Printer Friendly

US postexposure effect on conditioned flavor preference in the rat.

After a flavor is paired with a substance such as sucrose, rats may acquire preference for that flavor rather than other flavors. Recent studies (see Sclafani, 1990, for a review) demonstrate that this learning is robust and they conceptualize it as an example of Pavlovian conditioning, where the flavor is a conditioned stimulus (cs) and the paired substance is an unconditioned stimulus (us). Most of these studies concentrate on which aspect (taste or nutritional consequence) of the US causes the conditioned preference. In contrast, comparatively little is known about the content of the association acquired in this paradigm.

Fedorchak and Bolles (1987) found that food satiation treatment after flavor-nutrient pairing attenuates conditioned flavor preference (see also Capaldi, 1991; Capaldi, Owens, & Palmer, 1994; Deems, Oetting, Sherman, & Garcia, 1986; Mehiel, 1991). This finding is consistent with the results reported in other Pavlovian first-order conditioning preparations (e.g., Cleland & Davey, 1982; Hilliard & Domjan, 1995; Holland, 1981; Holland & Rescorla, 1975; Stanhope, 1989, 1992; Weingarten & Martin, 1989) that postsatiation of the US attenuates conditioned responding and suggests that rats associate a flavor with the stimulus properties of the nutrient. This is because, if all that the rats had learned was an association between the flavor and reaction evoked by the nutrient, the satiation treatment should not affect the flavor preference. Thus, the rats have to maintain some information about the stimulus properties of the nutrient and make conditioned responses according to the current status of the nutrient.

In the studies described above, the value of a US was modified by satiation of the subjects' need. The experiments reported here used a different postconditioning manipulation of the US representation: presentation of a US itself after pairing training. Several studies demonstrate that US postexposure affects conditioned responding established by the pairing (see Delamater & LoLordo, 1991, for a review). For example, Rescorla (1973, 1974) showed that postexposure of the same aversive US used in conditioning attenuated conditioned fear in the test (see also Overmier, Payne, Brackbill, Linder, & Lawry, 1979; Randich & Haggard, 1983; Randich & Rescoria, 1981). The similar attenuation effect was reported in conditioned taste aversion (Colby & Smith, 1977; Mikulka, Leard, & Klein, 1977). One account of the US postexposure effect is that US presentation habituates the US representation and that the degraded US representation evokes less conditioned responding (Rescoria, 1973, 1974).

To our knowledge, there is no study examining the US postexposure effect on conditioned flavor preference. If the postexposure of a US attenuates conditioned flavor preference, then it supports the view that the current status of the US representation determines the performance. That is, information about the US representation must be kept in this learning. Three experiments reported here explored this possibility.

Experiment 1

As described above, postexposure of the same US used in conditioning attenuates conditioned responding. In Experiment 1, after a flavor was paired with sucrose solution, the same concentration of sucrose solution was exposed and then conditioned flavor preference was tested. Control rats received the same treatment, but tap water rather than sucrose solution was postexposed for them. Less flavor preference in the former group would indicate the US postexposure effect is similar to that observed in conditioned fear and taste aversion (Colby & Smith, 1977; Mikulka et al., 1977; Overmier et al., 1979; Randich & Haggard, 1983; Randich & Rescorla, 1981; Rescorla, 1973, 1974).

In addition to these groups, two groups of rats received different sucrose solutions as postexposure treatments. For them, the concentration of the solution was either lower or higher than that used in the pairing phase. Concentration is assumed to be an intensity dimension of the reinforcing property of sucrose solution for hungry rats: the higher the concentration, the more the intensity (e.g., Collier & Bolles, 1968). Literature on the effect of a less or more intense US postexposure is limited to fear conditioning (but, see DeCola & Fanselow, 1995). Postexposure of a less intense shock than that used in conditioning attenuates conditioned fear (Hendersen, 1985; Randich & Haggard, 1983; Randich & Rescorla, 1981) and this effect is sometimes larger than the effect of postexposure of the same intense shock ("deflation" effect: Randich & Rescorla, 1981).

The effect of a more intense shock US is complicated. Some researchers (Bouton, 1984; Hendersen, 1985; Rescorla, 1974; Sherman, 1978) have demonstrated "inflation" of the fear, compared with no shock treatment (see also DeCola & Fanselow, 1995, for a demonstration of inflation in taste aversion). Randich and Rescorla (1981) observed the same effect but only on the later trials, and Randich and Haggard (1983) reported less fear throughout the testing. However, the fear induced by postexposure of more intense shocks was stronger than that of the same intense shock in almost all studies where this comparison was available (Randich & Rescoda, 1981; Rescorla, 1974; Sherman, 1978). Although Randich and Haggard (1983) reported the opposite pattern on the first trial, this pattern was reversed on the later trials.

Deflation and inflation effects have been interpreted by an eventmemory model (Randich & Rescorla, 1981; Rescorla, 1974). Postexposure of a less or more intense US replaces the original US and the modified US representation evokes less or more conditioned responding.

Thus, the results of these studies and the event-memory model predict that the concentration (i.e., intensity) of postexposed sucrose solution determines conditioned flavor preference in testing: the lower the concentration, the more the attenuation by the postexposure. These sucrose-exposure groups were also compared with a water-exposed control group.

Method

Subjects. The subjects were two squads of 24 male Wistar rats with histories of conditioned suppression training under water deprivation. The assignment of the rats to groups in the present experiment was orthogonal. Prior to this experiment, they were maintained in ad-lib food and water schedules for 2 weeks and then their bodyweights were reduced to 80-85% by limited feeding. On the first conditioning day of this experiment, the first squad was approximately 140 days old and 415-580 g and the nd was approximately 120 days old and 406-550 g.

Apparatus. The rats were housed in individual hanging cages (30 cm long, 20 cm wide, and 19 cm high) and training and tests were conducted there. Usually, ad-lib tap water was provided by a nozzle protruding from a small hole in the back wall of each cage. When necessary, a metal cover of the wall was used to prevent drinking from the nozzle. Experimental fluids were presented from the front wall by plastic bottles with metal spouts, which protruded into the cages at approximately 6 cm above the mesh floors. Fluids were flavored either with vanilla essence (Kyoritsu Shokuhin, Japan) or with strawberry essence (Asaoka Koushinryou, Japan) by adding 2 ml of essence into 100-ml tap water. A series of fluorescent lamps in the animal room provided illumination from 08:00 hrs to 20:00 hrs.

Procedure. The experiment was run in two identical replications, the first with Squad I and the second with Squad 2. At the outset, the rats of each squad were assigned in equal numbers to four groups (Groups 0%, 2.5%, 5%, 10%). Each daily treatment started at 19:00 hrs.

Flavor-sucrose pairing training continued for 4 days. The nozzle of the back wall was made inaccessible 30 rain before each session. Then two bottles were concurrently set for each cage (distance between the two spouts was 5 cm). One bottle contained 5% (w/w) sucrose diluted in tap water flavored with Essence A, and the other contained tap water flavored with Essence B. For half of the rats in each group, A was vanilla and B was strawberry. The identities of essence were switched for the remaining rats. All rats were allowed to drink from the two bottles for 20 min. After the session, some food blocks were put into each cage to maintain the rats' bodyweights, and tap water was made ad-lib by allowing access to the nozzle of the back wall. The positions (left or right) of the bottles were counterbalanced within each group and alternated across days.

On the next 4 days of postexposure, all rats were allowed 20-rain access to a single bottle. The bottle contained unflavored tap water for Group 0% and unflavored sucrose solution for Groups 2.5%, 5%, and 10%. Concentration of the solution used for each of these three groups was designated by each group name. Access to the nozzle of the back wall and postsession feeding were the same as in the pairing phase.

After the last session of the postexposure, some food blocks were given and ad-lib access to the nozzle of the back wall was prohibited. The nozzle was never accessible during the next 2 days of testing: The test was conducted under food- and water-deprived conditions. The intention of the water deprivation was to lessen the group difference in the total amount of fluid intake in the test. Flavor preference was tested on each day by concurrent presentation of two bottles for 20 min. One bottle contained tap water flavored with A, the other, tap water flavored with B. The positions of the bottles were alternated across these 2 days.

Measurement and data analyses. In this and subsequent experiments, the amount of fluid intake was measured by weighing each bottle using an electric balance to the nearest 0.1 g. Our major interest was the flavor preference in the testing and it was indexed by ratios of the form a/(a+b) where "a" was the intake of fluid flavored with A and "b" was that of fluid flavored with B. Because Flavor A was paired with sucrose, values higher than 0.5 indicate preference for the flavor paired with sucrose. This index was also calculated for the data from the pairing phase, and in that case it reflects unconditioned preference for the sucrose over the tap water in addition to the conditioned flavor preference. In addition, the amounts of fluid intake in each phase were analyzed. Split-plot factorial analyses of variances (ANOVAs) were performed on the preference ratios and the intakes, using the 0.05 significance level. For simplicity, only significant F values will be reported below.

Results

Although the procedures for the two squads were identical, there were some differences in their results. The data were described separately when this factor was significant. We have no proper account of the replication effects, but as described below this factor did not change our conclusion about the postexposure effect on conditioned flavor preference.

Preference ratios during the flavor-sucrose pairing phase were analyzed with an ANOVA with replication, group, and day (repeated) as factors. The effects of replication, F(1, 40) = 14.61, and day, F(3, 120) = 27.18, were significant, but the effect of group and all interactions were not. Thus all groups showed equal preference for the flavored fluids containing sucrose. The ratios of these days were shown in the left portion of each panel of Figure 1: The left panel is for the first squad and the right panel for the second. Higher preference ratios on the later days suggest acquisition of conditioned flavor preference and/or dissipation of neophobia about flavored sucrose solution.

An ANOVA on the total amount of fluid intake yielded significant effects of replication, F(1, 40) = 4.10, and day, F(3, 120) = 92.31, and a nonsignificant group effect and nonsignificant interactions. The intakes of 4 days of each squad were shown in the left portion of each panel of Figure 2. More total intakes on the later days are probably caused by adaptation to drinking from the flavored bottles.

The center portion of each panel of Figure 2 shows intakes of fluid during the postexposure. An ANOVA yielded significant main effects of replication, F(1,40) = 21.92, group, F(3, 40) = 102.45, and day, F(3, 120) = 7.86, and significant Replication x Group, F(3, 40) = 7.40, Group x Day, F(9, 120) = 3.90, and Replication x Group x Day, F(9, 120) = 2.45, interactions. The Replication x Day interaction was not significant. Newman-Keuls post-hoc tests revealed that Group 0% drank significantly less than the other groups on every day in the both squads. Other significant group differences were: Group 5% [greater than] Group 2.5% (Days 1 and 3 of the first squad and Day 2 of the second), Group 5% [greater than] Group 10% (Days 2, 3, and 4 of the first squad), and Group 10% [greater than] Group 2.5% (Days 2 and 4 of the second squad). The lower consumption of fluid in Group 0% than the other groups was expected, because the fluid of this group was tap water and that of the others was sucrose solution. An unexpected result was that Group 5% drank more than Group 10% in the first squad despite its lower concentration of sucrose. However, there was no difference between these groups in the second squad, so that the differences in the first squad were perhaps caused by a sampling error.

The results of the most interest, flavor preference on the 2 test days, are shown in the right portion of each panel of Figure 1. All rats showed preference for the flavor which had been paired with the sucrose over the unpaired flavor. An ANOVA revealed that the preference ratios were affected by the effects of replication, F(1, 40) = 15.78, and group, F(3, 40) = 7.57. The main effect of day and all interactions were nonsignificant. Newman-Keuls post-hoc tests showed that Groups 0% and 2,5% preferred the conditioned flavor more than Groups 5% and 10%. The other group comparisons were not significant. Although Figure 1 suggests a difference between Groups 0% and 2.5% in the first squad, a Newman-Keuls test of this difference revealed no significance. Thus, postexposure of the sucrose with high concentration attenuated the conditioned flavor preference which had been established by pairing the flavor and the sucrose.

The total fluid intake on the tests days were shown in the right portion of Figure 2. An ANOVA yielded a significant Group x Day interaction, F(3, 40) = 5.04. Neither main effects nor the other interactions were significant. Newman-Keuls post-hoc tests revealed that Group 10% drank more than Groups 5% and 2.5% on the 1st test day. There were no other significant group differences.

Discussion

Attenuation of conditioned performance by postexposure of the same US reported in other Pavlovian preparations (Colby & Smith, 1977; Mikulka et al., 1977; Overmier et al, 1979; Randich & Haggard, 1983; Randich & Rescorla, 1981; Rescorla, 1973, 1974) was demonstrated in conditioned flavor preference. Postexposure of sucrose solution attenuated the flavor preference established by flavor-sucrose pairing in solution, when the concentration of the postexposed solution was the same (5%).

When the concentration of sucrose solution used in the postexposure phase was lower (2.5%) than that used in the pairing phase (5%), there was no attenuation of flavor preference. Previous studies on conditioned fear demonstrated attenuation of the fear by postexposure of a less intense US (Hendersen, 1985; Randich & Haggard, 1983; Randich & Rescorla, 1981). Furthermore, Randich and Rescorla (1981) reported deflation of conditioned fear: larger attenuation than that produced by postexposure of the same intensity US. The disagreement between these studies and our results of neither attenuation nor deflation effects will be treated in the General Discussion section.

Postexposure of 10% sucrose solution attenuated conditioned flavor preference as that of 5% sucrose solution did. However, studies on conditioned fear (Bouton, 1984; Hendersen, 1985; Randich & Rescorla, 1981; Rescorla, 1974; Sherman, 1978; but see Randich & Haggard, 1983) and taste aversion (DeCola & Fanselow, 1995) have found that postexposure of a more intense US inflates conditioned performance. Thus, one should predict the inflation of flavor preference by postexposure of a more concentrated sucrose solution. One possibility for our failure to observe this inflation effect in Experiment 1 is that the 10% solution was not intense enough to demonstrate this effect. The amount of intake in the postexposure phase might have suggested this possibility: The intake of 10% solution was not larger than that of 5% solution.

Experiment 2

In Experiment 2, conditioned performance of three groups of rats were compared. Two of them received the same treatment as that for Groups 0% and 5% in Experiment 1, respectively. A more concentrated sucrose solution (20%) in the postexposure was used for the remaining group to maximize the possible inflation of conditioned flavor preference. In addition, preference for 20% solution over 5% solution was examined after the flavor preference testing to confirm that the former is more intense than the latter.

Method

Subjects and apparatus. The subjects were 24 male Wistar rats with histories of escape training under water deprivation. Prior to this experiment, they were maintained in ad-lib food and water schedules for 4 weeks and then their bodyweights were reduced to 80-85% by limited feeding. On the first conditioning day of this experiment, they were approximately 140 days old and 419-525 g. The same kinds of home cages, bottles, and flavor essences as in Experiment 1 were used.

Procedure. Three groups (Groups 0%, 5%, and 20%) of 8 rats received flavor-sucrose pairing, sucrose postexposure, and flavor preference tests. In this experiment, the rats were assigned to these groups after the pairing phase, matched in the total intake and in the preference ratio. The assignment was also orthogonal with respect to their previous histories. Procedures for Groups 0% and 5% were exactly the same as in Experiment 1. Group 20% received similar treatments to these groups, but 20% sucrose solution was used in the postexposure phase. That is, four flavor-sucrose pairing days were followed by four 0%, 5%, or 20% sucrose postexposure days and by two flavor preference test days.

After the second test session, ad-lib access to the nozzle of the back wall was allowed again, and the rats were maintained under this condition for the next 3 days. The first sucrose preference test with two bottles was conducted on the middle day of this nonthirsty phase. Then, access to the nozzle of the back wall was prohibited again and the second test of sucrose preference was conducted on the next day under this thirsty condition as in the flavor preference tests. Note that all rats in this study were under food deprivation throughout and that "nonthirsty" means being under food deprivation and "thirsty" means being under food and water deprivation. Procedures of the sucrose preference tests were the same as of the flavor preference tests except that one bottle contained unflavored 5% sucrose solution and the other unflavored 20% sucrose solution. The positions of the bottle were counterbalanced across rats and were alternated across days.

Results

Preference ratios during the flavor-sucrose pairing phase were analyzed with an ANOVA with group and day (repeated) as factors. Only the effect of day was significant, F(3, 63) = 46.06. The ratios of 4 days were shown in the left panel of Figure 3. An ANOVA on the total amount of fluid intake also yielded only a significant effect of day, F(3, 63) = 97.99. The intakes of 4 days were shown in the left portion of Figure 4. Because the groups were matched in these measures after this phase, the nonsignificant group effect and the nonsignificant interaction were not surprising. The higher preference ratios and the more intakes in the later days of this phase accorded with the results in the corresponding phase of Experiment 1.

The center portion of Figure 4 shows intakes of fluid during the postexposure. An ANOVA yielded a significant effect of group, F(2, 21) = 60.85, and a significant Group x Day interaction, F(6, 63) = 3.93. The main effect of day was not significant. Newman-Keuls post-hoc tests revealed that Group 0% drank significantly less than the other groups in all days and that Group 20% drank less than Group 5% on the 1st day. The former result was also demonstrated in Experiment 1 and it reflects unconditioned preference for sucrose solution over tap water. The latter result might have been from neophobia about 20% solution, because this solution was novel for the rats.

The results of the most interest, flavor preferences in the testing, are shown in the right panel of Figure 3. All rats except one showed preference for the flavor which had been paired with sucrose over the unpaired flavor. The exception was a rat of Group 20%, which showed less preference for the paired flavor on the first test. An ANOVA yielded a significant effect of day, F(1, 21) = 5.40, and a significant Group x Day interaction, F'(2, 21) = 8.90. The effect of group was not significant. Newman-Keuls post-hoc tests showed that on the 1st day both Groups 0% and 5% preferred the conditioned flavor more than Groups 20%. The difference between Groups 0% and 5% on this day was marginal, p = 0.054. Thus, postexposure of sucrose solution with higher concentration than that used in the pairing phase enhanced the attenuation effect on conditioned flavor preference.

The right portion of Figure 4 showed the total fluid intake in the flavor testing. An ANOVA yielded nonsignificant main effects. The Group x Day interaction was also nonsignificant.

All rats showed preference for 20% sucrose over 5% sucrose on the nonthirsty day (a binomial test, p [less than] 0.0001), and all except one of Group 20% showed the same preference on the thirsty day, p [less than] 0.0001. Mean preference ratios under the nonthirsty and thirsty days were 0.81 and 0.75, respectively. An ANOVA on this measure yielded a significant effect of day, F(1, 21) = 6.13, but the effect of group and the Group x Day interaction were nonsignificant. The total intakes on the nonthirsty day were 23.6, 30.7, and 30.1 g for Groups 0%, 5%, and 20%, respectively. The corresponding values on the thirsty day were 29.3, 33.3, and 35.3 g. An ANOVA on this measure yielded significant effects of group, F(2, 21) = 8.08, and day, F(1, 21) = 23.45. The Group x Day interaction was nonsignificant. Newman-Keuls post-hoc tests revealed significantly less intake in Group 0% than the other groups, implying neophobia about unflavored sucrose solution in this group. The lower value of the preference ratio and the more intake on the thirsty day than the nonthirsty day suggest that preference for high concentrated sucrose solution was hindered by motivation to drink any fluid.

Discussion

Postexposure of sucrose solution after flavor-sucrose pairing in solution attenuated conditioned flavor preference and this effect was positively related to the concentration of the sucrose: The highly concentrated (20%) solution induced more attenuation than the same (5%) solution used in the pairing phase, though the effect was short-lived and was observed only on the first test day. Preference for 20% solution over 5% solution demonstrated in the subsequent tests suggests that the former was a more intense US.

Studies on conditioned fear and taste aversion have shown inflation of performance by postexposure of a more intense US (Bouton, 1984; Hendersen, 1985; DeCola & Fanselow, 1995; Randich & Rescorla, 1981; Rescorla, 1974; Sherman, 1978). Thus the prediction from these studies was that postexposure of highly concentrated sucrose solution should enhance conditioned flavor preference. Our results revealed the opposite. Randich and Haggard's (1983) results, however, accord with ours. They observed more attenuation of conditioned fear by postexposure of a more intense US on the first test trial. This issue will be treated in the General Discussion section.

Experiment 3

Randich and Rescorla (1981) demonstrated that in a conditioned suppression preparation at least a portion of the US postexposure effect is caused by erasure of an "incubation" effect. With an electric shock as a US, time passage after the last conditioning trial increases conditioned fear (McMichael, 1966). They attributed the cause of this incubation effect to spontaneous recovery of US representation which had been degraded by the repeated presentation during conditioning. The US postexposure redegrades the US representation and decreases conditioned responding.

To our knowledge, there is no study that reports incubation in conditioned flavor preference. But, incubation has been observed in conditioned taste aversion (Biederman, Milgram, Heighington, Stockman, & O'Neill, 1974; Marcant, Schmaltz, Roy, & Leconte, 1985). In Experiment 3, we tested whether erasure of incubation is a cause of the attenuation of conditioned flavor preference observed in the preceding experiments. According to this hypothesis, the sucrose representation which has been degraded during the flavor-sucrose pairing phase would recover in the postexposure phase for the rats that drank tap water, but sucrose postexposure would erase this incubation effect.

Three groups received flavor-sucrose pairings. One of them (Group Imm) was tested the day after the last pairing. The remaining two groups were tested after 4-day postexposure of tap water (Group 0%) or sucrose solution (Group 5%). Thus the treatments of these two groups were the same as those of the corresponding groups of the previous experiments. If the attenuation effect of postexposure is caused by erasure of the incubation, flavor preference at testing should be largest in Group 0% and smallest in Group Imm. Group 5% should be between these groups depending on the effectiveness of the postexposure: Perfect erasure should make it indistinguishable from Group Imm.

In contrast, if there is no incubation effect in this setting, conditioned preference of Group Imm should be the same as that of Group 0% and more than that of Group 5%. Finally, if the time passage simply attenuates conditioning, Group Imm should show the best performance because it received the testing earlier than Groups 0% and 5%.

Method

Subjects and apparatus. The subjects were 18 male Wistar rats with histories of conditioned suppression training under water deprivation. The assignment of the rats to groups in the present experiment was orthogonal. Prior to this experiment, they were maintained in ad-lib food and water schedules for 2 weeks and then their bodyweights were reduced to 80-85% by limited feeding. On the first pairing day of this experiment, they were approximately 140 days old and 391-678 g. The same kind of home cages, bottles, and flavor essences as in the previous experiments were used.

Procedure. The rats were assigned in equal numbers to three groups (Groups Imm, 0%, and 5%) at the outset. Procedures for Groups 0% and 5% were the same as in Experiments 1 and 2. Group Imm animals received the same pairing treatment but they were tested on the next 2 days. For this group, access to the nozzle of the back wall was prohibited after the last session of pairing: They were tested under a food- and water-deprived condition as the rats of Groups 0% and 5% were.

Data analyses. As in the previous experiments, split-plot factorial ANOVAs were applied to all measures acquired. However, planned orthogonal comparisons with F statistics were executed for the preference ratios of the testing, because the three explicit hypotheses on the test performance were in question. A rejection criterion was set at p [less than] 0.05 as before.

Results

Preference ratios during the flavor-sucrose pairing phase were analyzed with an ANOVA with group and day (repeated) as factors. Only the effect of day was significant, F(3, 45) = 9.11. Thus all groups showed equal preference for the flavored fluid containing sucrose. The ratios of 4 days were shown in the left panel of Figure 5. An ANOVA on the total amount of fluid intake also revealed only a significant effect of day, F(3, 45) = 53.01. The intakes were shown in the left portion of Figure 6. These patterns of the preference ratios and the intakes were the same as in Experiments 1 and 2.

Groups 0% and 5% received the postexposure and the former drank more than the latter during the training: The fluid intake of each day of this 4-day phase was shown in the center portion of Figure 6. An ANOVA yielded significant effects of group, F(1, 10) = 118.24, and day, F(3, 30) = 3.28, and a significant Group x Day interaction, F(3, 30) = 4.86.

The results of the most interest, flavor preferences in the testing, are shown in the right panel of Figure 5. All rats showed preference for the flavor which had been paired with the sucrose over the unpaired flavor. An ANOVA yielded a marginal effect of group, F(2, 15) = 3.03, p = 0.08. The effect of day and the Group x Day interaction were nonsignificant. As described before, this experiment was designed to test specific hypotheses, so planned comparisons were executed. Pairwise comparisons revealed significant difference only between Groups 5% and Imm, F(1, 15) = 5.88. The nonpairwise contrast between the data of Groups Imm and 0% and those of Group 5% was significant, F(1, 15) = 5.36, but the other two nonpairwise contrasts (Group Imm vs. Groups 0% and 5%; Groups Imm and 5% vs. Group 0%) were not. These results suggest that the magnitude of conditioned flavor preference was Group Imm = Group 0% [greater than] Group 5%.

An ANOVA on the total fluid intake in the testing yielded a significant effect of day, F(1, 15) = 9.55. The effect of group and the Group x Day interaction were nonsignificant. The fluid intake on the 2 test days were shown in the right portion of Figure 6.

Discussion

This experiment showed that the effect of the 4-day passage on conditioned flavor preference was negligible, but that the sucrose postexposure attenuates the flavor preference. These results signify that the underlying mechanism of the sucrose postexposure is not caused by erasure of the sucrose representation incubation during the postexposure, because if there had been any incubation effect in this setting Group 0% would have shown more flavor preference than Group Imm did.

General Discussion

Conditioned flavor preference was demonstrated in our setting because all rats except one showed preference for the flavor which had been paired with the sucrose over the unpaired flavor. Compared with other studies in this field, the conditioned preference observed here was small. A cause of this may be that the conditioned preference was tested under water deprivation. Motivation to drink any fluid might have made detection of conditioned preference difficult.

Although this possibility had been noticed, we were afraid that testing without water deprivation would result in group differences in the total intake on the test days. During the postexposure phase, Group 0% drank less than the other groups. Thus, it was very likely that Group 0% would be reluctant to drink from the bottles in the testing, resulting in less intake in this group. Although a preference ratio was used as an index to standardize conditioned preference, a similar amount of total intake among groups makes the conclusion sounder. Our procedure of water deprivation was successful, because the total intakes in the testing were almost the same among groups. Only one exception was Group 10% in Experiment 1, which drank more than the other two postexposed groups on the 1st test day only.

Our finding of a US postexposure effect in conditioned flavor preference indicates the generality of this effect across different preparations for learning and it gives additional support to the claim that conditioned flavor preference is a form of Pavlovian conditioning.

An interesting result of this study is that the attenuation effect of US postexposure on conditioned performance was positively related to US intensity (i.e., concentration of sucrose solution). Though 2.5% solution did not attenuate conditioned preference compared with a control condition, 5% and 10% solution attenuated it, and attenuation induced by 20% solution was much more.

Studies about conditioned fear, however, have reported a negative relationship between the attenuation effect and US intensity (Hendersen, 1985; Randich & Rescorla, 1981; Rescorla, 1974; Sherman, 1978). Furthermore, postexposure of a more intense US inflates conditioned fear compared with no US treatment (Bouton, 1984; Hendersen, 1985; Randich & Rescorla, 1981; Rescorla, 1974; Sherman, 1978; see also DeCola & Fanselow, 1995).

Randich and Haggard (1983) observed more attenuation of conditioned fear on the first test trial induced by postexposure of a more intense US. Delamater and LoLordo (1991) speculated that habituation occurred in all groups of Randich and Haggard's experiment and that it was positively related to US intensity (cf. Davis & Wagner, 1968). As habituation of a US degrades the US representation and evokes less conditioned responding (Rescorla, 1973), intensity of a postexposed US should positively relate to attenuation of conditioned performance. This account clearly fits the results we reported here.

There are other accounts of the attenuation effect on conditioned performance of US postexposure. One of them, erasure of incubation, was rejected in Experiment 3 because the incubation was not observed in our setting.

Another account of the attenuation effect is that US postexposure makes the background context more excitatory and this context excitation hampers conditioned performance. In our setting, all procedures were conducted in the home cages where the rats lived. Although excitation of the home cage would have been extinguished between sessions, bottle stimuli might have played the contextual role (cf. Archer, Sjoden, Nilsson, & Carter, 1980). Thus, attenuation of conditioned performance in postexposure groups could be caused by additional excitation of both bottles, resulting in nondifferential performance. However, if the bottle stimuli had acquired some excitation by sucrose postexposure, the total fluid intake should have been more in the postexposure groups than in nonexposed rats. This was not the case, and this failure to detect a group difference cannot be ascribed to ceiling effect because rats drank much more in the pairing phase. It is therefore unlikely that concealment of conditioned preference by context excitation is the principal cause of the US postexposure effect reported here.

Attenuation of conditioned performance by US postexposure is predicted by the comparator hypothesis (Miller & Matzel, 1988). According to this hypothesis, conditioned excitation is observed when the associative strength of a CS is larger than that of a comparator stimulus. If the background context is considered as the comparator stimulus, increment of its associative strength by US postexposure makes conditioned excitation weak. This hypothesis, however, cannot account for the attenuation of conditioned flavor preference reported in this study, because the conditioned preference was indexed by a preference ratio, a/(a+b), where "a" was the intake of a sucrose-paired flavor and "b" was that of an unpaired flavor. If conditioned intake is a difference between associative strength of each flavor ([V.sub.A] or [V.sub.B]) and that of the context ([V.sub.C]), an increment of the latter should increase, rather than decrease, the preference ratio, ([V.sub.A]-[V.sub.C])/[([V.sub.A]-[V.sub.C])+([V.sub.B]-[V.sub.C])]. If conditioned intake is a quotient of associative strength of each flavor and that of the context, an increment of the latter should have no effect on the preference ratio, ([V.sub.A]/[V.sub.C])/[([V.sub.A]/[V.sub.C])+([V.sub.B]/[V.sub.C])]. Thus, the comparator hypothesis cannot account for decrement of the preference ratio reported here, as long as its comparison is based upon a simple subtraction or division rule.

An alternative successful account of the US postexposure effect might be derived from a within-compound learning view (Rescorla & Durlach, 1981). Simultaneous pairing of two stimuli may result in a bidirectional association between these stimuli and postexposure of one stimulus may break the association (Rescorla & Freberg, 1978). Thus, postexposure of a US may attenuate conditioned performance, as postexposure of a CS does (extinction).

A related, but more cognitive, account of US postexposure effect has been proposed by Baker and Mercier (1989). They have claimed that animals may include all of the experimental episodes in the analysis of contingency between a CS and a US. Postexposure of the US degrades the overall contingency retrospectively, resulting in attenuation of conditioned performance.

Although we cannot determine which of these three accounts - US habituation (Rescorla, 1973), within-compound learning (Rescorla & Durlach, 1981), or retrospective processing (Baker & Mercier, 1989) - is adequate for the attenuation of conditioned performance by US postexposure, demonstration of the effect on conditioned flavor preference means that the rats maintain some information about the US representation after conditioning and show flavor preference according to the current status of the US. Furthermore, the positive relationship observed here between US intensity and the attenuation effect calls for further exploration of US postexposure effect in other conditioning preparations than fear conditioning with rats.

Requests for reprints should be sent to Sadahiko Nakajima, who is now at the Department of Psychology, Kwansei Gakuin University, Uegahara, Nishinomiya, Hyogo 662, Japan.

Nobuyuki Kawai conducted the experiments and the statistical analyses, and Sadahiko Nakajima conceived the design and prepared the article. Portions of the study were reported at the 7th Meeting of the International Society for Comparative Psychology in August, 1996. We were supported by the JSPS Fellowship for the Japanese Junior Scientists and the JSPS Postdoctoral Fellowship for Research Abroad, respectively. We are grateful to K. Matthew Lattal for comments on the manuscript and for refinement of English expression, and to H. Imada for encouragement of the study.

References

ARCHER, T., SJODEN, P. O., NILSSON, L. G., & CARTER, N. (1980). Exteroceptive context in taste-aversion conditioning and extinction: Odour, cage, and bottle stimuli. Quarterly Journal of Experimental Psychology, 32, 197-214.

BAKER, A. G., & MERCIER, P. (1989). Attention, retrospective processing and cognitive representations. In S. B. Klein & R. R. Mowrer (Eds.), Contemporary learning theories: Pavlovian conditioning and the status of traditional learning theory (pp. 85-116). Hillsdale, N J: Erlbaum.

BIEDERMAN, G. B, MILGRAM, N. W., HEIGHINGTON, G. A., STOCKMAN, S. M., & O'NEILL, W. (1974). Memory of conditioned food aversion follows a U-shape function in rats. Quarterly Journal of Experimental Psychology, 26, 610-615.

BOUTON, M. E. (1984). Differential control by context in the inflation and reinstatement paradigms. Journal of Experimental Psychology: Animal Behavior Processes, 10, 56-74.

CAPALDI, E. D. (1991). Hunger and the learning of flavor preferences. In R. C. Bolles (Ed.), The hedonics of taste (pp. 127-142). Hillsdale, N J: Erlbaum.

CAPALDI, E. D., OWENS, J., & PALMER, K. A. (1994). Effects of food deprivation on learning and expression of flavor preferences conditioned by saccharin or sucrose. Animal Learning & Behavior, 22, 173-180.

CLELAND, G. G., & DAVEY, G. C. L. (1982). The effects of satiation and reinforcer devaluation on signal-centered behavior in the rat. Learning and Motivation, 13, 343-360.

COLBY, J. J., & SMITH, N. F. (1977). The effect of three procedures for eliminating a conditioned taste aversion in the rat. Learning and Motivation, 8, 404-413.

COLLIER, G., & BOLLES, R. (1968). Hunger, thirst, and their interaction as determinants of sucrose consumption. Journal of Comparative and Physiological Psychology, 66, 633-641.

DAVIS, M., & WAGNER, A. R. (1968). Startle responsiveness after habituation to different intensities of tone. Psychonomic Science, 12, 337-338.

DeCOLA, J.P., & FANSELOW, M. S. (1995). Differential inflation with short and long CS-US intervals: Evidence of a nonassociative process in long-delay taste avoidance. Animal Learning & Behavior, 23, 154-163.

DEEMS, D. A., OETTING, R. L., SHERMAN, J. E., & GARCIA, J. (1986). Hungry, but not thirsty, rats prefer flavors paired with ethanol. Physiology and Behavior, 36, 141-144.

DELAMATER, A. R., & LoLORDO, V. M. (1991). Event revaluation procedures and associative structures in Pavlovian conditioning. In L. Dachowski & C. F. Flaherty (Eds.), Current topics in animal learning: Brain, emotion, and cognition (pp. 55-94). Hillsdale, NJ: Erlbaum.

FEDORCHAK, P. M., & BOLLES, R. C. (1987). Hunger enhances the expression of calorie- but not taste-mediated conditioned flavor preferences. Journal of Experimental Psychology: Animal Behavior Processes, 13, 73-79.

HENDERSEN, R. W. (1985). Fearful memories: The motivational significance of forgetting. In F. R. Brush & J. B. Overmier (Eds.), Affect, conditioning, and cognition: Essays on the determinants of behavior (pp. 43-53). Hillsdale, N J: Erlbaum.

HILLlARD, S., & DOM JAN, M. (1995). Effects on sexual conditioning of devaluing the US through satiation. Quarterly Journal of Experimental Psychology, 48B, 84-92.

HOLLAND, P. C. (1981). The effect of satiation after first- and second-order appetitive conditioning. Pavlovian Journal of Biological Science, 16, 18-24.

HOLLAND, P. C., & RESCORLA, R. A. (1975). The effect of two ways of devaluing the unconditioned stimulus after first- and second-order appetitive conditioning. Journal of Experimental Psychology: Animal Behavior Processes, 1,355-363.

MARCANT, P., SCHMALTZ, G., ROY, J. C., & LECONTE, R (1985). Long-term retention of a poorly learned saccharin aversion: Evidence for an incubation effect. Behavioural Processes, 11, 15-20.

McMICHAEL, J. S. (1966). Incubation of anxiety and instrumental behavior. Journal of Comparative and Physiological Psychology, 61,208-211.

MEHIEL, R. (1991). Hedonic-shift conditioning with calories. In R. C. Bolles (Ed.), The hedonics of taste (pp. 107-126). Hillsdale, N J: Erlbaum.

MIKULKA, R J., LEARD, B., & KLEIN, S. B. (1977). Illness-alone exposure as a source of interference with the acquisition and retention of a taste aversion. Journal of Experimental Psychology: Animal Behavior Processes, 3, 189-201.

MILLER, R. R., & MATZEL, L. D. (1988). The comparator hypothesis: A response rule for the expression of associations. In G. H. Bower (Ed.), The psychology of learning and motivation (Vol. 22, pp. 51-92). San Diego: Academic Press.

OVERMIER, J. B., PAYNE, R. J., BRACKBILL, R. M., LINDER, B., & LAWRY, J. A. (1979). On the mechanism of the post-asymptotic CR decrement phenomenon. Acta Neurobiologiae Experimentalis, 39, 603-620.

RANDICH, A., & HAGGARD, D. (1983). Exposure to the unconditioned stimulus alone: Effects on retention and acquisition of conditioned suppression. Journal of Experimental Psychology: Animal Behavior Processes, 9, 147-159.

RANDICH, A., & RESCORLA, R. A. (1981). The effects of separate presentations of the US on conditioned suppression. Animal Learning & Behavior, 9, 56-64.

RESCORLA, R. A. (1973). Effect of US habituation following conditioning. Journal of Comparative and Physiological Psychology, 82, 137-143.

RESCORLA, R. A. (1974). Effect of inflation of the unconditioned stimulus value following conditioning. Journal of Comparative and Physiological Psychology, 86, 101-106.

RESCORLA, R. A., & DURLACH, P. J. (1981). Within-event learning in Pavlovian conditioning. In N. E. Spear & R. R. Miller (Eds.), Information processing in animals: Memory mechanisms (pp. 81-111). Hillsdale, NJ: Erlbaum.

RESCORLA, R. A., & FREBERG, L. (1978). The extinction of within-compound flavor associations. Learning and Motivation, 9, 411-427.

SCLAFANI, A. (1991). Conditioned food preferences. Bulletin of the Psychonomic Society, 29, 256-260.

SHERMAN, J. E. (1978). US inflation with trace and simultaneous fear conditioning. Animal Learning & Behavior, 6, 463-468.

STANHOPE, K. J. (1989). Dissociation of the effect of reinforcer type and response strength on the force of a conditioned response. Animal Learning & Behavior, 17, 311-321.

STANHOPE, K. J. (1992). The representation of the reinforcer and the force of the pigeon's keypeck in first- and second-order conditioning. Quarterly Journal of Experimental Psychology, 44B, 137-158.

WEINGARTEN, H. P., & MARTIN, G. M. (1989). Mechanisms of conditioned meal initiation. Physiology and Behavior, 45, 735-740.
COPYRIGHT 1997 The Psychological Record
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1997 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:unconditioned stimulus
Author:Kawai, Nobuyuki; Nakajima, Sadahiko
Publication:The Psychological Record
Date:Jun 22, 1997
Words:7079
Previous Article:A matching law analysis of the effect of amphetamine on responding reinforced by the opportunity to run.
Next Article:Making Sense in Law: Linguistic, Psychological and Semiotic Perspectives.
Topics:


Related Articles
Conditioned suppression and the effects of pentobarbital with picrotoxin, flumazenil, and R05-3663.
Conditional control of freezing by food deprivation intensity stimuli in Pavlovian conditioning procedures.
Value transfer in discriminative conditioning with pigeons.
The effects of terminal-link stimulus arrangements on preference in concurrent chains.
Ketaset-Rompun anesthesia induces a conditioned taste aversion in rats.
The effects of sexually explicit words on the formations of stimulus equivalence classes.
The operant-respondent distinction revisited: toward an understanding of stimulus equivalence.
Stimulus equivalence and the blocking effect.
RESPONSE PRIMACY IN FEAR CONDITIONING: DISENTANGLING THE CONTRIBUTIONS OF UCS VS. UCR INTENSITY.
DIFFERENTIAL OUTCOMES EFFECT ON INSTRUMENTAL SERIAL FEATURE-AMBIGUOUS DISCRIMINATION IN RATS.

Terms of use | Copyright © 2017 Farlex, Inc. | Feedback | For webmasters