F(6,60) = 2.267, p = 0.049). Newman-
Keuls post hoc analysis of the pooled PA2-PA4 values indicated that the ME group
differed significantly from the MC and FE groups (p = 0.002 and p = 0.002,
respectively) but not from the FC group (p = 0.119). The FE group differed
significantly from the ME, MC and FC groups (p - 0.002, p = 0.000 and p = 0.000,
56
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
respectively). The MC and FC groups also differed significantly from each other (p
= 0.034).
The FE group consumed significantly less of the sucrose solution than either of the
control groups on all three tests, and significantly less than the ME group on PA3
and PA4. The ME group drank significantly less than the MC group on PA2 but did
not otherwise differ from either of the control groups. The MC and FC groups
differed in their consumption of the sucrose solution on PA2, but not on PA3 or 4
(Table 6 and Figure 10).
3.3.4. Discussion
These results indicate that there is no sex difference in the effect of 500 pg/kg
estradiol injections on taste reactivity among Sprague-Dawley rats. They also
suggest that estradiol injections cause a decrease in ingestive orofacial responses and
an increase in MM (an ambiguous response), but do not cause an increase in aversive
or non-ingestive responses compared to control animals. The failure of the estradiol-
treated animals to increase their display of aversive responses (in fact, they decreased
across the 4 test days) indicates that a conditioned shift in the animals’ palatability
evaluation of the sucrose solution did not occur.
The fact that the ME and FE groups consumed significantly less sucrose than the
MC and FC groups, respectively on PA2, indicates that estradiol causes a
conditioned avoidance of a paired sucrose solution. However, the fact that the ME
57
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and FC groups did not differ from each other in their consumption of the sucrose
solution on PA2 suggests that there is a sex difference for the effects of estradiol on
this avoidance. Following PA2, only the FE group showed a continued significant
decrease in consumption of the sucrose solution, indicating that possibly the FE
animals experienced a stronger effect of the estradiol injections.
It is also interesting that the control groups differed significantly in their
consumption of the sucrose solution on PA2; females consumed less than males.
This finding suggests a sex effect that cannot be accounted for by the estradiol
treatment. One possible source of this disparity between the control groups may be
the liquid diet. It is possible that there are differential gastrointestinal effects of the
liquid diet for males and females, and this effect may involve illness or malaise of
some kind. However, to our knowledge, there has been no research to date that has
examined the possibility of a sex difference with regard to the gastrointestinal
consequences of a liquid diet. Additionally, a sex effect of the liquid diet would
account for the lowered consumption of the female groups in general compared to
the male groups (see Figure 10). Under this hypothesis, the significantly lower
consumption of the sucrose solution by the FE group would reflect the combined
effects of the estradiol treatment and the liquid diet. This hypothesis is also
consistent with the observation that they were the only animals to display the few HS
and FW responses seen in this experiment. It should be noted again that bottle test
results in this study must be viewed with caution since it involved the forcible pre-
58
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
exposure of the CS (through taste reactivity testing), which may confound the bottle
test results (see 2.1.3. Aversion versus avoidances).
3.4. General discussion
Taken together, the results of experiments 1 and 2 suggest that estradiol does not
induce an aversion (as measured by the taste reactivity test) to a paired sucrose
solution but it does cause a conditioned avoidance of that solution in a subsequent
bottle test.' The failure to find significant effects at the level of individual orofacial
response measures (with the exception of MM, which will be discussed shortly)
among the larger samples in experiment 2 supports our hypothesis that such findings
in experiment 1 might have reflected variation attributable to the smaller sample
sizes.
The present results are not consistent with the Ossenkopp et al. (1996) finding that
estradiol produces a robust negative shift in palatability in addition to the avoidance
of the sucrose solution in the two-bottle preference test. The most likely explanation
for this difference is that it is caused by methodological differences. The most likely
difference in methodology between Ossenkopp et al. (1996) and the present study
which might account for the observed difference in results is unknown. However, it
should be noted that the dose of estradiol used in the present experiment (500 pg/kg)
was 5 times stronger than the lOOpg/kg dose used by Ossenkopp et al. (1996). The
relationship of the present findings to those reported by King et al. (1997) is unclear.
59
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
These authors reported that estradiol injections caused an avoidance of a paired
saccharin solution in the bottle test but did not cause aversive orofacial responses and
did not alter ingestive orofacial responses. Specifically, they reported that estradiol-
treated animals displayed TP and MM responses. However, they do not report
whether or not there was any increase or decrease in these ingestive responses
following the estradiol-saccharin pairing.
The pattern of orofacial responses in the present study are in fact most similar to
those reported by Cross-Mellor et al. (1999). As described in the Estrogen and
cholecystokinin (2.3.1) section above, these authors found that injections of CCK
caused a significant reduction in the display of ingestive responses and no increase in
aversive responses. They suggested that such results are consistent with the
hypothesis that CCK operates through a satiety mechanism.
We believe that the present findings are similarly consistent with the hypothesis
that estradiol acts through a satiety mechanism. As we discussed in the Scoring of
taste reactivity data (3.1.8) section above, we have found during pilot studies in this
lab that the MM generally precedes both ingestive and aversive responses patterns
and may not be definitively associated with either pattern. It is possible that the
King et al. (1997) failure to observe a decrease in ingestive responses might be due
to their inclusion of the MM response in the ingestive profile. It is unclear how our
conception of the MM as an ambiguous response compares to that of Cross-Mellor et
al. (1999). However, the fact that the estradiol-treated animals reduced their display
60
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of ingestive orofacial responses and avoided the sucrose solution in the bottle test
without evidence of a negative hedonic shift is more consistent with a conditioned
satiety hypothesis than any kind of aversive explanation. Furthermore, with possible
methodological differences in the scoring of ingestive responses aside, the findings
of King et al. (1997) and the present study that estradiol produces a conditioned
avoidance without a concomitant negative shift in taste reactivity, combined with the
similar effects of CCK reported by Cross-Mellor et al. (1999), offers further support
for the hypothesized interaction between CCK and estradiol to produce reductions in
food consumption.
4. HUMAN FOOD AVERSIONS AND THE ARTICULATED THOUGHTS
IN SIMULATED SITUATIONS PARADIGM
4.1. Human food aversions
4.1.1. Taste hedonics
Conditioned food aversions among humans are quite common and share the
essential features of CTA (robust single trial learning, taste cues, illness association,
etc.) that have been characterized for non-human species (Chambers et al., 1995;
Garb et al., 1974; Logue et al., 1983; Logue et al., 1981). The literature describes
studies of human CTA that range from the therapeutic induction of illness (such as
the aversion therapies used in some alcohol and smoking treatment centers) to its
61
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
role in serious human diseases such as anorexia nervosa (De Silva and Rachman,
1987). It is interesting to note that studies by Bernstein and Borson (1986) and
Gustavson et al. (1989) have suggested that estrogen-induced conditioned food
aversions may play a role in the development of anorexia nervosa.
Although the study of food aversions among humans is necessarily constrained by
ethical considerations, a few studies have explored the experimental induction of
CTA within specific populations of subjects. Bernstein (1978) reports the pairing of
a novel candy flavor with chemotherapy-induced nausea in children undergoing
cancer treatment in order to divert the resulting food aversion from becoming
associated with more nutritious foods in their diets. Boland et al. (1978) and Mellor
and White (1978) used lithium and motion sickness, respectively, to induce
conditioned taste aversions to alcohol. In general, however, human populations are
not appropriate for the careful experimental manipulations required to explore the
relationship between shifts in hedonic evaluation and avoidance of illness-paired
foods such as those described in the previous sections of this thesis.
Most studies of human food aversions have relied on questionnaire formats to
study the more typical demographic groups. De Silva et al. (1987), Garb et al.
(1974) and Logue et al. (1981), for example, all used questionnaires to estimate the
incidence of CTA and the target foods that produce them in university populations.
While such studies certainly provide much information on the characterization of
CTA in humans, they offer little insight into the relationship between hedonics and
62
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
avoidance. Furthermore, Chamberlain and Haaga (1999) and Davison, Vogel and
Coffman (1997) have pointed out that a limitation of “pencil and paper” methods is
that they may be constrained by the experimenter’ s prior judgements as to the types
of responses that subjects might give. This problem can be compounded in studies
of conditioned food aversion among humans by simple errors of language
interpretation. Yeomans and Symes (1999) have reported that there are two
interpretations of the term “palatability” in common English usage and that this
difference affects the reliability of evaluations of hedonic responses to food. Zylan
(1996) found that survey type (discrete choice versus open-ended questions)
significantly affected the response to the statement “I usually stop eating when
The author found that 64.4% of the subjects chose “fullness” when it was
an alternative choice, but only 26.9% wrote in this reason.
Although studies of CTA among humans have not focused specifically on subjects’
hedonic evaluation of illness-paired foods, many studies have looked at the hedonic
evaluation of taste among humans in its own right, particularly with regard to innate
taste preferences and aversions. Graillon, Barr, Young, Wright and Hendricks
(1997) demonstrated the ability of an intraoral application of sucrose solution to
increase mouthing and hand-mouth contact and reduce crying in human newborns.
Steiner (1974) reported work with human newborns, very similar to the animal taste
reactivity work of Grill and Berridge (1978), Grill and Norgren (1985) and Spector
et al. (1988), in which the facial responses of infants to intraoral applications of
63
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
sweet, sour and bitter tastes were video-recorded and later analyzed for facial motor
reactions. The author noted a marked relaxation of the facial muscles and retraction
of the mouth angles (as in a smile) as well as sucking and licking sequences in
response to sweet tastes. Sour tastes elicited cyclical pursing of the lips, while bitter
tastes produced depression of the mouth angles and an “arch-like” opening of the
mouth. The authors describe this latter response as a characteristic face of “dislike”
or “aversion.” These responses to sweet, sour and bitter tastes appeared identical in
neurOlogically intact, hydroaneeephalic and anecephalic neonates. Furthermore, the
author pointed out that in similar studies with adults, subjects were often moved to
embarrassment by their inability to control their “grimaces” in response to bitter taste
solutions. Based on these findings, the author suggests that not only is human taste
reactivity innate but also that it is controlled in a reflex-like fashion by primitive
brain structures.
Perl, Shay, Hamburger and Steiner (1992) and Steiner (1974) extended the above
findings to include orofacial reactions to odor stimuli. The authors found that
orofacial reactions similar to those produced by intraoral taste applications were
evoked by food-like odors among neonates, adults and demented elderly patients.
Soussignan and Schaal (1996) conducted a study of children’s facial responses to
pleasant and aversive odors. Children in this experiment were confronted with
pleasant (fruity and floral) and aversive (fecal and fishy) odors and asked to provide
a hedonic rating of them on a five point scale. Additionally, the children’s facial
64
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
responses to the odors were surreptitiously video-recorded for later analysis using a '
much more stringent coding method than that used by Perl et al. (1992) and Steiner
(1974). These authors found that the facial responses of the children were not highly
stereotyped and varied as a function of gender or social condition (i.e., whether or
not there was an adult in the room with the subject during testing).
The Soussignan et al. (1996) study was designed to test the hypothesis proposed by
Steiner (1974) that human facial responses to hedonic stimuli are innate, reflex-like
and not subject to modulation by higher cortical structures. Although their results
challenged this hypothesis, the children did display fundamentally differentiated
facial responses. Specifically, the authors found that the children spontaneously
reacted with differential facial responses to odorants that they concurrently
differentiated at the hedonic level through verbal report. In comparison with
pleasant and control odors, unpleasant odors elicited an increased amount of negative
facial responses even though they were subject to masking when there was an adult
present in the room during testing (Soussignan et al. 1996). This finding raises the
question of whether or not reliable behavioral indices can be found that correlate
with a human subject’s verbal reports of hedonic evaluation of aversive food items
since behavior may be suppressed in order to conform to social expectations.
The modulation of innate hedonic responses by higher brain structures in different
social situations suggests that the cognitions mediating this process may be a useful
tool in the study of CTA among humans since social situations can be manipulated
65
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and cognitions may indicate both hedonic evaluation and motivation. It may be that
subjects’ cognitions can be used to explore the motivational process involved in the
avoidance of illness-paired foods in lieu of, and possibly with more sensitivity, than
behavioral measures.
4.1.2. Articulated Thoughts in Simulated Situations
Regarding human studies of CTA, the task remains of finding a suitable method to
explore subjects’ hedonic evaluations and avoidance of illness-paired food items
short of exposing them to the foods and possibly causing them to become ill. The
Articulated Thoughts in Simulated Situations (ATSS) cognitive assessment
technique (Davison, Robins and Johnson, 1983) provides a controlled experimental
environment where this might be achieved.
The ATSS procedure requires a subject to listen to an audiotape in which actors
portray a social situation that either directly or indirectly involves the listener. The
subjects imagine that they are actively involved in the scenario they are listening to
and are periodically prompted (every 15-20 seconds) to respond aloud with his/her
thoughts in response to what they have just heard (i.e. tapping into their thoughts in
response to the situation that they are imagining they are involved in and saying
these thoughts out loud). The subjects’ responses are recorded and later coded and
analyzed for the cognitions of interest by coders who are unaware of the subjects’
experimental conditions (Davison et al., 1997).
66
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Regarding the coding of ATSS data, the creation of a specific content analytic
scheme depends on the specific research interest of the experimenter (Davison et al.,
1997). A general approach is to prepare transcripts of subject responses to the
scenarios and then code the recorded thoughts according to predetermined content
categories. The coding procedure can be conducted using the transcripts, the
audiotaped thoughts, or both. The content categories will depend on the cognitions
being sought. The ATSS procedure is very flexible in this way and also with regard
to the presentation of the thought-provoking stimuli. For example, although ATSS
generally employs pre-recorded audiotape presentations of the scenarios, the present
study required that the scenarios be read “live” due to the fact that the provocative
stimuli are food items, which were different for each subject. Bates, Campbell and
Burgess (1990) report satisfactory results with a more significant departure from
previous ATSS procedures. In their study, socially anxious and non-anxious subjects
watched a videotape depicting the social scenario rather than listening to an audio
tape, and stopped the tape themselves to think aloud as they experienced thoughts.
Concerning the ability of imagined situations to effectively evoke hedonic
responses in subjects, a study by Redd, Dadds, Futterman, Taylor and Bovjerg
(1993) suggests that this is eminently possible. The authors describe a study in
which mental images of chemotherapy induced nausea in chemotherapy patients
with a history of classically conditioned nausea to clinic stimuli. The procedure was
similar to an ATSS procedure insofar as subjects were instructed to immerse
67
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
themselves in the scenes and “speak in the present, really be there,” (Redd et al.,
1993, p. 630) and the format was open-ended. The subjects themselves described the
surroundings, instructed only to imagine being at the clinic. Not only did all subjects
with a history of anticipatory nausea report feeling ill, but two of them had to
discontinue the imagery due to overwhelming urges to vomit. Furthermore, the
ability of the imagined scenes to induce nausea persisted even when subjects were
explicitly instructed that the task was designed to evoke only feelings of relaxation.
In addition to the subjects’ self-reports they were also videotaped and the tapes later
analyzed for nausea behaviors, which the authors defined as “frown/grimace, verbal
complaint, clasping stomach, covering mouth, requesting help and sweating” (Redd
et al., 1993, p. 632). However, the authors report that the nausea behaviors were not
consistently present in the subjects who reported nausea.
Weddington, Miller and Sweet (1984) have also reported that thoughts associated
with chemotherapy provoked anticipatory nausea reactions in over half (9/17) of a
group of cancer patients. Dobkin, Zeichner, and Dickson-Pamell (1985) have
reported that cognitive stimuli and visual stimuli (such as visualization of needle
insertions) can stimulate anticipatory nausea reactions. Using the covert
sensitization technique, Clarke and Hayes (1983) found that subjects reduced both
their consumption and hedonic rating of a novel taste that was paired with only
nausea imagery.
68
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
A pilot study of conditioned food aversions conducted in this lab, employing the
stimulus-input portion of the ATSS procedure (i.e., the scenarios without the subject
response portions) and fMRI technology, indicates that the ATSS paradigm is suited
to such investigations. Using aversive and preferred food scenarios with a sample of
4 subjects, we found significant activation of the temporal region (which is involved
in the processing of negative emotional stimuli) during the aversive scenarios as
compared to preferred scenarios or rest periods (Chambers, Davison, Kunz and
Singh, unpublished data).
4.1.3. Hypotheses
In this experiment, male and female subjects listened and responded aloud with
their thoughts to scenarios involving target foods that the subjects had previously
reported as being aversive, disliked, or preferred by them. We hypothesized that
subjects would generate more cognitions characteristic of avoidance, disgust, nausea,
and negative evaluation of the scenario events, foods and characters during aversive
conditions relative to disliked or preferred conditions. Conversely, we hypothesized
that cognitions characteristic of positive evaluation of the scenario events, foods and
characters, and positive hedonic evaluation of foods would be higher during
preferred conditions relative to aversive or disliked conditions.
69
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2. Methods
4.2.1. Subjects
Thirty eight female and 6 male undergraduate and graduate students, ranging in
age from 18 to 39 years (mean age 21.33 years), were selected for participation from
the Department of Psychology at the University of Southern California. Subjects
were selected from approximately 250 two-page screening surveys that had been
distributed in undergraduate courses and among graduate students. The following
selection criteria were used: the subject reported a specific food aversion, completely
filled out the screening questionnaire, and the food items could be
reasonably/believably inserted into the ATSS scenarios.
4.2.2. Materials
The screening questionnaire required subjects to list two food items each for
aversive (stimulate a nauseous reaction), disliked (strong dislike but no nauseous
reaction), and liked (strong preference) categories. Additionally, the subject was
asked to report his/her age at the time the aversion/dislike/preference occurred and
his/her degree of familiarity with the food prior to the palatabilty decision/shift for
each food item. Discrete choice categories were used for the latter two response
items.
The procedure also included a post-experimental questionnaire. This questionnaire
consisted of discrete choice categories and open-ended comments sections which
70
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
asked subjects to report: (a) the detail of the visual images they generated; (b)
perceived realism of the scenarios; (c) emotional discomfort during aversive and
disliked scenarios; (d) anxiousness about expressing their thoughts aloud; (e) desire
to censor their thoughts or feelings while speaking out loud; (f) what and how long
ago they last ate something; and (g) their current state of hunger or satiety. Subjects
were also asked if they would like to participate in a similar fMRI study sometime in
the future. Subjects were also given an informed consent form to read and sign.
The testing environment consisted of a testing room and control room. Subjects sat
at a table in the testing room and listened to the ATSS scenarios read by the
experimenter from the control room using a Sony STR-DE325 Audio/Video Control
Center (Sony Broadcast and Professional Company, Park Ridge, New Jersey) and
Unidyne III Dynamic Microphone (Shure Brothers, Inc., Evanston, IE). Subjects’
responses to the scenarios were audio-taped. A PCM microphone (Realistic, Inc.,
Fort Worth, TX), placed on the table in front of the subjects fed to a Marantz
PMD420 stereo cassette recorder (Marantz America, Roselle, IE) in the control room
and recorded onto Maxell XL II 90-minute audio tapes (Maxell Corporation of
America, Fair Lawn, NJ). A tone, signaling subjects to alternately begin and cease
verbal responses to the scenarios was generated manually by the experimenter from
the control room.
71
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Copies of the screening questionnaire, post-experimental questionnaire, informed
consent form, and the ATSS instructions and scenario scripts are included as
Appendices A, B, C, and D, respectively.
4.2.3. Procedure
The testing procedure was identical for all subjects. The subject was seated at a
table in the testing room, given the informed consent form to read and sign, and then
briefly interviewed with regard to the subject’s screening questionnaire. The
interview served to familiarize the subject with the experimenter and surroundings,
and help the experimenter decide which food from each category was best suited for
use in the ATSS procedure and how to best present them. For example, if a subject
listed “tomatoes” as an aversive food item, the experimenter ascertained during the
interview in exactly what context the subject found tomatoes aversive, for example,
whether they were most aversive when served whole, sliced, or as part of a sauce. In
this way, the ATSS scenarios were tailored to each subject’s target food items.
Following the interview, the subject was read instructions for the ATSS procedure.
The subject was told to imagine that they were actually in the social situations
presented by the experimenter (as opposed to being in the testing room and merely
thinking about such situations) and to verbalize their thoughts as freely and
uncensored as possible. It was stressed that there were no right or wrong answers
and that anything that might come to the subject’s mind in response to the scenarios
72
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
was worthy of mention. It was also stressed that it was important to express as many
of their thoughts as they could. It was explained to the subject that his/her responses
would be audio taped. Previous research, both in Davison’s ATSS lab and in other
laboratories, confirms the ability of nearly all subjects to immerse themselves in the
procedure and provide data that can be reliably content-analyzed and that
demonstrate concurrent, predictive, and construct validity (for a review, see Davison
et al., 1997).
In order to familiarize the subject with the ATSS procedure, the experimenter went
into the control room and conducted a “practice scenario.” Sitting alone in the
testing room, the subject listened to a four-part scenario that depicted a social
situation, which involved the subject and a food item from the subject’s aversive
food category. The first segment of the scenario was read by the experimenter and
then a tone was presented, signaling the subject to verbalize his/her thoughts aloud
for 30 seconds. After 30 seconds had elapsed, the tone was presented again,
signaling the subject to cease talking, and then the second segment of the scenario
was read by the experimenter. The practice trial proceeded in this fashion until the
subject had responded to all four segments of the scenario. The experimenter
returned to the testing room to discuss any problems the subject might have had with
the practice scenario and answer any questions before beginning the data collection.
For example, if the subject was not very talkative during the response periods of the
scenario, the experimenter encouraged them to try and speak more. We found during
73
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
pilot studies in this lab that it is generally easier for subjects to verbalize about
aversive food items than preferred food items and it was for this reason that an
aversive food was used during the practice scenario. The aversive food item used in
the practice scenario was not the aversive food item to be used during data
collection.
After answering any of the subject’s questions and verifying that the subject
wished to continue with the experiment, the experimenter returned to the control
room. The data collection portion of the experiment consisted of three separate four-
segment scenarios depicting social situations in which the subject is confronted with
the prospect of having to consume a food item from their screening questionnaire.
Each scenario was presented three times, once for each food category (aversive,
disliked and preferred). So the subject responded to four segments of nine scenarios
for a total of 36 30-second responses. Each scenario proceeded in the same fashion
as the practice scenario with the exception that the subj ect was instructed to engage
in a simple counting task (“countdown from 300 by 3 s,” for example) between
scenarios. This task served to clear the subject’s mind of the scenario they had just
listened to prior to presentation of the next scenario. Following the presentation of
the final scenario, the subject was asked to fill out the post-experimental
questionnaire (described above) and thanked for their participation.
74
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.2.4. Scoring of A TSS data
Since there have been no previous investigations of this nature into human food
aversions, we devised a coding scheme composed of 7 categories of articulated
thoughts that we believed would identify the cognitions of interest to CTA research.
They are described in the following paragraphs.
1 . Escape/avoidance cognitions (ESC) were defined as expressions by the subject
that indicated a rejection or avoidance of the food. ESC responses included refusal
to eat, excuses for not eating, leaving the table, spitting food out, masking flavor, or
tactics designed to divert the attention of scenario characters from the subject’s
failure to eat the food. “I’ll just spit it into my napkin when she’s not looking,” for
example. The number of ESC responses was tallied for each 30-second segment.
ESC responses were assumed to reflect the cognitive correlate of behavioral
avoidance.
2. Negative expressions (NE) were defined as any negative thoughts expressed by
the subject in response to the scenarios. NE responses included negative evaluations
of the food, anxiety about the situation, negative anticipation of the food, anger
towards scenario characters or the self, or guilt at not being able to eat the food. “I
hate mussels,” for example. The number of NE responses was tallied for each 30-
second segment.
3. Disgust expressions (DE) were defined as any verbal expression indicating a
negative visceral reaction to the food. The range of possible DE responses is large
75
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
and includes a variety of “gut-level” utterances and words that reflect hedonic
evaluations of food. Examples of DE responses include “ewwww,” “blech,” “ick,”
“yuck,” “nasty,” “disgusting,” gross,” “foul,” etc. The number of DE responses was
tallied for each 30-second segment.
4. Verbal digressions (VD) were defined as a failure of the subject’s articulated
thoughts to address the target food within the context of the scenario. VD responses
were avoidance cognitions that exceeded the scope of the ESC category and were
differentiated from ESC by the subject going markedly off task during the response
segments. VD responses included expressions of disbelief in the scenario events or a
preponderance of thoughts only tangentially related to the scenario. “Well, I would
never even be in this situation,” for example. The occurrence of verbally digressive
responses was scored on a “yes/no” scale for each segment. Like ESC, VD was
assumed to reflect the cognitive correlate of behavioral avoidance, though possibly
more profound. That is, avoidance of the food to the point of avoiding even thoughts
of it. It is of interest to note that such off-task comments in previous ATSS
experiments have generally been regarded as a failure of the subject to adhere to the
procedural requirements of ATSS. In the present study, we construed such
articulated thoughts as relevant data.
5. Positive expressions (PE) were defined as any positive thoughts expressed by
the subject in response to the scenarios. PE responses included positive evaluations
of the food, enthusiasm about the situation, positive anticipation of the food, or
76
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
positive articulated thoughts about scenario characters or the self. “I love pizza,” for
example. The number of PE responses was tallied for each 30-second segment.
6. Expressions o f pleasure (PLE) were defined as any verbal expression indicating
a positive visceral reaction to the food. Like DE, the range of possible PLE
responses is large and includes a variety of “gut-level” utterances and words that
reflect hedonic evaluations of food. Examples of PLE responses include “mmmm,”
“yum,” “ooooh,” “tasty,” etc. The number of PLE responses was tallied for each 30-
second segment.
7 . Reports o f nausea (NAU) were defined as any report of a nauseous internal
physiologic state or expectation of this state in response to the target food. NAU
reports included feeling sick to the stomach, the urge to vomit, and retching and
tightening of the throat (or other physiological signs of nausea). NAU reports were
taken to be the putative sign of conditioned food aversion (Garcia, 1955). The
number of NAU reports were tallied for each segment.
Verbatim transcriptions were made of each subject’s articulated thoughts at each
response segment. The transcripts were prepared with all information about the
conditions removed. Two raters performed the task of coding the subjects’ responses
into the above categories. The raters were trained using data obtained from two
subjects during a pilot study for the present investigation. After agreement on the
classification scheme was achieved, each rater independently coded the articulated
thoughts of all subjects. Reliability analyses were conducted on each third of the
77
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
data, and retraining was conducted as necessary to control rater drift. The results for
each rater were averaged for the final analyses.
4.2.5. Statistical analyses
Subjects’ responses in each coding category for each segment were summed by
condition, resulting in 7 scores for each of the 3 conditions. These data were
analyzed with a two-factor (sex X condition) ANOVA with repeated measures on
condition. Combined data (both sexes together) were analyzed with a 1-factor
(condition) ANOVA with repeated measures on condition. A Newman-Keuls test
was used for paired comparisons among conditions for all articulated thoughts
categories. Intraclass correlation coefficients were used to assess inter-rater
reliability for frequently occurring categories and agreement ratios were used for
infrequently occurring categories. Agreement ratios were used because we have
found that the intraclass correlation coefficient as well as Kappa and Pearson product
moment correlation statistics are confounded by the large number of “zero value”
observations found among the infrequently occurring categories. Intraclass
correlation coefficients measured the consistency of agreement between the raters
while agreement ratios measured the absolute agreement between the raters.
78
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
4.3. Results
4.3.1. Reliability
The intraclass correlation coefficient showed that mean overall consistency in the
observations between the raters, for all thought categories across all segments, was
acceptable (a=0.89). However, consistency in the observations between the raters
for the frequently occurring categories (ESC, NE, DE, and PE) individually was
fairly low (a=0.69, a=0.49, a=0.71 and a=0.63, respectively). Mean absolute
agreement between the raters for the infrequently occurring categories (VD, PLE and
NAU) was very high (means = 0.98, 0.99 and 0.97, respectively).
4.3.2. Articulated thoughts
Tables 7, 8 and 9 show the means and standard errors, analyses of variance, and
Newman-Keuls pairwise comparisons for each of the articulated thought categories
across conditions, respectively. Figures 11 through 17 show the mean (± SE)
articulated thoughts across conditions for the ESC, NE, DE, VD, PE, PLE and NAU
categories, respectively. There were no main or interaction effects for males (n=6)
and females (n=38) on any of the articulated thought categories (Table 8) so they
were pooled for further analysis.
One-factor ANOVA indicated that there were significant interaction effects for the
mean frequency of articulated thoughts in all of the thought categories across
conditions (Table 8). Newman-Keuls pairwise comparison (Table 9) of conditions
79
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 7.
Means and standard errors of articulated thought categories for each condition.
Thought Category
Aversive Disliked Preferred
M (SE) M (SE) M (SE)
Escape/avoidance 28.318(2.109) 24.034 (2.084) 0.0114(0.0114)
Negative expression 21.068 (1.525) 18.034 (1.407) 1.59(0.27)
Disgust expression 9.95 (2.34) 6.977(1.692) 0.0568 (0.0334)
Verbal Digression 0.43 (0.0755) 0.27 (0.0679) 0.16(0.0558)
Positive expression 0.00 (0.00) 0.0795 (0.0486) 55.545 (3.3014)
Pleasure expression 0.00 (0.00) 0.00 (0.00) 1.67(0.51)
Nausea 4.94 (0.63) 1.57(0.37) 0.0341 (0.0252)
Table 8.
Results of 2-factor- and 1-factor-ANOVAs of articulated thought categories across conditions.
Thought Category
Male versus Female Pooled Data (all subjects)
F(2,84)
P
F(2,86)
P
Escape/avoidance 0.114 0.893 113.886 0.000**
Negative expression 0.605 0.549 115.740 0.000**
Disgust expression 2.588 0.081 16.336 0.000**
Verbal digression 1.275 0.285 5.456 0.006**
Positive expression 1.130 0.328 282.007 0.000**
Pleasure expression 0.703 0.498 10.591 0.000**
Nausea 1.754 0.179 48.365 0.000**
^Significance at the alpha = .01 level.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Table 9.
Results of Newman-Keuls pairwise comparisons of articulated thought categories by condition.
Condition
Thought Category
Condition Aversive Disliked Preferred
Escape/avoidance
Aversive — 0.037* 0.000**
Disliked 0.037* — 0.000**
Preferred 0.000** 0.000**
Negative expression
Aversive — 0.030* 0.000**
Disliked 0.030* — 0.000**
Preferred 0.000** 0.000** "
Disgust expression
Aversive 0.098 0.000**
Disliked 0.098 — 0.000**
Preferred 0.000** 0.000** —
Verbal Digression
Aversive - 0.059 0.004**
Disliked 0.059 — 0.174
Preferred 0.004** 0.174 —
Positive expression
Aversive — 0.977 0.000**
Disliked 0.977 — 0.000**
Preferred 0.000** 0.000** —
Pleasure expression
Aversive — 1.000 0.001**
Disliked 1.000 — 0.000**
Preferred 0.001** 0.000** —
Nausea
Aversive — 0.000** 0.000**
Disliked 0.000** — 0.004**
Preferred 0.000** 0.004** -
♦Significance at the alpha=.05 level. ** Significance at the alpha = .01 level.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Mean Escape/Avoidance Repsonses by Condition
40-i
Disliked Aversive
Condition
Figure 11. Mean (+/- SE) ESC responses by condition. A, D,
and P indicate a significance difference from the aversive, disliked
or preferred conditions, respectively. There were virtually no ESC
responses during the preferred condition.
Mean Negative Expression by Condition
30-.
Aversive Disliked Preferred
Condition
Figure 12. Mean (+/- SE) NE responses by condition. A, D,
and P indicate a significance difference from the aversive,
disliked or preferred conditions, respectively.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Mean Disgust Expressions by Condition
15-|
Aversive Disliked
Condition
Figure 13. Mean (+/- SE) DE responses by condition. A, D,
and P indicate a significance difference from the aversive,
disliked or preferred conditions, respectively. There were
virtually no DE responses during the preferred conditions.
Mean Verbal Digressions by Condition
0.75-t
Disliked Preferred Aversive
Condition
Figure 14. Mean (+/- SE) VD responses by condition. A, D,
and P indicate a significance difference from the aversive,
disliked or preferred conditions, respective^. The disliked
condition did not differ significantly from the other conditions.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Mean Positive Expressions by Condition
■ B
M
a
75-
50-
25-
AD
Aversive Disliked
Condition
Preferred
Figure IS. Mean (+/- SE) PE responses by condition. A, D,
and P indicate a significance difference from the aversive,
disliked or preferred conditions, respectively. There were
virtually no PE responses during the aversive and disliked
conditions.
Mean Pleasure Expressions by Condition
• o '
1
= J ■
o O
o b
W a
c a 2
Aversive Disliked
Condition
Preferred
Figure 16. Mean (+/- SE) PLE responses by condition. A, D,
and P indicate a significance difference from the aversive,
disliked or preferred conditions, respectively. There were
virtually no PLE responses during the aversive and disliked
conditions.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Mean Nausea Reports by Condition
7.5n
Condition
Figure 17. Mean (+/- SE) NAU responses by condition. A, D,
and P indicate a significance difference from the aversive,
disliked or preferred conditions, respectively. There were
virtually no NAU responses during the preferred condition.
indicated that the aversive, disliked and preferred conditions differed significantly
from each other for the ESC, NE and NAU categories (Figures 11,12 and 17,
respectively). The highest frequency of articulated thoughts occurred in the aversive
condition and the lowest frequency in the preferred condition, with the disliked
condition in between. For the DE, PE and PLE categories, the aversive and disliked
conditions did not differ significantly from each other, however both of these
conditions differed significantly from the preferred condition. The highest frequency
of articulated thoughts occurred in the aversive and disliked conditions and the
lowest frequency occurred in the preferred condition for DE (Figure 13). The
opposite was true for the PE and PLE categories, with the highest frequency of
articulated thoughts occurring in the preferred condition and the lowest frequency
85
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
occurring during the aversive and disliked conditions (Figures 15 and 16). For the
VD category, the aversive and preferred conditions differed significantly, with the
highest frequency of articulated thoughts occurring in the aversive condition and the
lowest frequency occurring in the preferred condition. The disliked condition did not
differ significantly from either the aversive or preferred conditions (Figure 14).
4.4. Discussion
The results ofthe present study indicate that subj ects generated more cognitions
characteristic of avoidance, nausea, and negative hedonic evaluation of a food (as
well as negative evaluation of events and characters associated with that food) in
response to imagined situations involving a food they have previously identified as
aversive than they did in response to the same imagined situations involving a food
they have previously identified as preferred. Conversely, subjects generated more
cognitions characteristic of positive hedonic evaluation of the preferred food (as well
as positive evaluation of events and characters associated with that food) than they
did for the aversive food under the same test conditions.
The results with respect to foods disliked by these subjects are less clear. While
the mean frequency of ESC, NE and NAU responses significantly increased during
aversive relative to disliked conditions, the mean frequency of DE and VD responses
did not. Similarly, the PE and PLE responses did not differ significantly between the
aversive and disliked conditions. The similar frequency of DE responses among the
86
Reproduced with permission o fth e copyright owner. Further reproduction prohibited without permission.
aversive and disliked conditions suggests that the hedonic evaluation of foods that
have caused illness and foods that are simply disliked is the same. That is, subjects
may rate disliked and aversive foods as equally unpleasant regardless of whether or
not they were reported to be associated with nausea and therefore are presumed to
have been paired with illness. This hypothesis also accounts for the similarity of the
PE and PLE responses between the aversive and disliked conditions. There were
virtually no PE and PLE responses generated during these conditions. What is
interesting is that subjects showed increased avoidance of the nausea-associated
foods.
It is also possible that the similar frequency of DE responses observed among the
aversive and disliked conditions in this experiment is due to the colloquial use of
disgust expressions in common English, especially as relates to the discussion of
food items. There is a relative poverty in the English language of expressions used
to differentiate foods that make one feel nauseous from foods that one dislikes. With
the exception of specific indicators of nausea, such as “Til puke if I eat that” (which
were coded in the NAU category and were significantly higher in aversive
conditions) subjects must rely on the same general expressions to describe their
hedonic evaluation of both aversive and disliked foods. For example, a subject may
describe both pork chops (which presumably made them ill) and lima beans (which
they dislike strongly) as “disgusting,” with the difference being that they also report
the urge to vomit in response to the image of pork chops.
87
Reproduced with permission o fth e copyright owner. Further reproduction prohibited without permission.
The similar frequency of VD responses among the aversive and disliked conditions
may be due to an inability of the subjects to adhere to the ATSS procedure. As
described in the Scoring o f ATSS data (4.1.4) section above, such off-task
occurrences have not generally been regarded as relevant data in ATSS experiments.
However, it is interesting to note that there were significantly fewer VD responses in
the preferred condition relative to the aversive condition.
5. DIRECTIONS FOR FUTURE RESEARCH
5.1. Animal studies
As we have seen, the evidence which indicates that estradiol exerts its effects on
feeding through a satiety mechanism is mounting. However, the taste reactivity
methodology used in these experiments may need to be modified to facilitate further
inquiries in this area since brief infusions of small amounts of the CS may not fully
activate the satiety mechanism. As described in the Aversion versus avoidance
(1.1.3) section above, Grill and Berridge (1985) have shown that rats will decrease
their display of digestive responses and increase their display of aversive responses
as they become satiated during a meal. So it is possible that estradiol-treated rats
might show aversive orofacial responses to longer/larger infusions of the CS. If this
were the case, time courses of CS infusions could be given that might further
differentiate between the behavioral effects of LiCl and estradiol. That is, the rats’ .
88
Reproduced with permission o fth e copyright owner. Further reproduction prohibited without permission.
responses could be examined over a longer infusion period to see if there is a
hedonic shift from beginning to end.
These issues underscore the idea of lattice hierarchies described in the Aversion
versus avoidance (1.1.3) section of this thesis. Indeed, taste reactivity and bottle
tests are not the only behavioral measures that suffer from such ambiguity. Strieker
and Verbalis (1991) report that rats display the behavioral sequence of satiety
(grooming followed by inactivity) following treatment with LiCl. It may be that the
best approach in comparing estradiol and LiCl will be to examine the behavioral
constellation induced by each. There are several other common methodologies in the
CTA literature that might be suitable.
The defensive burying response has been used by Parker (1988) to demonstrate
that amphetamine produces an avoidance but not an aversion when paired with a
sucrose solution, indicated by the fact that animals avoided the solution but did not
bury it. In the same experiment she found that animals not only avoided a LiCl-
paired solution but also buried it, which is consistent with the known aversive effects
of LiCl. Defensive burying is a species-specific response of the rat to intrinsically
noxious stimuli or conditioned noxious stimuli (Bowers et al., 1992). This method
offers two improvements over previous methods. First, it circumvents the “forced
choice” problems which may be associated with the taste reactivity test previously
described. Second, it tells us definitively what the animal’s evaluation of the tastant
is. Although we still don’t know why the animal might not bury the target, we are
89
Reproduced with permission o fth e copyright owner. Further reproduction prohibited without permission.
sure of why it does. As discussed in the Estrogen and cholecystokinin (2.1.3) section
above, Bowers et al. (1992) report that rats will defensively bury a food that has been
paired with exogenous CCK administration. However, we do not believe anyone has
yet examined the effects of estradiol on this response.
Also from the pharmacological literature comes the conditioned place test. In this
test, the experimental substance is paired with particular environmental cues
(Jorenby, Steinpreis, Sherman and Baker, 1990). It is believed to measure the
approach or avoidance behavior that is displayed toward an environment that has
been paired with the physiological effects of a chemical (Costello, Carlson, Glick
and Bryda, 1989). Although it is generally used to examine the animal’s hedonic
evaluation of a psychoactive drug, it has also proved effective when used with
aversive agents used in the CTA paradigm such as radiation and LiCl (Fudula, Teoh,
Edgar and Iwamoto, 1985).
The conditioned place preference technique possesses advantages similar to those
described for the defensive burying procedure. Another interesting aspect of this test
is that the conditioned and non-conditioned chambers can be created in different
arrangements from both being equally neutral to the conditioned chamber being
inherently aversive to the animal (Clarke and Fibiger, 1987). For example, the
conditioned chamber may be brightly illuminated while the non-conditioned
chamber is left dark. The implications for ascertaining the degree of an animal’s
distaste or preference for a flavor are obvious. To our knowledge, the conditioned
90
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
place preference procedure has not yet been applied to the study of estrogen and
satiety.
Satiety may also be distinguished from aversion by examining meal initiation and
termination (Booth, 1977,1985; Smith et al., 1979). If estrogen produces
conditioned satiety then we might expect estrogen-treated animals to initiate meals
and terminate them rapidly since it is the animal’s evaluation of the satiating effects
of the solution, not its palatability, which has changed. However, if estrogen
produces aversion then we might expect the animal will fail to initiate a meal. Since
meal initiation must be distinguished from exploratory sampling of the solution, and
the total volume consumed may be similar for both situations, lickometers may be of
use. Lickometers provide information about the frequency with which the animal
makes contact with the spout and should be able to distinguish between tentative
sampling and a robust but abbreviated bout of drinking.
In conclusion, we believe that the evaluation of multiple conditioned behavioral
measures may be the key to increasing our understanding of the effects of estrogen
on feeding behavior in the rat. We are more likely to uncover meaningful
differences in the constellation of behaviors evoked by estrogen and LiCl than we are
in the individual measures used in the present experiments.
91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5.2. Human studies
The sensitivity of the ATSS procedure to the aversive (nausea reports) and
avoidance (escape cognitions) behaviors elicited by foods that are reported by human
subjects to have previously been associated with illness represents a significant
development in the study of human CTA. First, the flexibility of the ATSS scenarios
allows for any number of refinements to the procedure used in the present study as
well as a seemingly endless array of more divergent inquiries. For example, the
finding of Clarke and Hayes (1983) that subjects reduced their consumption and
hedonic rating of a novel taste paired only with nausea imagery suggests that the
ATSS procedure may even be used to explore the cognitions involved in the
acquisition of human food aversions. The ATSS procedure may also be useful in
identifying and treating the cognitions associated with eating disorders.
The stimulus presentation portion of the ATSS procedure has implications for the
use of brain imaging techniques as well. The present study shows that the ATSS
scenarios can reliably induce the physiological symptoms of nausea (as verified by
subject reports) and is thus well-suited to the study of the neurological substrates of
food aversion in humans. Indeed, we are currently developing an ATSS
methodology for use with fMRI, in which both the stimulus presentation and subject
response portions of the scenarios will be used with live human subjects while in the
MRI machine.
92
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
REFERENCES
Adachi, A., Kobashi, M., Miyoshi, N. & Tsukamoto, G. (1991). Chemosensitive
neurons in the area postrema of the rat and their possible functions. Brain Research
Bulletin. 26: 137-40.
Bates, G. W., Campbell, I. M. & Burgess, P. M. (1990). Assessment of
articulated thoughts in social anxiety: Modification of the ATSS procedure. British
Journal of Clinical Psychology. 29: 91-8.
Bernstein, I. L. (1978). Learned taste aversions in children receiving
chemotherapy. Science. 200: 1302-3.
; Bernstein, I. L. (1985). Learned food aversions in the progression of cancer and
its treatment. Annals of the New York Academy of Science. 443: 348-56.
Bernstein, I. L. & Borson, S. (1986). Learned food aversion: A component of
anorexia syndromes. Psychological Review. 93(4): 462-72.
Bernstein, I. L., Courtney, L. & Braget, D. J. (1986). Estrogens and the Leydig
LTW(m) tumor syndrome: Anorexia and diet aversions attenuated by area postrema
lesions. Physiology and Behavior. 38: 159-63.
Berridge, K. C., Flynn, F. W., Schulkin, J. & Grill, H. J. (1984). Sodium
depletion enhances salt palatability in rats. Behavioral Neuroscience. 98: 652-660.
Blaustein, J. & Wade, G. (1976). Ovarian influences on the meal pattern of
female rats. Physiology and Behavior. 17: 201-8.
Bless, E. P., McGinnis, K. A., Mitchell, A. L., Hartwell, A. & Mitchell, J. B.
(1997). The effects of gonadal steroids on brain stimulation reward in female rats.
Behavioral Brain Research. 82: 235-44.
Boland, F. J., Mellor, C. & Revusky, S. (1978). Chemical aversion treatment of
alcoholism: Lithium as the aversive agent. Behavioural Research and Therapy. 16:
401-6.
Bonavera, J. J., Dube, M. G., Kalra, P. S. & Kalra, S. P. (1994). Anorectic effects
of estrogen may be mediated by decreased neuropeptide-Y release in the
hypothalamic paraventricular nucleus. Endocrinology. 134(6): 2367-70.
93
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Booth, D. A. (1977). Satiety and appetite are conditioned reactions.
Psychosomatic Medicine, 39(2): 76-81.
Booth, D. A. (1985). Food-conditioned eating preferences and aversions with
interoceptive elements: Conditioned appetites and satieties. Annals of the New York
Academy of Sciences, 443: 22-41.
Borison, H. L. (1989). Area postrema: Chemoreceptor circumventricular organ of
the medulla oblongata. Progress in Neurobiology, 32: 351-90.
Borison, H. L. & Wang, S. C. (1953). Physiology and pharmacology of vomiting.
Pharmacological Review. 5: 193-230.
Bowers, R. L., Herzog, C. D., Stone, E. H. & Dionne, T. J. (1992). Defensive
burying following injections of cholecystokinin, bombesin, and LiCl in rats.
Physiology and Behavior, 51: 969-72.
Brizzee, K. R., Ordy, J. M. & Mehler, W. R. (1980). Effect of ablation of area
postrema on frequency and latency of motion sickness-induced emesis in the squirrel
monkey. Physiology and Behavior. 24: 849-53.
Butera, P. C. & Beikirch, R. J. (1989). Central implants of diluted estradiol:
Independent effects on ingestive and reproductive behaviors of ovariectomized rats.
Brain Research, 491: 266-73.
Butera, P. C., Campbell, R. B. & Bradway, D. M. (1993). Antagonism of
estrogenic effects on feeding behavior by central implants of anisomycin. Brain
Research, 624: 354-6.
Butera, P. C., Willard, D. M. & Raymond, S. A. (1992). Effects of PVN lesions
on the responsiveness of female rats to estradiol. Brain Research. 576: 304-10.
Carter, D. A., Williams, T. D. M. & Lightman, S. L. (1986). A sex difference in
endogenous opioid regulation of the posterior pituitary response to stress in the rat.
Journal of Endocrinology, 111: 239-44.
Chamberlain, J. & Haaga, D. A. F. (1999). Convergent validity of cognitive
assessment methods. Behavior Modification, 23(2): 294-315.
Chambers, K. C. (1990). A neural model for conditioned taste aversions. Annual
Review of Neuroscience. 13: 373-85.
94
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Chambers, K. C. & Bernstein, I. L. (1995). Conditioned flavor aversions. In R.
L. Doty (Ed.), Handbook o f Olfaction and Gustation, New York: Marcel Dekker,
Inc.
Chambers, K. C. & Bernstein, I. L. (in press). Conditioned taste aversions. In R.
L. Doty (Ed.), Handbook o f Olfaction and Taste. New York: Marcel Dekker, Inc.
Chambers, K. C., Hayes, U. L., Kunz, R. D. & Wang, Y. (1998). Conditions
under which estradiol induces conditioned taste aversions and accelerates extinction
of LiCl-induced aversions. Society for Neuroscience Abstracts. 24: 941.
Chambers, K. C., Hayes, U. L., Kunz, R. D. & Wang, Y. (1999). Pre-exposure to
estradiol before acquisition of a LiCl-induced conditioned taste aversion accelerates
extinction and decreases c-fos expression in the area postrema. Society for
Neuroscience Abstracts. 25: 1358.
Chambers, K. C., Kunz, R. D., Makena, A. & Wang, Y. Cooling lesions of the
area postrema do not attenuate estradiol-induced conditioned taste aversions and
estradiol does not increase c-fos expression in the area postrema. (unpublished
manuscript)
Chambers, K. C., Makena, A., Kunz, R. D. & Wang, Y. (in press). Estradiol-
induced conditioned taste avoidance is not mediated by the area postrema. Society
for Neuroscience Abstracts. 26.
Chambers, K. C. & Yuan D. L. (1999a). Estradiol accelerates extinction of a
conditioned taste aversion in female and male rats. Hormones and Behavior, 36(1):
1-16.
Chambers, K. C. & Yuan D. L. (1999b). Estradiol accelerates extinction of LiCl-
induced conditioned taste aversions through its illness-associated properties.
Hormones and Behavior. 36(3): 287-98.
Clarke, P. B. S. & Fibiger, H. C. (1987). Apparent absence of nicotine-induced
conditioned place preference in rats. Psychopharmacology. 92: 84-8.
Clarke, J. C. & Hayes, K. (1984). Covert sensitization, stimulus relevance and
the equipotentiality premise. Behavioral Research Therapy. 22(4): 451-4.
Conde, G. L., Herbison, A. E., Femandez-Galaz, C. & Bicknell, R. J. (1996).
Estrogen uncouples noradrenergic activation of fos expression in the female rat
preoptic area. Brain Research. 735: 197-207.
95
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Costello, N. L., Carlson, J. N., Glick, S. D. & Bryda, M. (1989). Dose-dependent
and baseline-dependent conditioning with d-amphetamine in the place conditioning
paradigm. Psvchopharmacology. 99: 244-7.
Cross-Mellor, S. K., Kent, W. D. T., Ossenkopp, K. P. & Kavaliers, M. (1999).
Differential effects of lipopolysaccharide and cholecystokinin on sucrose intake and
palatability. American Journal of Physiology. 277: R705-15.
Dagnault, A. & Richard, D. (1994). Lesions of the hypothalamic paraventricular
nuclei do not prevent the effect of estradiol on energy and fat balance. American
Journal of Physiology. 267(1 Pt. 1): E32-8.
Dagnault, A. & Richard, D. (1997). Involvement of the medial preoptic area in
the anorectic effects of estrogens. American Journal of Physiology. 272(1 Pt. 2):
R311-7.
Davison, G. C., Robins, C. & Johnson, M. (1983). Articulated thoughts during
simulated situations: A paradigm for studying cognition in emotion and behavior.
Cognitive Therapy and Research. 7: 17-40.
Davison, G. C., Vogel, R. S. & Coffman, S. G. (1997). Think-aloud approaches
to cognitive assessment and the articulated thoughts in simulated situations
paradigm. Journal of Consulting and Clinical Psychology. 65(6): 950-8.
De Silva, P. & Rachman, S. (1987). Human food aversions: Nature and
acquisition. Behaviour Research and Therapy. 25(6): 457-68.
Deutsch, J. A. & Hardy, W. T. (1977). Cholecystokinin produces bait shyness in
rats. Nature. 266: 196.
Disshon, K. A. & Dluzen, D. E. (1997). Estrogen as a neuromodulator of MPTP-
induced neurotoxicity: Effects upon striatal dopamine release. Brain Research. 764:
9-16.
Dobkin, P., Zeichner, A. & Dickson-Pamell, B. (1985). Concomitants of
anticipatory nausea and emesis in cancer patients in chemotherapy. Psychological
Reports. 56: 671-6.
Dorso, V. A. C. & Paimeto-Neto, J. (1992). Lithium effects on estrogen-induced
dopaminergic supersensitivity in rats. Brain Research Bulletin. 29: 239-41.
96
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Dulawa, S. C. & Vanderweele, D. A. (1994). Cholecystokinin and estradiol
synergystically potentiate satiety in rats. Peptides, 15(5): 913-8.
Dunn, L. T. & Everitt, B. J. (1988). Double dissociations of the effects of
amygdala and insular cortex lesions on conditioned taste aversion, passive
avoidance, and neophobia in the rat using the excitotoxin ibotenic acid. Behavioral
Neuroscience, 102: 3-23.
Eckel, L. A. & Ossenkopp, K-P. (1996). Area postrema mediates the formation
of rapid, conditioned palatability shifts in lithium treated rats. Behavioral
Neuroscience. 110: 202-12.
Eckersell, C. B. & Micevych, P. E. (1997). Opiate receptors modulate estrogen-
induced cholecystokinin and tachykinin but not enkephalin messenger mRNA levels
in the limbic system and hypothalamus. Neuroscience, 80(2): 473-85.
Fink, H., Rex, A., Voits, M. & Voigt, J. P. (1998). Major biologic actions of
CCK - A critical evaluation of research findings. Experimental Brain Research,
123: 77-83.
Fudala, P. J., Teoh, K. W. & Iwamoto, E. T. (1985). Pharmacologic
characterization of nicotine-induced conditioned place preference. Pharmacology,
Biochemistry and Behavior. 22: 237-41.
Ganesan, R. & Simpkins, J. W. (1990). The role of conditioned taste aversion in
the suppression of food intake by estradiol. Physiology and Behavior, 48: 647-52.
Ganesan, R. & Simpkins, J. W. (1991). Conditioned taste aversion induced by
estradiol pellets. Physiology and Behavior. 50: 849-52.
Garb, J. L. & Stunkard, A. J. (1974). Taste aversions in man. American Journal
of Psychiatry. 131: 1204-07.
Garcia, J. (1955). Conditioned aversion to saccharine resulting from exposure to
gamma radiation. Science. 122: 157-58.
Garcia, J., Hankins, W. G. & Rusiniak, K. W. (1974). Behavioral regulation of
the milieu interne in man and rat. Science. 185: 824-31.
Geary, N., Trace, D., McEwan, B. & Smith G. P. (1994). Cyclic estradiol
replacement increases the satiety effect of CCK-8 in ovariectomized rats.
Physiology and Behavior. 56(2): 281-9.
97
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Geiseknan, P. J. (1996). Control of food intake. Obesity. 25(4): 815-29.
Goodman, L. S. & Gilman, A., (Eds.). (1975). The Pharmacological Basis of
Therapeutics. New York: Macmillan Publishing Company, Inc.
Graillon, A., Barr, R. G., Young, S. N., Wright, J. H. & Hendricks, L. A. (1997).
Differential response to intraoral sucrose, quinine and com oil in crying newborns.
Physiology and Behavior. 62(2): 317-25.
Grant, V. L. (1987). Do conditioned taste aversions result from activation of
emetic mechanisms? Psvchopharmacologv. 93(41: 405-15.
Grigson, P. S. (1997). Conditioned taste aversions and drugs of abuse: A
reinterpretation. Behavioral Neuroscience. 111(11: 129-36.
Grill, H. J. & Berridge, K. C. (1985). Taste reactivity as a measure of the neural
control of palatability. Progress in Psychobiology and Physiological Psychology, 11:
1-61.
Grill, H. J. & Norgren, R. (1978). The taste reactivity test. I. Mimetic responses
to gustatory stimuli in neurologically normal rats. Brain Research. 143: 263-79.
Gu, Y., Gonzalez, M. F., Chen, D. Y. & Deutsch, J. A. (1993). Expression of c-
fos in brain subcortical structures in response to nauseant lithium chloride and
osmotic pressure in rats. Neuroscience Letters. 157: 49-52.
Gustavson, C. R., Gustavson, J. C., Young, J. K., Pumariega, A. J. & Nicolaus, L.
K. (1989). Estrogen induced malaise. In J. M. Lakoski, J. R. Perez-Polo & D. K.
Rassin (Eds.), Neural control of reproductive function (pp. 501-23). New York:
Alan R. Liss, Inc.
Gustavson, C. R. & Gustavson, J. C. (1985). Predation control using conditioned
food aversion methodology: Theory, practice and implications. Annals of the New
York Academy of Science. 443: 348-56.
Hermann, G. E., Kohlerman, N. J. & Rogers, R. C. (1983). Hepatic-vagal and
gustatory afferents in the brainstem of the rat. Journal of the Autonomic Nervous
System. 9(2-3): 477-95.
98
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Holland, K. L., Norby, L. A. & Micevych, P. E. (1998). Peripubertal ontogeny
and estrogen stimulation of cholecystokinin and preproenkephalin mRNA in the rat
hypothalamus and limbic system. The Journal of Comparative Neurology. 392: 48-
57.
Hunt, T. & Amit, Z. (1987).. Conditioned taste aversion induced by self
administered drugs: Paradox revisited. Neuroscience and Behavioral Reviews. 11:
107-30.
Jorenby, D. E., Steinpreis, R. E., Sherman, J. E. & Baker, T. B. (1990). Aversion
instead of preference learning indicated by nicotine place conditioning in rats.
Psvchopharmacology. 101: 533-38.
King, J. L., Grigson, P. S., Grill, H. J. & Flanagan-Cato, L. M. (1997). Estrogen
pairing does not condition a taste aversion in female rats. Society for Neuroscience
Abstracts. 23: 1347.
Kritzer, M. F. & Kohama, S. G. (1999). Ovarian hormones differentially
influence immunoreactivity for dopamine p-hydroxylase, choline acetyltransferase,
and serotonin in the dorsolateral prefirontal cortex of adult rhesus monkeys. The
Journal of Comparative Neurology. 409: 438-51.
Kunz, R. D. (1997). The effect of 30mm estradiol implants in producing
conditioned taste aversion in female Fischer 344 rats. Unpublished manuscript,
University of Southern California, Los Angeles.
Lammers, C. H., D’Souza, U., Qin, Z. H., Lee, S. H., Yajima, S. & Mouradian,
M. M. (1999). Regulation of striatal dopamine receptors by estrogen. Synapse. 34:
222-7.
Leibowitz, S. F., Weiss, G. F. & Suh, J. S. (1990). Medial hypothalamic nuclei
mediate serotonin’s inhibitory effect on feeding behavior. Pharmacology.
Biochemistry, and Behavior. 37: 735-42.
Levine, A. S. & Billington, C. J. (1997). Why do we eat? A neural systems
approach. Annual Review of Nutrition. 17: 597-619.
Linden-Hirschberg, A. (1998). Hormonal regulation of appetite and food intake.
Annals of Medicine. 30: 7-20.
99
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Logue, A.W., Logue, K. R. & Strauss, K. E. (1983). The acquisition of taste
aversion in humans with eating and drinking disorders. Behavioural Research and
Therapy. 21: 275-89.
Logue, A.W., Ophir, I. & Strauss, K. E. (1983). The acquisition of taste aversions
in humans. Behavioural Research and Therapy. 19: 319-33.
Martin, J. R., Cheng, F. Y. & Novin, D. (1978). Acquisition of learned taste
aversion following bilateral subdiaphragmatic vagotomy in rats. Physiology and
Behavior. 21: 13-7.
Mellor, C. S. & White, H. P. (1978). Taste aversion to alcoholic beverages
conditioned by motion sickness. American Journal of Psychiatry. 135:125-6.
Melton, P. A. & Riley, A. L. (1994). Receptor mediation of the stimulus
properties of cholecystokinin. Pharmacology. Biochemistry and Behavior. 48(1):
275-9.
Misevych, P. E., Eckersell, C. B., Brecha, N. & Holland, K. L. (1997). Estrogen
modulation of opioid and cholecystokinin systems in the limbic-hypothalamic
circuit. Brain Research Bulletin. 44(4): 335-43.
Mook, D. G. (1996). Motivation: The organization o f action. New York: W.W.
Norton and Company, Inc.
Moore, B. O. & Deutsch, J. A. (1985). An antiemetic is antidotal to the satiety
effects of cholecystokinin. Nature. 315: 321.
Mordes, J. P., Longscope, C., Flatt, J. P., MacLean, D. B. & Rossini, A. A.
(1984). The rat LTW(m) Leydig cell tumor: Cancer anorexia due to estrogen.
Endocrinology. 115: 167-73.
Morley, J. E. (1981). The endocrinology of the opiates and opioid peptides.
Metabolism. 30(2): 195-209.
Morley, J. E., Levine, A. S., Grace, M., Kneip, J. & Gosnell, B. A. (1984). The
effect of ovariectomy, estradiol and progesterone on opioid modulation of feeding.
Physiology and Behavior. 33: 237-41.
100
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Moms, R., Frey, S., Kasambir, T. & Petrides, M. (1999). Ibotenic acid lesions of
the basolateral, but not the central, amygdala interfere with conditioned taste
aversion: Evidence from a combined behavioral and anatomical tract-tracing
investigation. Behavioral Neuroscience. 113: 291-302.
Niijima, A. & Yamamoto, T. (1994). The effects of LiCl on the activity of the
afferent nerve fibers from the abdominal visceral organs in the rat. Brain Research
Bulletin. 35: 141-5.
Ogawa, H., Hayama, T. & Ito, S. (1987). Response properties of the parabrachio-
thalamic taste and mechanoreceptive neurons in rats. Experimental Brain Research.
68: 449-57.
Oro, A. E., Simerly, R. B. & Swanson, L. W. (1988). Estrous cycle variations in
levels of cholecystokinin immunoreactivity within cells of three interconnected
sexually dimorphic forebrain nuclei. Neuroendocrinology. 47: 225-35.
Ossenkopp, K. P., Rabi, Y. J. & Eckel, L. A. (1996). Oestradiol-induced taste
avoidance is the result of a conditioned palatability shift. Neuroreport. 7(15-17):
2777-80.
Parker, L. A. (1988). Defensive burying of flavors paired with lithium but not
amphetamine. Psvchopharmacology, 96: 250-2.
Parker, L. A. (1995). Rewarding drugs produce taste avoidance, but not taste
aversion. Neuroscience and Behavioral Reviews. 19(1): 143-51.
Parker, L. A. & MacLeod, K. B. (1991). Chin rub CRs may reflect conditioned
sickness elicited by a lithium-paired sucrose solution. Pharmacology. Biochemistry
and Behavior. 40: 983-6.
Pelchat, M. L., Grill, H. J., Rozin, P. & Jacobs, J. (1983). Quality of acquired
responses to tastes by rattus norvegicus depends on type of associated discomfort.
Journal of Comparative Psychology. 97(2): 140-53.
Perl, E., Shay, U., Hamburger, R. & Steiner, J. (1992). Taste- and odor-
reactivity in elderly demented patients. Chemical Senses. 17: 779-94.
101
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Pfaus, J. G., Marcangione, C., Smith, W. J., Manitt, C. & Abillamaa, H. (1996).
Differential induction of fos in the female rat brain following different amounts of
vaginocervical stimulation: Modulation by steroid hormones. Brain Research. 741:
314-30.
Piva, F., Limonta, P., Dondi, D., Pimpinelli, F., Martini, L. & Maggi, R. (1995).
Effects of steroids on the brain opioid system. Journal of Steroid Biochemistry and
Molecular Biology. 53(1-6): 343-8.
Qian, M., Johnson, A. E., Kallstrom, L., Carrer, H. & Sodersten, P. (1997).
Cholecystokinin, dopamine D2 and n-methyl-d-aspartate binding sites in the nucleus
of the solitary tract of the rat: Possible relationship to ingestive behavior.
Neuroscience. 77(41: 1077-89.
Redd, W. H., Dadds, M. R., Futterman, A. D., Taylor, K. L. & Bovbjerg, D. H.
(1993). Nausea induced by mental images of chemotherapy. Cancer. 72(2): 629-36.
Ritter, S., McGlone, J. J. & Kelley, K. W. (1980). Absence of lithium-induced
taste aversion after area postrema lesion. Brain Research. 201: 501-6.
Sakai, N. & Yamamoto, T. (1999). Possible routes of visceral information in the
rat brain in formation of conditioned taste aversion. Neuroscience Research. 35: 53-
61.
Shapiro, R. E. & Miselis, R. R. (1985). The central neural connections of the area
postrema of the rat. The Journal of Comparative Neurology. 234: 344-64.
Simerly, R. B., Chang, C., Muramatsu, M. & Swanson, L. W. (1990).
Distribution of androgen and estrogen receptor mRNA-containing cells in the rat
brain: An in situ hybridization study. The Journal of Comparative Neurology.
294(1): 76-95.
Smith, G. P. & Gibbs, J. (1979). Postprandial satiety. Progress in Psychobiology
and Physiological Psychology. 8: 179-242.
Soussignan, R. & Sehaal, B. (1996). Children’s facial responsiveness to odors:
Influences of hedonic valence of odor, gender, age, and social presence.
Developmental Psychology. 32(2): 367-79.
102
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Spector, A. C., Breslin, P. & Grill, H. J. (1988). Taste Reactivity as a Dependent
Measure of the Rapid Formation of Conditioned Taste Aversion: A Tool for the
Neural Analysis of Taste-Visceral Associations. Behavioral Neuroscience. 102(6):
942-952.
Steiner, J. E. (1974). Discussion paper: Innate, discriminative human facial
expressions to taste and smell stimulation. Annals of the New York Academy of
Sciences. 237: 229-33.
Strieker, E. M. & Verbalis, J. G. (1991). Caloric and noncaloric controls of food
intake. Brain Research Bulletin. 27: 299-303.
Swank, M. W. & Bernstein, I. L. (1994). C-Fos induction in response to a
conditioned stimulus after single trial taste aversion learning. Brain Research. 636:
202- 8 .
Swank, M. W., Schafe, G. E. & Bernstein, I. L. (1995). C-Fos induction in
response to taste stimuli previously paired with amphetamine or LiCl during taste
aversion learning. Brain Research. 673: 251-61.
Tetel, M. J., Getzinger, M. J. & Blaustein, J. D. (1994). Estradiol and
progesterone influence the response of ventromedial hypothalamic neurons to tactile
stimuli associated with female reproduction. Brain Research. 646: 267-72.
Travers, J. B. & Norgren, R. (1983). Afferent projections to the oral motor nuclei
in the rat. Journal of Comparative Neurology. 220: 280-98.
Wade, G. N. (1972). Gonadal hormones and behavioral regulation of body
weight. Physiology and Behavior. 8: 523-34.
Wade, G. N. & Gray, J. M. (1979). Gonadal effects on food intake and adiposity:
a metabolic hypothesis. Physiology and Behavior. 22: 583-93.
Wade, G. N. & Schneider, J. E. (1992). Metabolic fuels and reproduction in
female mammals. Neuroscience and Biobehavioral Reviews. 16: 235-72.
Wagner, E. J., Manzanares, J., Moore, K. E. & Lookingland, K. J. (1994).
Neurochemical evidence that estrogen-induced suppression of kappa-opioid-
receptor-mediated regulation of tuberoinfundibular dopaminergic neurons is
prolactin-independent. Neuroendocrinology, 59: 197-201.
103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Wang, S. C. & Chinn, H. I. (1954). Experimental motion sickness in dogs:
Functional importance of chemoceptive emetic trigger zone. American Journal of
Physiology. 178: 111-6.
Wang, Y., Lavond, D. G. & Chambers, K. C. (1996). Cooling blockade of the
parabrachial nucleus attenuates conditioned taste aversion in freely moving rats.
Society for Neuroscience Abstracts. 22: 941.
Wang, Y., Lavond, D. G. & Chambers, K. C. (1997a). The effects of cooling the
area postrema of male rats on conditioned taste aversions induced by LiCl and
apomorphine. Behavioural Brain Research. 82:149-58.
Wang, Y., Lavond, D. G. & Chambers, K. C. (1997b). Cooling the area postrema
induces conditioned taste aversions in male rats and blocks the acquisition of LiCl-
induced aversions. Behavioral Neuroscience. 111(4): 768-76.
Weddington, W. Miller, N. & Sweet, D. (1984). Anticipatory nausea and
vomiting associated with cancer chemotherapy. Journal of Psychosomatic Research.
28(1): 73-7.
Weizenbaum, F., Benson, B., Solomon, L. & Brehony, K. (1980). Relationship
among reproductive variables, sucrose taste reactivity and feeding behavior in
humans. Physiology and Behavior. 24: 1053-6.
Yamamoto, T., Shimura, T., Sako, N., Yasoshima, Y. & Sakai, N. (1994a)
Neural substrates for conditioned taste aversion in the rat. Behavioural Brain
Research. 65: 123-37.
Yamamoto, T., Shimura, T., Sakai, N. & Ozaki, N. (1994b). Representation of
Hedonics and Quality of Taste Stimuli in the Parabrachial Nucleus of the Rat.
Physiology and Behavior. 56(6): 1197-202.
Yeomans, M. R. & Symes, T. (1999). Individual differences in the use of
pleasantness and palatability ratings. Appetite. 32(3): 383-94.
Zylan, K. D. (1996). Gender differences in the reasons given for meal
termination. Appetite. 26(1): 37-44.
104
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Appendix A
Food Aversions and Food Preferences Questionnaire
Social Security Number:______ __________ Sex:_______ Age:________ Race: __________
When people become sick after eating a specific food or drink, they may develop an intense dislike, called
an "aversion", for that food or beverage. That is, the foods or drinks make the individuals nauseous when they
see, smell, or taste them, or even when they think about them. We are conducting research to understand this
phenomenon because it plays a role in certain eating disorders and in anorexia (failure to eat) after medical
treatments like radiation and chemotherapy. The experiment we will be conducting does not involve actual
contact with any food or drink. Everything is done in imagination. Would you please answer the following
questions:
1 . Are there specific foods or drinks that you have a strong aversion to. That is, are there foods or drinks that
stimulate a nauseous reaction in you? If so, please print them in order of aversiveness.
most aversive next most aversive
(if you have another one)
If you indicated at least one food or drink as aversive to you, continue with the rest of this questionnaire.
2. How familiar were you with each of the foods or drinks before you developed an aversion to it?
Food or drink Degree of familiarity (check one)
______________ ___ Never eaten it before
most aversive______________________ Had eaten it once or twice before
Had eaten it several times before
Had eaten it often
____________ ___ Never eaten it before
next most aversive__________________ Had eaten it once or twice before
(if applicable)__________________ ___ Had eaten it several times before
Had eaten it often
3. For each of the foods or drinks in Question 1, please indicate as best you can when the aversion began:
Food or Drink Age (circle one range of ages for each food or drink)
0-5 6-10 11-15 16-20 over 21
most aversive
, 0-5 6-10 11-15 16-20 over 21
next most aversive
(if applicable)
4. Now we are switching to another topic: Are there foods or drinks that you strongly dislike but that don’t
stimulate nauseous reactions in you? If there is more than one, please print them in order of dislike.
most disliked next most disliked
(if you have another one)
(continued on next page)
105
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5. How familiar were you with each of these foods or drinks before you developed a dislike for it?
Food or drink Degree of familiarity (check one)
___________ ___ Never eaten it before
most disliked Had eaten it once or twice before
Had eaten it several times before
Had eaten it often
Never eaten it before
next most disliked_______________ ___ Had eaten it once or twice before
(if applicable)__________________ ___ Had eaten it several times before
Had eaten it often
6. For each of the foods or drinks in Question 4, please indicate as best you can when the dislike (but not
aversion) began:
Food or Drink Age (circle one range of ages for each food or drink)
______________ 0-5 6-10 11-15 16-20 over 21
most disliked
0-5 6-10 11-15 16-20 over 21
next most disliked
(if applicable)
7. Now we are changing the direction of the questions to food or drink preferences. What specific foods or
drinks do you really like? If there is more than one, please print them in order of preference.
most liked next most liked
8. How familiar were you with each of the foods or drinks before you developed a preference for it?
Food or drink Degree of familiarity (check one)
_____ _ _. Never eaten it before
most liked Had eaten it once or twice before
Had eaten it several times before
Had eaten it often
Never eaten it before
next most liked_________________ ___ Had eaten it once or twice before
(if applicable) Had eaten it several times before
Had eaten it often
9. For each of the foods or drinks in Question 7, please indicate as best you can when the strong liking began:
Food or Drink Age (circle one range of ages for each food or drink)
_ _ _ _ _ _ 0-5 6-10 11-15 16-20 over 21
most liked
______________ 0-5 6-10 11-15 16-20 over 21
next most liked
106
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Appendix B
Post-Experimental Questionnaire
Please circle the alternative that best matches what you experienced.
1. How detailed were most of the visual images that you were able to generate?
a. I really didn’t generate any visual images.
b. I had some flashes of images but they were brief and sketchy or fuzzy.
c. I pictured only the people speaking and not the setting around them.
d. I pictured clearly the people and the setting around them— as if I was actually watching
the scenes unfold.
Comments:
2. How realistic did the situations presented seem to you?
a. Not at all— I could not believe any of it.
b. Slightly realistic — a few parts were believable but most were not.
c. Somewhat realistic — most of the segments were believable.
d. Very believable.
Comments:
3. How would you rate your emotions or feelings while hearing and thinking aloud to the situations
in which food that you have an aversion to or that you dislike were mentioned?
1 2 3 4 5
calm/not at all somewhat very bothered
bothered bothered
4. To what extent did you feel anxious about expressing your thoughts out loud?
1 2 3 4 5
not at all anxious somewhat anxious very
anxious
(continued on next page)
107
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
5. To what extent did you feel like censoring your thoughts or feelings while speaking out loud?
1 2 3 4 5
not at all some of the all of the time
time
6. How long ago did you last eat something?
a. I ate something just before coming to this experiment.
b. I ate something about an hour ago.
c. I ate something about two hours ago.
d. I ate something about three hours ago.
e. I ate something more than three hours ago.
7. What did you eat when you last ate something? (Please describe)
8. How hungry or full do you feel right now?
a. Very hungry
b. Hungry
e. A little hungry
d. Neither hungry nor lull (sated)
e. A little full (sated)
f. Full (sated)
g-
Very full, kind of stuffed
9. We may be inviting some of the subjects in this experiment to participate in a related study that
involves brain imaging, that is, having your brain activity measured by what is called magnetic
resonance imaging. This is painless and safe. It would take place up at the medical school, and
we would provide transportation back and forth. You would be paid $25.00 for between two and
three hours away from campus. Basically, this study would be similar to what you have just done
but with the addition of brain imaging. Would you be interested in our contacting you later this
semester or in the spring semester to see if this could be arranged?
No, thanks.
_________ _ Yes, I’d be willing to be contacted.
If you check Yes, please give us your local phone number and your email address:
Local phone:____ __________________________
Email (please print clearly):__________________________
108
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Appendix C
INFORMED CONSENT FORM
PRINCIPAL INVESTIGATOR: Gerald C. Davison, Ph.D. DEPT: Psychology
24-HOUR TELEPHONE NUMBERS: (213) 740-3970
PURPOSE OF THE STUDY: You are invited to participate in this experiment, which is designed to
learn more about people’s thoughts when confronted by images of foods that they dislike or prefer.
PROCEDURE: If, after reading this form and having the experiment described to you, you decide
against participating, you are free to end your participation. And if at any time during the study itself
you wish to end your participation, you may do so. You will still get a full hour’s credit for this
experiment. Of course, we hope you will want to continue because your participation can help us
answer some questions about how people think about different foods. This information may
contribute to greater understanding of such problems as overeating and other eating disorders.
Basically, you will be asked to imagine several social situations, like going to dinner at a
friend’s house. Sometimes the food described will be something you like, sometimes it will be
something you dislike, and sometimes it will be something that you have an aversion to. While
imagining each situation, you will be asked to verbalize out loud what is going through your mind,
and you will be audiotaped and videotaped while doing so. Afterwards, our research team will be
able to code what you say and how you say it.
RISKS: Some of the simulated situations you will be hearing may cause you some discomfort because
they refer to a food that you have an aversion to.
BENEFITS: There are no direct benefits to your participating beyond the knowledge that you will
helping us understand why people like and dislike certain foods.
ALTERNATIVE PROCEDURES: You need not participate in this study.
CONFIDENTIALITY: So that you will feel free to tell us whatever is going through your mind as the
different scenarios unfold, we would like to assure you that your responses will not be attached to
your name. Any information that is obtained in connection with this study and that can be identified
with you will remain confidential.
OFFER TO ANSWER QUESTIONS: The faculty members in charge of this experiment are Dr.
Gerald C. Davison (213-740-3970) and Dr. Kathleen C. Chambers (213-740-7344). You may contact
them at any time with any questions or concerns regarding your participation. You will be given a
copy of this form to keep. If you have any questions about this study, please feel free to ask them at
any time.
COMPENSATION: You will receive one credit hour for being in this experiment. Your psychology
instructor will inform you of the amount of extra credit you will receive in the computation of your
grade in his or her course.
(continued on next page)
109
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
COERCION AND WITHDRAWAL STATEMENT: We wish to reiterate and emphasize that your
participation is voluntary, that refusal to participate will involve no loss of the experimental credit just
mentioned, and that you may terminate participation at any time without loss of benefits.
CALIFORNIA LAW REQUIRES THAT YOU MUST BE INFORMED ABOUT:
1. The nature and purpose of the study.
2. The procedures in the study and any drug or device to be used.
3. Discomforts and risks to be expected from the study.
4. Benefits to be expected from the study.
5. Alternative procedures, drugs or devices that might be helpful and their risks and benefits.
6. The opportunity to ask questions about the study or the procedure.
7. The opportunity to withdraw at any time without affecting your future care at this institution.
8. A copy of the written consent form for the study.
9. The opportunity to consent freely to the study without the use of coercion.
AGREEMENT
YOUR SIGNATURE INDICATES THAT YOU HAVE DECIDED TO PARTICIPATE,
HAVING READ THE INFORMATION ABOVE.
Signature of Subject Social Security Number Date
Signature of Witness Date
Signature of Experimenter
110
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Appendix D
CTA-ATSS INSTRUCTIONS AND SCENARIO TRANSCRIPTS
Experimenter says what is in boldface.
Upon arrival at the Laboratory for Cognitive Studies in Clinical Psychology (SGM-718):
Hello, my name is Richard Kunz. Thanks for coming to our experiment. Let me show
you the lab set-up and then we’ll go over the Informed Consent procedures. [Show S the audio
equipment in both the control room and the subject room, also the video equipment in the subject
room.] Please come here [usher S into subject room] and sit down. Please read this Informed
Consent Form. [Let S read it and, if no questions, have S sign it and then you sign it. If S has
questions, explain what seems reasonable but if it seems that certain questions are best answered by
getting into the experiment proper, just tell them that more details are forthcoming. Then before
playing the practice tape for them, see if they will sign it. Remind them that they can stop the
experiment at any time without penalty.]
OK, now I am going to read you some instructions. If you have any questions, just ask me.
[Read the instruction below.]
In this study we are interested in what people think
about when they are confronted with foods that they like very
much and foods that they dislike very much. Rather than
actually show you such foods, we are going to have you
imagine them embedded in different social situations. And
while you are imagining these scenarios, we will be asking you
to tell us what you are thinking and feeling.
Often, when people are going about their daily
affairs, interacting with others and so forth, they have a kind
of internal monologue going through their heads, a constant
stream of thoughts or feelings which reflect their reactions to
something which is happening.
What we'd like you to do is to play a part in several
make-believe situations. Your part will involve listening to
situations and tuning in to what is running through your
mind, and then saying these thoughts out loud. Each situation
is divided into four segments. At the end of each segment, I
will signal you to think out loud about what is going through
your mind in reaction to what you have just heard. I’ll be
giving you half a minute to do this. Then I will ask you to stop
talking and I will continue with the story. Say as much as you
can until I ask you to stop. Of course, there are no right or
wrong answers, so please just say whatever comes to mind,
without judging whether it seems appropriate or not. The
more you can tell us the better.
Try to imagine as clearly as you can that it is really
you in the situation right now. Note that your task is not to
speak back to any one of the voices on the tape, as though you
were having a conversation with one of them. Rather, you
should tune in to your own thoughts and say them out loud.
I ll
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The microphone in front of you will enable us to tape your
comments.
To help us obtain as complete a picture as possible of
your reactions to these simulated situations, we also would
like to videotape you while you are listening to the stories and
verbalizing your thoughts and feelings. [Gesture to
videocamera]
Now, as to the nature of the stories you are about to
hear, ifyou agree to participate in this experiment. Some of
the stories will contain references to a food that you indicated
on a screening questionnaire that you like a lot; other stories
will refer to a food that you dislike. In fact, you will hear a
particular story more than one time, with different foods
being introduced into the narrative. Your job is just to listen
closely to each story, imagine as vividly as you can that it is
happening to you right now, and, when a tone signals every
fifteen or twenty seconds, to verbalize what you are thinking
and feeling in reaction to what you have just heard.
One other thing, between each story there will be a
break of about a minute, and during this break 1 will be
asking you to do a mental task that will kind of clear your
mind of the story you’ve just heard. This mental task will be
a simple counting task, and I will be instructing you what to
do when the time comes. Just do this task out loud until I
start the next story.
Now, this is a somewhat novel procedure, so to help
you learn it and get accustomed to it, I am going to go into the
control room and read you a practice scenario. This will give
you an idea of what the experiment is like. I won’t be tape-
recording or videotaping you because this is just practice. But
I would like you to verbalize your thoughts and feelings when
I tell you to do so. OK? When we finish this practice
scenario, I’ll come back here and see ifyou have any
questions.
[E goes into control room and reads practice story below,
actively coaching S in the ATSS procedure as needed. Two main
things to look out for: (1) S talks to the people in the simulated
situation. This is not what we want. We want their thoughts
about what is unfolding, the things that they are saying to
themselves; (2) Long silences. This is not good. Encourage S to
talk, e.g., “Please tell what’s going through your mind. Whatever
it is.” “Please try to fill the time by telling me your thoughts.”]
[Note: We are not having them close their eyes. This is a change.]
112
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CAMPING TRIP
[Use one of S’s preferred foods from their questionnaire, but not the one to be used in the experiment
proper.]
You have just arrived late at a camp-site to meet a good friend in a location you really like. You
are very hungry. Your friend says: “Gee the only food I’ve got left is _____ .” [SOUND TONE
AND SAY:] OK, now verbalize your thoughts and feelings. [Give S 30 seconds. If S is silent,
encourage them to speak.] [After 30 seconds, SOUND TONE AGAIN AND SAY:] OK, now back to
the story.
Your friend takes the out of a backpack and shows it to you, saying that you should eat
before it gets dark. [SOUND TONE AND SAY:] OK, now verbalize your thoughts and feelings.
[Give S 30 seconds. If S is silent, encourage them to speak.] [After 30 seconds, SOUND TONE
AGAIN AND SAY:] OK, now back to the story.
The _____ is in front of you in a tin camping container, and you look over at your friend, who
is eating it with gusto. [SOUND TONE AND SAY:] OK, now verbalize your thoughts and
feelings. [Give S 30 seconds. If S is silent, encourage them to speak.] [After 30 seconds, SOUND
TONE AGAIN AND SAY:] OK, now back to the story.
Your friend is finishing eating. He looks at you and says: “Why haven’t you eaten more______
? You realize that’s all we have till we leave for home tomorrow afternoon.” [SOUND TONE
AND SAY:] OK, now verbalize your thoughts and feelings. [Give S 30 seconds. If S is silent,
encourage them to speak.] [After 30 seconds, SOUND TONE AGAIN AND SAY:] OK, I’ll be
coming back into the room now.
[Then E returns to subject room and gives any feedback that seems necessary for
encouraging S to engage the ATSS procedure. If S didn’t talk much during practice tape, emphasize
in a nice but unequivocal way that we need to have them verbalize as much as they can during the
breaks between stimulus segments. “Fill the time” is a phrase that can be used. Then read the
following to S while you are in the Subject room with them:]
OK, do you have any questions at all about the
procedure? Remember: We are interested in what is on your
mind as you imagine yourself in these various situations. The
only way we can get a handle on your thoughts and feelings is
for you to tell us about them in as much detail as you can. Do
your best to imagine the situations vividly, imagine you are
right in them, and, when I signal a pause in the story, talk out
loud as much as you can about your reactions to what you
have just heard.
[If S didn’t verbalize much during the 30-second
response intervals of the practice tape, E encourages S to talk
more during the actual experiment.]
OK, now I am going to turn on the videocamera and it’ll just be running continuously
while I am reading you the stories and while you are articulating your thoughts and feelings
when I ask you to do so. [Make any necessary adjustments on the videocamera and turn it on. Then
leave the room, close the door, go into control room close the door, and turn on the tape recorder and
113
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
make any necessary adjustments. If necessary, have S say his or her name to get recording level.
And you can ask them if the volume is adequate from the speakers.]
[E then reads each stimulus transcript three times each — once where the food is the subject’s
preferred, once where the food is the subject’s disliked but not nauseating, and once where the food is
aversive to the subject. The order of presentation will be counterbalanced. S’s verbalizations will be
audiotaped and S’s nonverbals will be videotaped. Here are the three scripts:]
FANCY RESTAURANT WITH PROSPECTIVE EMPLOYER
You are at a fancy restaurant with a prospective employer, whom you are trying hard to
impress, and he/she says: “You should really order the______ because it’s one of the best dishes
they make.” [SOUND TONE AND SAY:] OK, now verbalize your thoughts and feelings. [Give S
30 seconds. If S is silent, encourage them to speak.] [After 30 seconds, SOUND TONE AGAIN
AND SAY:] OK, now back to the story.
You are looking at the menu in an effort to find see the variety of things that can be ordered.
You then notice that is being served at the next table. [SOUND TONE AND SAY:] OK,
now verbalize your thoughts and feelings. [Give S 30 seconds. If S is silent, encourage them to
speak.] [After 30 seconds, SOUND TONE AGAIN AND SAY:] OK, now back to the story.
You’ve decided to order th e. ______ , and the waiter is just placing the dish in front of you. You
can really smell the . [SOUND TONE AND SAY:] OK, now verbalize your thoughts and
feelings. [Give S 30 seconds. If S is silent, encourage them to speak.] [After 30 seconds, SOUND
TONE AGAIN AND SAY:] OK, now back to the story.
You are about halfway through the meal and your host asks with concern: “Is there anything
else you’d like to have besides _?” You look down at your plate, now empty except for a
few leftovers from the . [SOUND TONE AND SAY:] OK, now verbalize your thoughts
and feelings. [Give S 30 seconds. If S is silent, encourage them to speak.] [After 30 seconds,
SOUND TONE AGAIN AND SAY:] OK, now count backwards out loud, slowly from 100 by 3's.
[Do this for one minute and then say:] OK, please stop counting and listen to the next story.
DINNER AT FRIEND’S PARENTS’
You are at the home of the parents of a good friend, and dinner is about to be served. Your
friend says: “Hey, we’re having a special dish, [SOUND TONE AND SAY:] OK, now
verbalize your thoughts and feelings. [Give S 30 seconds. If S is silent, encourage them to speak.]
[After 30 seconds, SOUND TONE AGAIN AND SAY:] OK, now back to the story.
You are sitting next to your friend’s mother, and she is passing you a plate that has on it,
and she says: “You really have to try some of this!” [SOUND TONE AND SAY:] OK, now
verbalize your thoughts and feelings. [Give S 30 seconds. If S is silent, encourage them to speak.]
[After 30 seconds, SOUND TONE AGAIN AND SAY:] OK, now back to the story.
Although you have had as much as you want and have already declined to take more,
your friend’s mother is putting some on your plate and saying: “I made this especially
for you. Please take some more.” [SOUND TONE AND SAY:] OK, now verbalize your
114
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
thoughts and feelings. [Give S 30 seconds. If S is silent, encourage them to speak.] [After 30
seconds, SOUND TONE AGAIN AND SAY:] OK, now back to the story.
Your friend’s mother notices that you haven’t taken any more and looks over at her
husband with a hurt expression on her face. [SOUND TONE AND SAY:] OK, now verbalize
your thoughts and feelings. [Give S 30 seconds. If S is silent, encourage them to speak.] [After 30
seconds, SOUND TONE AGAIN AND SAY:] OK, now count backwards slowly, out loud from 96
by threes. [Do this for one minute and then say:] OK, please stop counting and listen to the next
story.
REHEARSAL DINNER FOR A WEDDING
You are entering a nice restaurant for a dinner that is being held the day before the wedding of
one of your favorite relatives. As you enter the room reserved for this dinner, the bride’s
mother tells you that is part of the meal. [SOUND TONE AND SAY:] OK, now verbalize
your thoughts and feelings. [Give S 30 seconds. If S is silent, encourage them to speak.] [After 30
seconds, SOUND TONE AGAIN AND SAY:] OK, now back to the story.
There is no way to order a meal that doesn’t have _____, and now you see it being served to the
people at your table. [SOUND TONE AND SAY:] OK, now verbalize your thoughts and
feelings. [Give S 30 seconds. If S is silent, encourage them to speak.] [After 30 seconds, SOUND
TONE AGAIN AND SAY:] OK, now back to the story.
The server has placed your plate in front of you, and now you can get a good look at the______
and can smell it as well. [SOUND TONE AND SAY:] OK, now verbalize your thoughts and
feelings. [Give S 30 seconds. If S is silent, encourage them to speak.] [After 30 seconds, SOUND
TONE AGAIN AND SAY:] OK, now back to the story.
Dessert is about to be served, and as the server comes by to pick up your dinner plate, she asks
you: “Did you enjoy the You nod yes. Everyone at your table is looking at you. [Give
S 30 seconds. If S is silent, encourage them to speak.] [After 30 seconds:] OK, now count
backwards slowly, out loud from 92 by threes. [Do this for one minute and then say:] OK, please
stop counting and listen to the next story.
[After the three scenarios are completed with each of the three kinds of foods:]
Good. I’ll be rejoining you in your room now. [E goes into Subject room and hands S
Post-Experimental questionnaire.] The last thing I am going to ask you to do is just to fill out this
very short questionnaire. Please try to be as accurate and frank as possible. When you’re
through, just tell me and I’ll get it from you. [Hand S the questionnaire and leave the room while
they are filling it out.]
115
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Affection and conflict in family relationships

PDF

Expectancies for alternative behaviors predict drinking problems: A test of a cognitive reformulation of the matching law

PDF

Applying the theory of reasoned action to condom use: The effect of immediate consequences on intention-behavior consistency

PDF

Effects of bilateral stimulation and stimulus redundancy on performance in processing nonword letter trigrams

PDF

Content analysis of articulated thoughts of chronic worriers

PDF

A typology of maritally violent men and correlates of violence

PDF

Inactivation of the interpositus nucleus prevents transfer of the rabbit's classically conditioned eyeblink response from a light to a tone-conditioned stimulus

PDF

Articulated thoughts about intentions to commit anti-gay hate crimes

PDF

Head injury and dementia: A co-twin control study of Swedish twins

PDF

Effects of threat and self-focus on consensual bias in majorities

PDF

Intragroup evaluations, attitude source, and in-group member derogation

PDF

Effects of personal resource sufficiency on perceived difficulty and desirability of earthquake preparedness

PDF

Adolescents' social attitudes: Genes and culture?

PDF

Alcohol's impact on triggered displaced aggression

PDF

Depression in offspring following severe prenatal stress

PDF

History of depression, antidepressant treatment, and other psychiatric illness as risk factors for Alzheimer's disease in a twin sample

PDF

Effects of a corneal anesthetic on the extinction of the classically conditioned response in the rabbit

PDF

Ethnic identity, acculturation, self-esteem and perceived discrimination: A comparison study of Asian American adolescents

PDF

A one-year public relations campaign developed for the Jewish Free Loan Association emphasizing fundraising and image-building

PDF

Atypicality: Benefit and bane. Provocation, trigger, typicality, and the expression of aggression and generalization

Asset Metadata

Creator Kunz, Richard Daniel (author)

Core Title Hedonic aspects of conditioned taste aversion in rats and humans

School Graduate School

Degree Master of Arts

Degree Program Psychology

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

Tag OAI-PMH Harvest,psychology, behavioral,psychology, physiological

Language English

Contributor Digitized by ProQuest (provenance)

Advisor [illegible] (committee chair), Davison, Gerald C. (committee member), Lavond, David (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c16-38007

Unique identifier UC11337470

Identifier 1406455.pdf (filename),usctheses-c16-38007 (legacy record id)

Legacy Identifier 1406455.pdf

Dmrecord 38007

Document Type Thesis

Rights Kunz, Richard Daniel

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