The acquisition of taste aversions in humans with eating and drinking disorders

A written questionnaire or interview concerned with acquisition of illness-induced (taste) aversions to foods and drinks was given to three groups of people with eating and drinking disorders. These groups consisted of 101 male and 1 female hospitalized alcoholics, 8 male and 8 female college-student heavy consumers of alcohol and 18 females with anorexia nervosa and/or bulimia. In most respects taste-aversion acquisition in these three groups was similar to taste-aversion acquisition in a general college-student population previously studied by Logue, Ophir and Strauss (1981), and to taste-aversion acquisition in other species. In all three groups the aversions were more likely to be reported as having been formed through forward rather than simultaneous or backward conditioning, and long-delay learning was frequent. The aversions usually formed to the tastes rather than to the appearance or other aspects of the foods and drinks. Extinction appeared more effective in decreasing the aversions than did forgetting. While the illness responsible for the aversions forming was usually attributed to the subsequently aversive food or drink, in at least one third of the cases subjects reported that something else might have caused their illness. Aversions were more likely to have formed to relatively less familiar and less preferred foods and drinks. However, the hospitalized alcoholics reported fewer aversions, less generalization of aversions, and stronger nausea as the cause of the aversions than did Logue et al.'s (1981) subjects. About 15% of these subjects reported taste aversions to alcoholic beverages. The college-student heavy consumers of alcohol reported no generalization of their taste aversions, but in other respects were similar to Logue et al.'s subjects. Thirty-one percent of these subjects reported taste aversions to alcoholic beverages. The anorexic and bulimic subjects were also similar to Logue et al.'s subjects with the exception that they, like the hospitalized alcoholics, reported stronger nausea as the cause of the aversions. These data may help to understand and treat people with eating and drinking disorders     

Types of food aversions: animal, vegetable, and texture

Despite a growing body of research investigating the origins and effects of food aversions, few research instruments have been developed to measure aversions to specific types or categories of food. Undergraduates (N = 209) responded to a series of food aversion questionnaires. The results suggest that people tend to be averse to 2 types of foods (vegetables and meats or fats) and to the texture and taste of certain foods (e.g., oysters). Aversions were slightly more prevalent among women than among men and were correlated with lower educational levels. The authors provide a means of advancing future research on this problem by reliably identifying 3 categories of food aversions. Future researchers should evaluate additional food categories and expand the focus on food aversions beyond the current concern with learned avoidance of specific food items.     

Types of Food Aversions: Animal,Vegetable, and Texture

Despite a growing body of research investigating the origins and effects of food aversions, few research instruments have been developed to measure aversions to specific types or categories of food. Undergraduates (N = 209) responded to a series of food aversion questionnaires. The results suggest that people tend to be averse to 2 types of foods (vegetables and meats or fats) and to the texture and taste of certain foods (e.g., oysters). Aversions were slightly more prevalent among women than among men and were correlated with lower educational levels. The authors provide a means of advancing future research on this problem by reliably identifying 3 categories of food aversions. Future researchers should evaluate additional food categories and expand the focus on food aversions beyond the current concern with learned avoidance of specific food items.     

Human Flavor-Aversion Learning: A Comparison of Traditional Aversions and Cognitive Aversions

Traditional flavor aversions form when an organism consumes a flavor prior to an illness-producing unconditioned stimulus (US). Rozin (1986), however, introduced a second type of taste-aversion learning produced by a disgust-eliciting US. In the present work, Rozin's disgust categorization is extended to include other human situations (negative information, forced consumption) in which taste aversions are mediated by cognitive processes. College students completed a self-report questionnaire that addressed their experiences with taste-aversion learning. The results show that cognitive aversions were stronger and longer-lasting than traditional aversions, they formed to foods of animal origin more than to other substances, and they could be produced without nausea. Finally, individuals who reported having a cognitive aversion had significantly more taste aversions than individuals who reported having a traditional aversion. These results provide additional evidence that taste-aversion learning can occur without the direct mediation of an illness-producing US.     

One-trial acquired likes and dislikes in humans: Disgust as a US,food predominance,and negative learning predominance

Self-reports were solicited by questionnaire on experiences in which likes or dislikes for anything except people were produced as a result of a single event. The same questionnaire also asked for cases of one-experience acquisition of, or loss of, fears. Analysis of 440 reports indicates that the learned taste aversion phenomenon accounts for most rapid acquired dislikes; foods were the subject of 77% of all rapid acquired dislikes. A new form of food “aversion” learning appeared (22 cases), in which the US is a disgusting stimulus rather than illness. This may be mediated by nausea, the same US as in illness-produced aversions. There were more than three times as many cases of one-trial dislikes as opposed to likes. This could result from a bias in remembering or sampling, a bias in the occurrence of events in the world, and/or a bias to learn more rapidly about negative events. In contrast to acquired likes and dislikes, foods were the subject of one-trial acquired fear in only 1 case out of 164.     

Taste aversions to mother's milk: the age-related role of nursing in acquisition and expression of a learned association.

The development of taste aversion learning to novel cues contained in mother's milk was examined in rat pups. Pups receiving distinctive mild by experimenter-delivered oral infusions followed by toxicosis formed an aversion to the dam's diet. Robust aversions were learned as early as Day 10 and were retained for at least 11 days. When the same distinctive milk was obtained directly from a foster mother through nursing, however, only weanling-age pups (over 20 days) formed an aversion. X-ray analysis of nipple location in the mouths of suckling pups suggested that pups between the ages of 10 and 21 days receive milk at a similar tongue locus. Flavored milk was then delivered at specific time intervals in controlled quantities through tongue cannulas implanted at loci corresponding to the nipple position shown by the X-rays. Cannulated preweanling pups that were attached to a nipple during milk delivery failed to associate the taste cue with illness, whereas both preweanlings off the nipple and weanlings on the nipple acquired aversions to the taste cue in the milk. The evidence obtained in these experiments suggests that pups of all ages are incapable of expressing a taste aversion in a nursing situation and that preweanling pups in particular are also dificient in acquiring aversions within a suckling context. The inability of preweanling pups to acquire taste aversions in a nursing situation appears to result from a failure to associate taste cues with illness rather than a failure to detect taste cues obtained from a nipple.     

Formation of food aversions in cancer patients receiving repeated infusions of chemotherapy

Patients receiving emetogenic chemotherapy for cancer have been found to develop aversions to normal dietary items consumed in close temporal relation to treatment administrations. These aversions are presumed to develop via conditioning processes as demonstrated in experimental studies of food aversion learning. The present study used a prospective, longitudinal design to evaluate the possible role of conditioning in the formation of aversions to normal dietary items in women receiving adjuvant chemotherapy for breast cancer. Patients were monitored for the development of aversions to foods and beverages consumed in the 24 hr periods before and after each of eight consecutive chemotherapy infusions beginning with the initial infusion. Data on the prevalence, course, and prediction of aversions to normal dietary items are reported. These results pointed to similarities and differences between aversions formed to normal dietary items during chemotherapy treatment and aversions formed to taste stimuli during experimental conditioning studies. In addition to their theoretical significance, results also suggest possible strategies for preventing the clinical problem of aversions to normal dietary items in chemotherapy patients.     

Relative potency of foods and drinks as targets in aversion conditioning

The potency of food stimuli as targets in aversion learning was examined using an interference paradigm. In the first study, foods interfered with liquid aversions but liquids did not interfere with food aversions. In the second study, food aversions were found to be resistant to interference by other foods. These findings suggest that foods are relatively potent targets in aversion conditioning in that they are resistant to interference by both foods and drinks. The final study examined the contribution of flavor intensity and nutrient density to the potency of aversion conditioning. The relative potency of foods over drinks may reflect differences in their intensity as well as their different roles in toxin avoidance and nutrient selection.     

Simultaneous odor-taste and taste-taste compounds in poison-avoidance learning

In six experiments, rats received an odor or a taste alone or in simultaneous compound with another taste prior to lithium chloride-induced illness. Aversions to the target odor and the target taste were then assessed. In Experiments 1–3, the presence of the taste in compound with the target during conditioning attenuated the strength of the aversions to both the target odor and the target taste. Although these results were consistent with the familiar principles of compound conditioning, they contradicted previous reports of potentiation, rather than attenuation, of odor conditioning by taste. In Experiment 4, taste again attenuated the conditioning of odor when a second odor was presented during the CS-US interval. In Experiment 5, we looked for odor potentiation with a within-subject design, and in Experiment 6, we examined an alternative method of presenting odor. In neither case did we find odor potentiation by taste. Over the range of conditions investigated here, a taste often attenuates, and never potentiates, the conditioning of aversions to both odors and tastes.     

Disruption of conditioned taste aversion: evidence that ECS weakens the gustatory engram

Two experiments are reported concerning the neural substrate of conditioned taste aversion (CTA) and the amnesic mechanisms involved when such learning is disrupted by electroconvulsive shock (ECS). Subjects were adult male rats. In the first experiment it is demonstrated that an aversion established to a 2.5% sucrose solution can simulate the learning loss reported in earlier studies which was caused by interpolating ECS between tasting and illness when the aversion was established using a 10% sucrose cue. It is concluded that if ECS acts to disrupt the memory of the taste cue, it possibly reduces it to only about one-quarter of its original strength, a substantial deficit not readily apparent simply from the degree of aversion displayed. In the second experiment, CTAs were established using either a 2.5% sucrose cue or a 10% cue with ECS interpolated during the taste-illness interval. Animals in the two groups were subsequently confronted with a range of sucrose stimuli and their respective aversions were compared. Near identical responses were observed under all conditions tested. These findings are consistent with the theory that ECS disrupts CTA by weakening the gustatory engram.     

Formation of associations of colored and flavored food with induced sickness in five avian species

Domestic ducks, geese, pigeons, quail, and chickens were given colored, flavored, or colored and flavored food and then injected with lithium chloride. Each species showed learning of color and taste aversions. Flavor facilitated the formation of color aversions in ducks, geese, and pigeons but not in quail or chickens. Color interfered with the formation of flavor aversions in quail and chickens but not in the other three species. These findings indicate that all birds can probably associate both colored and flavored food with induced sickness and that colored food is more easily associated with induced sickness than is colored water. Moreover, these findings suggest that the capacity to associate colored and flavored food with induced sickness and the interaction between color and flavor in food vary between species. Birds, such as quail and chickens, that eat relatively tasteless food rely more on color than on flavor cues when forming learned food aversions. Birds that can select their food on the basis of taste, such as ducks and geese, rely more on flavor than on color when forming aversions to food. Birds, such as pigeons, that are initially raised on tasty food before switching to predominantly tasteless food show tendencies that are similar to those of ducks and geese.     

Developmental flavor experience affects utilization of odor, not taste in toxiphobic conditioning

Feeding experiences were varied in developing rats and the effects upon flavor neophobia and lithium chloride-induced flavor aversions were observed. In Experiment 1, nursing experience of neonate rats was reduced by artificial feeding via intragastric cannula; the rats then were tested with apple juice paired with lithium chloride injection at weaning or maturity. Conditioned aversions were not affected, but neophobia to novel apple juice was attenuated in artificially-reared rats tested at maturity. In Experiment 2, rats received enriched feeding experience after weaning, which consisted of (a) obtaining many complex flavors, a few of which were paired with poisoning, effortlessly in the home cage, or (b) foraging for various foods on an elevated maze. No dramatic effects on neophobia or conditioned taste aversion for saccharin water were apparent. In Experiment 3, rats were given experience after weaning with vanilla-scented water either paired or unpaired with quinine water, and then tested with the odor of almond or that odor compounded with saccharin water for neophobia and lithium-induced aversions. Flavor-experienced rats exhibited more pronounced odor conditioning and more resistance to extinction of the odor aversion after both simple and compound conditioning. In contrast, saccharin taste aversions were relatively unchanged. Apparently, enriched feeding and drinking experience facilitates the utilization of odor more than taste cues.     

A comparison of two methods of assessing representation-mediated food aversions based on shock or illness

In experiments that measured food consumption, Holland (1981; Learning and Motivation, 12, 1-18) found that food aversions were formed when an exteroceptive associate of food was paired with illness, but not when such an associate was paired with shock. By contrast, measuring the ability of food to reinforce instrumental responding, Ward-Robinson and Hall (1999; Quarterly Journal of Experimental Psychology, 52B, 335-350) found that pairing an associatively activated representation of food with shock readily established an aversion to that food. Two experiments considered the origins of these apparently discrepant results. The results did not support either the possibility that instrumental reinforcement power is a more sensitive measure of aversion learning than consumption, or the hypothesis that illness particularly devalues properties of food representations that determine consumption (such as palatability) whereas shock devalues more general properties critical to reinforcement. The results suggested instead that whereas the effects of pairings of a food associate with illness are mediated by changes in the value of the food itself, the effects of pairings with shock are mediated by the conditioning of fear or other competing responses to the site of food delivery, and not by modification of the value of food itself.     

Flavor aversions produced by contingent drug injection: Relative effectiveness of apomorphine, emetine, and lithium

Shortly after rats began drinking saccharin solution, different groups of them were injected with different doses of apomorphine, emetine, or lithium. A control group was not injected at all. Six days later, the rats were given free access to saccharin solution and to water. Aversions to saccharin solution were obtained in all injected groups and tended to become more pronounced as the dose level increased. At similar dose levels, lithium produced the most pronounced aversions, apomorphine produced the weakest aversions, and emetine was intermediate. A follow-up study with squirrel monkeys confirmed that lithium produces more pronounced aversions than either emetine or apomorphine. From these results, it seems worthwhile to try lithium in the chemical aversion treatment of alcoholism.     

Early learning as ontogenetic adaptation for ingestion by rats

Ontogenetic adaptation is an “ecological” concept in which mammalian maturation is seen as a coordinated sequence of specializations (stages) that enables the infant to survive within a sequence of distinct niches created by the parent(s) and the rest of the environment. Natural selection is presumed to operate at each point in development. The development of ingestive behavior in the rat pup as a series of ontogenetic adaptations enabling the infant while in the nursing niche to obtain and utilize mother's milk and progress gradually into the feeding niche in which a variety of solid foods must either be obtained or avoided is discussed. Data to show that learning is an integral part of these ontogenetic adaptations for ingestion are reviewed and interpreted. Preweanling rats that are capable of learning and remembering toxiphobic food aversions do not display an aversion to a flavor conditioned stimulus (CS) paired with illness if the CS is presented during suckling. In contrast, 20-day-old pups (weanlings) trained identically do show conditioned taste aversions under these conditions. A perspective of ontogenetic adaptations to this phenomenon of “blockade of toxiphobia by suckling” and its dissolution at weaning is applied. The generality of the blockade phenomenon and its mechanisms are explored, and it is shown that feeding experience regulates the stage-appropriate learning strategies used by the pups.     

Selective associations in one-day-old rats: taste-toxicosis and texture-shock aversion learning

One-day-old rats learned aversions to a novel taste paired with lithium-induced distress and to a tactile stimulus paired with brief electric shocks. However, aversions did not develop when taste was paired with shock or when the tactile stimulus was paired with lithium treatment. The aversions occurred only when lithium treatment immediately followed taste exposure and when shock was concurrent with exposure to the tactile stimulus. These findings indicate that selective associations in aversion learning are mediated by innate mechanisms that govern conditioning in the absence of extensive ontogenetic experience. The present research also shows that selective associations are sufficient for the occurrence of long-delay learning.     

Flavor-illness aversions: potentiation of odor by taste is disrupted by application of novocaine into amygdala

Taste and odor have different properties in toxiphobic conditioning. When each is used alone, taste becomes aversive when followed by immediate or delayed poison, while odor becomes aversive only if followed by immediate poison. However, if odor and taste are presented as a compound and followed by delayed poison, then odor does become aversive when tested alone. It is as if taste has potentiated the odor signal. Several experiments assessed the role of the amygdala in this potentiation effect by anesthetizing the amygdala with 10% novocaine. Novocaine applied 30 min before presentation (Pre-CS) of an odor-taste compound disrupted the potentiated odor aversion but not the taste aversion. In contrast, novocaine applied 1 min after the compound odor-taste or 1 min prior to LiCl poison did not dissociate odor and taste aversions; both odor and taste aversions were facilitated. Novocaine applied 30 min before an odor alone also disrupted an odor aversion induced by immediate LiCl. But identical treatment did not disrupt odor avoidance conditioned by immediate foot-shock, suggesting that amygdala anesthesia does not simply produce anosmia. Pre-CS novocaine treatment also disrupted flavor neophobia prior to conditioning. The results suggest that novocaine applied to the amygdala disrupts the integration of odor with taste and illness during toxiphobic conditioning.     

Experimentally induced taste aversions in humans: effects of overshadowing on acquisition.

This study used vection-induced symptoms of motion sickness as an unconditioned stimulus to condition food aversions in humans and to evaluate the efficacy of an overshadowing agent (novel flavored candy: CS2) to attenuate acquisition of the aversion. Subjects unfamiliar with a target food (CS1) were assigned to one of the following three groups which were identical except for order of exposure to stimuli: Taste Aversion Group (CS1-US-CS2-Test), Control Group (US-CS1-CS2-Test), and Overshadowing Group (CS1-CS2-US-Test). Subjects were tested on aversion ratings and consumption of the target flavor and ratings of the overshadowing agent. Subjects in the Taste Aversion group rated the target flavor as significantly more aversive and consumed less of it, although not significantly so, that did those in the Control group. The Overshadowing group consumed significantly more of the target food than did the Taste Aversion Group. Considering only subjects unfamiliar with the overshadowing agent, those in the Overshadowing group rated the agent (CS2) as significantly more aversive than the Taste Aversion and Control groups. Implications of these findings to taste aversions in humans are discussed.     

Taste aversions conditioned by the aversiveness of insulin and formalin: role of CS specificity

Experimenters in the past have reported that when insulin is used as the unconditioned stimulus (US), rats will learn an aversion to a sodium chloride but not a sucrose solution, whereas with formalin as the US, they will learn an aversion to a sucrose but not a saline solution. The present experiments failed to confirm these findings. Aversions to sucrose were conditioned with insulin and aversions to sodium chloride were conditioned with formalin. The use of a more concentrated sucrose solution in the present study may have been responsible for the successful sucrose-aversion conditioning with insulin. Although the source of the discrepancy in findings concerning aversion conditioning with formalin remains unclear, experiments ruled out numerous possibilities. These experiments also showed that sodium chloride aversion conditioning with formalin is a highly robust phenomenon that occurs with a variety of conditioned stimulus durations and formalin doses, with distributed and massed training, in male and female rats, and even if saline is not the only novel solution presented during conditioning. Furthermore, the aversion can be detected with both single-stimulus and choice test procedures.     

Acquisition and extended retention of a conditioned taste aversion in preweanling rats

The time course of memory decay for infant rats may shed light on the processes responsible for infantile amnesia. A taste aversion conditioning procedure appropriate for both neonatal and adult rats was employed in four experiments to investigate the ontogeny of extended retention. In Experiment 1, rats trained at 1, 10, 20, or 60 days of age were tested for retention of the taste aversion 25 days later. At testing, only those rats conditioned when 20 or 60 days old demonstrated significant taste aversions. Experiments 2 and 3 established that rats 14-15 days old and older were able to retain significant taste aversions following a 25-day retention interval. Younger rats did, however, acquire and retain the aversion for several days and showed a gradual retention loss over progressively longer retention intervals (Experiment 4). These findings suggest that preweanling rats demonstrate initial consolidation, storage, and retrieval of conditioned taste aversions. It is only after this initial period that retention deficits become evident.