Poison-induced neophobia in rats: Role of stimulus generalization of conditioned taste aversions

Rats repeatedly injected with lithium chloride were subsequently tested drinking novel and familiar solutions of both casein hydrolysate and vinegar. Injections in the absence of edibles result in only a small, and sometimes not reliable, increased avoidance of the novel casein and vinegar solutions. In contrast, if subjects acquired an aversion to saccharin as a result of the lithium injections, this learned aversion generalized to casein hydrolysate, with the generalization greatly enhanced by novelty of the casein flavor. However, the saccharin aversions did not generalize to the novel vinegar solution nearly as much as to the novel casein flavor. These results suggest that previous observations of poison-induced neophobia were probably in part a result of the stimulus generalization of conditioned taste aversions and that in addition to test stimulus novelty some other factor, such as stimulus salience or similarity to the conditioned aversive flavor, is also involved in the generalization of learned taste aversions.     

Effects of context exposure during conditioning on conditioned taste aversions

Rats were used in a conditioned taste aversion procedure in order to examine the effects of context exposure duration during the conditioning sessions on conditioned responding. One flavor was paired with lithium chloride during a long session in one context, whereas another flavor was conditioned during a short session in another context. Testing occurred in the home cage. The results showed that conditioning during short sessions produced strong conditioned taste aversions. Conditioning during long sessions produced strong conditioned taste aversions when the conditioned-stimulus-unconditionedstimulus (CS-US) pairing occurred at the end of the lengthy session. Other results showed that context-US associations were formed during the short duration sessions and that these associations supported conditioned responding to the CS trained in that context. The results are discussed with respect to the different influences that contextual cues can exert on conditioned responding.     

Acquisition and retention of conditioned aversions to context and taste in laboratory mice

We compared the rate of acquisition and strength of retention of conditioned context aversion (CCA) with conditioned taste aversion (CTA) using pigmented, genetically heterogeneous mice (derived from Large and Small strains). Extending previous findings, in Experiment 1, mice accustomed to drinking from large glass bottles in the colony room learned to avoid graduated tubes after a single conditioning trial when drinking from these novel tubes was paired with injections of LiCl. The results also showed that CCA could be developed even when there was a 30-minute delay between conditioned stimulus and unconditioned stimulus. Retention of the aversion lasted for 4 weeks in both Immediate and Delay groups. Studies of conditioned saccharin aversion were conducted in Experiment 2. CTA acquisition was very similar to that observed in CCA and duration of aversion retention was similar in the CCA and CTA Delay groups, although at least 2 weeks longer in the Immediate group. Thus, CCA acquisition and retention characteristics are closer to those seen for CTA than has previously been reported. In Experiment 3, we examined whether albino mice (which are known to have weaker visual abilities compared to pigmented mice) would develop CCA comparable to those of pigmented mice. The development of conditioned aversion and its duration of retention was similar in albinos and pigmented mice. Nonspecific aversion emerged as an important contributor to strength of aversion during retention trials in both CCA and CTA paradigms with pigmented (but not albino) mice and deserves additional scrutiny in this field of inquiry.     

Possible role of confounded taste stimuli in conditioned taste aversions

In many publications, Sjödén, Archer, and their colleagues have studied conditioned taste aversions and found that in addition to taste, the bottle and/or spout containing a solution exerts strong control over the expression of an aversion. For example, Sjödén and Archer claim that this bottle stimulus will support a bottle-illness association even with long delays between the bottle and illness and after only a single conditioning trial. They have interpreted their results as indicating that contextual effects are important in taste-aversion learning. However, a confound in their procedure, stemming from the bottles they used, could explain their results in a simple way. Sjödén and Archer have emphasized that the bottle stimuli they used to distinguish between contexts consisted of one of two different sizes of drinking spouts, the larger of which made a clicking noise when licked. However, the larger spouts were always attached to plastic bottles and the smaller spouts were always attached to glass bottles. If discriminable tastes from the plastic bottles existed, then taste may have been inadvertently manipulated. Support for the likelihood of a plastic taste includes the seminal work on taste-aversion learning that stemmed from the serendipitous use of plastic bottles in the cage that rats were irradiated in and glass bottles in the home cages (see Garcia, McGowan, & Green, 1972). In these early demonstrations, rats learned to avoid the taste of water in the plastic bottles but not the taste of water in the glass bottles.     

The interaction of conditioned taste aversions and schedule-induced polydipsia: Availability of alternative behaviors

Animals poisoned following exposure to saccharin subsequently avoided the schedule-induced consumption of saccharin. While this suppression was transient for subjects who had access only to the saccharin solution during the free-food presentations, recovery of schedule-induced saccharin consumption was significantly retarded for subjects who had concurrent access to saccharin and a running wheel. It has been suggested that the transient suppression of schedule-induced polydipsia by conditioned taste aversions results from the pellet-induced tendency to drink within the schedule-induced polydipsia procedure. That access to the running wheel reduces schedule-induced polydipsia in general and prolongs the suppression of schedule-induced polydipsia by taste aversions supports this view.     

Neophobias and conditioned taste aversions in rats following exposure to novel flavors

Rats were successively exposed to three solutions with distinctively different flavors and then tested for both neophobia and propensity to form conditioned taste aversions to a fourth distinctively flavored solution. All permutations between the four solutions (salty, bitter, sweet, and sour) were examined. The prior exposures resulted in attenuation of neophobia to novel salty and sour solutions, but not to equally novel bitter or sweet solutions. These effects were found to depend upon thediversity of the prior ingestive events rather than upon either a single specific flavor experience or a summation of the reductions in generalized neophobia accrued by each substance separately; both of the latter findings are inconsistent with stimulus generalization being responsible for the observed attenuation of neophobia to salty and sour solutions following exposure to diverse different solutions. A further test of generalization between the salty and bitter solutions, consisting of associating one flavor with poison and extinguishing the avoidance response in half the animals prior to testing for generalization of conditioned taste aversion to the other flavor, also proved negative. Although these effects of exposure to flavors distinctly different from the test solution may be dependent upon solution concentrations, further research found that the same pretest exposures and same test concentrations failed to inhibit formation of conditioned taste aversions. A demonstration of “latent inhibition” attested to the sensitivity of our procedure to potential interference with acquisition of conditioned taste aversions. The results are considered in light of the relationship between neophobia and conditioned tasted aversions, the differential biological relevancy of specific tastes, and abstraction as a cognitive capability of rats. The possibility is raised that the defense against toxins is not the primary function of neophobia.     

The interaction of conditioned taste aversions and schedule-induced polydipsia: Effects of repeated conditioning trials

In Experiment 1, rats poisoned following schedule-induced saccharin consumption showed a moderate reduction in the schedule-induced consumption of saccharin. With repeated poisoning, schedule-induced saccharin polydipsia was markedly reduced. Acquisition of conditioned aversion under the schedule-induced procedure was significantly slower than acquisition under water deprivation. In addition, recovery of consumption of the previously poisoned solution during extinction was more rapid under schedule-induced polydipsia. Experiment 2 revealed that schedule-induced polydipsia was less sensitive to suppression by conditioned aversions than a prandial drinking condition in which subjects were equally food deprived but were given a mass feeding instead of spaced pellet deliveries, suggesting that the relative insensitivity of schedule-induced polydipsia to conditioned taste aversions is not simply a function of different levels of food deprivation. This relative insensitivity is offered as a partial basis for the occurrence and maintenance of schedule-induced alcohol polydipsia.     

The effects of conditioned taste aversions on the acquisition and maintenance of schedule-induced polydipsia

Conditioned taste aversions produced a moderate, but transient, suppression of schedule-induced polydipsia. This suppression was greater and longer lasting when rats were offered a choice between water and the previously poisoned solution on the polydipsia baseline. A final experiment demonstrated that taste aversions were more effective in suppressing schedule-induced consumption when superimposed on a developing schedule-induced drinking baseline as opposed to a stable pattern of schedule-induced drinking. It was suggested that schedule-induced polydipsia is insensitive to conditioned taste aversions. This conclusion was discussed in terms of schedule-induced alcohol consumption and its potential as an animal model of alcoholism.     

Circadian-temporal context and latent inhibition of conditioned taste aversion: Effect of restriction in the intake of the conditioned taste stimulus

Latent inhibition of conditioned taste aversion (CTA) is sensitive to changes in the temporal context. A change in the time of day of conditioning with respect to the time of day of the preexposure can disrupt the latent inhibition. This contextual change in the time of day may reveal a temporal specificity of latent inhibition. The optimum procedure to induce this temporal specificity is not well established. For example, it has been shown that a long period of habituation to temporal contexts is one factor that can determine the effect. However, the experimental conditions on the conditioning day that facilitate this phenomenon are unknown. The aim of this study is to elucidate whether a restriction in the intake of the conditioned taste stimulus affects the temporal specificity of latent inhibition. Two main groups of Wistar rats were tested in a latent inhibition of CTA paradigm, in which the temporal specificity of this phenomenon was analyzed by a change in the time of day of conditioning. The intake of the taste stimulus was restricted in the conditioning day in one of the groups, but this restriction was not applied in the other group. The results indicated temporal specificity of latent inhibition only in the group without restriction, but not in the group with limitation in the intake of the taste stimulus during conditioning. These findings can help to elucidate the characteristics of the procedure to induce temporal specificity of latent inhibition.     

Conditioned taste aversions: A bibliography

A bibliographic list of 403 articles dealing specifically with conditioned taste aversions from 1950–1975 is provided. In addition, the references are classified according to six major categories in a topical index. The major categories are Parameters of Conditioning, Physiological Manipulations, Pharmacological Interventions, Methodology, Comparative, and General Information. References were obtained from individual journals in psychology, physiology, pharmacology, and animal behavior and were supplemented and extended byPsychological Abstracts. A final source of references was provided by individual researchers who contributed preprints and reprints.     

Effect of taste context and ambient context changes on successive negative contrast

Rats shifted from 32% sucrose to 4% sucrose lick less than rats that experience oniy the 4% solution. Previous experiments have found this negative contrast effect to be reduced (“disinhibited”) by the addition of a novel tone in the postshift period. In Experiment 1 of this paper, the negative contrast effect was enhanced when a novel flavor was added to the sucrose solution in the postshift period. In Experiments 2–4, changes in the ambient context, even changes sufficient to produce disruptions in licking, did not alter the degree of negative contrast. Tiese results suggest that (1) rats compare rewards across substantially different contexts, (2) contrast may serve to enhance taste neophobia, and (3) a disinhibitory effect may be confined to the presentation of punctate, nontaste, novel stimuli within a familiar context.     

The insensitivity of schedule-induced polydipsia to conditioned taste aversions: Effect of amount consumed during conditioning

In Experiment 1, animals poisoned following schedule-induced or prandial-induced saccharin consumption subsequently showed identical aversions to saccharin when tested under water deprivation. In Experiment 2, animals conditioned to avoid saccharin to similar levels under water deprivation were differentially affected when saccharin was subsequently presented on the baselines of schedule-induced and prandial-induced drinking. Together, these data indicate that the differential effects observed on schedule-induced and prandial-induced drinking when animals are poisoned following consumption under these two schedules do not reflect the differential acquisition of taste aversions, but instead reflect the differential tendencies to drink induced by the spaced and massed feedings.     

Effect of taste preexposure on taste and odor aversions

Thirsty Sprague-Dawley rats drank flavored water in a wind tunnel prior to lithium-induced toxicosis. Flavors were presented for 5 min; 30 min later a toxin, lithium chloride, was injected. After the rats had recovered, subsequent aversions to the taste and the odor were assessed separately. In Experiment 1, extensive preexposure to the taste component of the flavor attenuated neophobia to the flavor and the subsequent taste aversion. However, the subsequent odor aversion was unaffected. Experiment 2 partially replicated the results of Experiment 1 and showed that, in a situation in which only taste-potentiated odor aversions are usually found, nonpotentiated aversions were evident. Experiment 3 found that, in addition to attenuating taste aversions, taste preexposure enhances the capacity of rats to learn nonpotentiated odor aversions. The results are interpreted with a neural-based model of conditioned flavor aversions.     

Independence of neophobia and taste aversion learning

Animals were presented with (1) one solution which differed from that of the test solution, (2) a series of distinctly flavored solutions whose flavors differed from that of the test solution, or (3) with a flavored solution whose flavor was the same as that of the test solution. When animals received the solution whose flavor was the same as that of the test solution prior to a test for neophobia and prior to a conditioning trial, neophobia was reduced and aversions were weakened. However, when animals received a solution or a series of solutions whose flavors differed from that of the test solution, neophobia was reduced but conditioned aversions were unaffected. Presentations of solutions that differed from the test solution following aversion formation left the association between the taste of the test solution and the effects of the aversion-inducing treatment intact. In a final experiment it was discovered that neophobia was reduced as much when animals drank solutions whose flavors changed every third day as when they drank the same solution throughout testing.     

Learning and nonlearned neophobia enhancement both contribute to the formation of illness-induced taste aversions by deer mice (Peromyscus maniculatus bairdi)

An investigation was made of the occurrence of learned and nonlearned aversions in the acquisition of illness-induced taste aversions in mice of the genusPeromyscus. It was determined: (1) that illness following the ingestion of a novel flavor both produced aversions specific to that flavor and also enhanced neophobia directed toward novel flavors in general; (2) that the specific aversion and the enhanced neophobia appeared to be mediated by independent processes, with no indication that the enhanced neophobia was dependent upon the integrity of the specific aversion; and (3) that illness following the ingestion of familiar water produced enhanced neophobia, which did not appear to be mediated by an aversion to water. It was noted that the results were fundamentally in agreement with those previously obtained with laboratory rats, except that a demonstration of the independence between the two types of aversions has not yet been reported in those animals.     

The effects of conditioned taste aversions on schedule-induced polydipsia: An analysis of the initiation and postpellet temporal distribution of licking

Animals poisoned following the schedule-induced consumption of saccharin initially continued to drink following spaced pellet deliveries. Neither the initiation of postpellet drinking (i.e., bout, initiation) nor the size and duration of the bouts was effected by the conditioned aversion procedure. With repeated conditioning trials (i.e., repeated pairings of saccharin and LiCl), schedule induced drinking was eventually reduced. The specific components underlying schedule-induced consumption, however, were differentially affected by the aversion training. Specifically, the decrease in schedule-induced drinking was effected primarily by a decrease in licking occurring between 10 and 60 sec after pellet delivery. Bout initiation and licking immediately postpellet (i.e., within the first 10 sec following pellet delivery) were most resistant to suppression and appeared to be responsible for the relative insensitivity of schedule-induced drinking to conditioned taste aversions. The differential effects of taste aversion conditioning on individual components of elicited behavior are discussed.     

Variations in the retention of taste aversions: Evidence for retrieval competition

In six experiments, we examined taste and compound taste/taste aversions at different retention intervals. In Experiment 1, saccharin aversions were significantly weaker 1 day after conditioning than 21 days after conditioning. This effect was determined not to be caused by the aftereffects of illness or differential hydration. With the use of a saccharin/denatonium compound, Experiment 2 demonstrated overshadowing of a denatonium aversion at 21- and 1-day retention intervals, Experiment 4 showed a potentiated saccharin aversion only at the 21-day retention interval, and both Experiments 2 and 4 revealed that the aversion of the taste-only controls was stronger at the later retention interval. Experiments 3 and 5 demonstrated that the differences at the two retention intervals were not caused by unconditioned changes in taste preference. Finally, Experiment 6 showed that extinction of the conditioning environment prior to testing results in stronger saccharin aversions than occur in nonextinguished controls. Collectively, these experiments suggest that testing within a 24-h period after conditioning will result in significantly weaker taste aversions. Also, these results support a retrieval-competition explanation that may account for the weakened aversions at the 1-day testing interval of both groups conditioned to single elements and those conditioned to compounds.     

Cross-modality transfer effects in conditioning-enhanced neophobia in chicks (Gallus domesticus): Evidence for the separability of novelty from specific stimulus characteristics

In this study, we investigated the separability of novelty from specific stimulus characteristics (e.g., color or taste quality) in the transfer of aversion effects. Ninety-six chicks (Gallus domesticus) received a novel visual (red water) or taste (3.0% vinegar) CS paired with an injection of lithium chloride or saline. The chicks were then tested for aversion to the CS or for conditioning-enhanced neophobia in response to a different novel visual cue (green water) or taste cue (1.0% saline). Aversions to the CSs were reliable and similar to each other. Reliable evidence of conditioning-enhanced neophobia occurred with respect to each test stimulus, irrespective of the type of CS, but conditioning with the vinegar CS produced reliably greater enhancement of neophobia than did conditioning with red water. For each CS, postconditioning neophobia was more persistent in testing with saline than with green water. The results for postconditioning neophobia suggested that novelty is a general stimulus property that is separable from specific stimulus characteristics.     

The effect of morphine preexposure on the acquisition of morphine-induced taste aversions: A nonassociative effect

In Experiment 1, the amount of time rats spent engaged in a range of behaviors was recorded immediately prior to and following the intraperitoneal administration of morphine sulfate (6 mg/kg) or distilled water. No behavioral differences were observed between these groups. In Experiment 2, preexposure to this low dose of morphine attenuated the subsequent acquisition of a morphine-induced taste aversion independent of the similarity of the preexposure and conditioning environments. These results with a dose of morphine that does not produce any behavioral effects, which in turn could potentially mask associative conditioning during preexposure, confirm that the attenuating effects of morphine preexposure on taste aversion learning are nonassociative.