Printer Friendly

Conditional control of freezing by food deprivation intensity stimuli in Pavlovian conditioning procedures.

Recent studies (Davidson, 1987; Davidson, Flynn, & Jarrard, 1992) show that stimuli that arise from different levels of food deprivation can act as conditional signals for shock. Rats were repeatedly shocked in a training box under deprivation Level A (either 0 or 24 hr without food), and they were not shocked in that box under the alternate deprivation level, Level B. Discrimination learning was revealed by a greater freezing response (a species-specific defense reaction) upon placement in the box under the shocked deprivation level than under the nonshocked level. The results of generalization tests, in which different interoceptive stimuli were induced through intubation of a high caloric load (corresponding with a low level of food deprivation) and injection of insulin (corresponding with a high food deprivation level), ensured that the observed control of freezing in the original discrimination was based on interoceptive consequences of the food deprivation levels and not on the external food manipulations that were required to induce the different deprivation levels.

At least two different associative mechanisms might underlie this conditional control. The first is that the food deprivation intensity stimuli became directly associated with the presence or absence of shock. On this view, the shock deprivation stimuli were functioning as simple conditioned excitors and/or the nonshock deprivation stimuli were conditioned inhibitors.

A second account is that the different deprivation intensity stimuli were functioning as so-called occasion setters (Bouton, 1993). Occasion setters are discrete or contextual stimuli that modulate conditioned responding to a target stimulus without being directly associated with the unconditioned stimulus (US) employed, or with the absence thereof. Instead, one interpretation of an occasion setter's modulatory potential is that it retrieves a specific relationship between the target stimulus and the US. The suggestion put forward here is that in the food deprivation experiments mentioned above the animals used the stimuli that arose from the different deprivation levels to retrieve whether the target stimulus, in this case the training box, would be shock-reinforced or not.

The purpose of the present experiments was to further pursue the suggested parallel between control of conditioned responding by food deprivation intensity stimuli and the modulation of such responding in standard occasion-setting experiments. To that end, three different types of procedures were used that in previous occasion-setting experiments using external contexts as putative occasion setters have been shown to reliably yield contextual control: An explicit discrimination procedure (Experiment 1), a latent inhibition (LI) procedure (Experiment 3), and a simple extinction procedure (Experiment 4). Experiment 2 examined whether deprivation intensity stimuli will acquire modulatory properties after simple excitatory conditioning employing just a single US. On the basis of the results of previous experiments manipulating external contexts, conditional control was not expected under the conditions of Experiment 2. Furthermore, the results of this experiment could be used to assess the extent of a possible nonassociative, or performance effect on freezing of the different deprivation levels used in the present experiments.

Experiment 1

In Experiment 1, we first tried to replicate previous findings regarding the potency of 0-hr and 24-hr food deprivation cues to control conditioned freezing. To that end, exactly the same procedure was employed as in Experiment 1 reported by Davidson et al. (1992).

Method

Subjects and Apparatus

The subjects were 12 experimentally naive female Lewis rats obtained from the University of Nijmegen. The mean body weight of the rats was 172 g (range: 158 - 197 g). The animals were housed individually in Makrolon cages in which water was available ad lib. They were maintained on a 12/12-hr light/dark cycle and the experimental events took place in the first portion of the dark phase of this cycle.

The apparatus used for conditioning and testing consisted of four identical boxes. Each box measured 24.5 x 25 x 24 cm and had aluminium end walls, and a ceiling and side walls made of clear Plexiglas. The floor consisted of 17 3-mm stainless-steel rods, spaced 1.3 cm apart. Through these rods, a 0.9-mA, 0.5-s scrambled electric shock could be delivered that served as the US. Each box was housed in a sound-attenuating chest and illumination was provided by a 2.8-W, 24-V houselight mounted on the ceiling of the chest. A ventilation fan provided a background noise in each box. The boxes were cleaned with tap water after each session. The same cleaning procedure held for each of the other experiments.

A camera was placed in front of the sound-attenuating chests and was used to monitor the rats' activity in the four boxes simultaneously during all sessions. A flashing light (1 s on, 4 s off) was used to pace the scoring of the rats' behavior (see Procedure). This light was positioned between the four environmental chests in such a manner that it was monitored by the camera, but not visible to the rats.

A Macintosh SE computer was used for controlling the shock US, houselight, and flashing light.

Procedure

Each rat was assigned to one of two groups (n = 6), matched on the basis of body weight. The animals in this experiment, as well as in each of the other three experiments, were trained and tested in squads of 4 subjects. Each squad was counterbalanced for group.

Discrimination training. On the day prior to the first conditioning session, food was removed from the home cage for all animals. On Day 1, 24 hr after removal of the food, each rat was individually placed in the training box for 4.5 min. Rats in Group 24-S received one footshock US during this session. The shock was presented 4 min after placement. Rats in Group 0-S were also placed in the training box but they did not receive a shock. All animals were then given access to food in the home cage, several minutes after the training session. On Day 2, approximately 24 hr after food had been made available, rats were again placed in the training box for 4.5 min. Group 0-S now received a shock after 4 min and Group 24-S did not.

On experimental Days 3 and 4, rats were brought under a 0-hr and a 24-hr food deprivation level, respectively. On Days 5 and 6, the deprivation levels were, respectively, 24 hr and 0 hr. On each of these days, rats were treated according to their group membership. Thus, rats in Group 0-S were consistently shocked under the 0-hr deprivation level and not shocked under the 24-hr deprivation level. The reverse was true for the animals in Group 24-S.

Tests. On each of 14 days, experimental Days 7-20, rats were individually placed in the training box for 4 min. No events were scheduled to occur during these tests. Prior to each session, the animals had been alternately brought under a 24-hr and a 0-hr deprivation level, starting with the 24-hr level on Day 7.

Dependent measure. The videotaped sessions of this experiment, as well as those of the other experiments to be described, were analyzed as follows. An observer scored the behavior as either "freezing" or "not freezing" once every 5 s for 4 min per session. Freezing is characterized by the absence of all visible movement, except for respiration. Albino rats in fearful circumstances sometimes demonstrate a slow, horizontal, rhythmical head movement ("pendulum motion") (Kolpakov, Borodin, & Barykina, 1977). This behavior was also scored as freezing. All other behavior was scored as not freezing.

The reliability of the scorings of the primary observer of this and the following experiments was checked by a second observer. For each of the experiments described in the present paper, at least one of the two observers was either unaware of group membership or of the purpose of the experiment. The data shown in the results sections of Experiments 1-4 are based on the scorings of the primary observer.

Regarding Experiment 1, a correlation coefficient was computed between the observers' freezing scores for each of the extinction test sessions. A rat's score on a given session was the total number of observations (out of 48) scored as freezing. The mean correlation was r = .91 (SEM: .03).

Statistics. The freezing data obtained during the last 32 observations of each session (i.e., the last two thirds of the sessions: 32 observations) of Experiments 1, 2, and 4, and the middle 16 observations of the sessions in Experiment 3 were analyzed nonparametrically. Significance of within-group differences, which were of primary importance because of their substantial sensitivity, were tested using a Wilcoxon matched-pairs signed-ranks test. Between-group differences were evaluated using Mann-Whitney tests. Results were considered significant when p [less than] .05.

Results and Discussion

The left panel of Figure 1 shows the results of the test sessions. The figure shows that during the last six test sessions, the rats in Group 0-S consistently froze more under their shocked deprivation level (0 hr) than under their nonshocked level (24 hr). The rats in Group 24-S, which had been shocked under 24 hr of food deprivation, did not consistently freeze differentially under the 24-hr deprivation level versus the 0-hr deprivation level. Pooled over the last three test sessions under each deprivation level (69.1% and 44.8% freezing under, respectively, the 0-hr and 24-hr deprivation level in Group O-S; in Group 24-S the corresponding percentages were 74.7 and 79.2), the rats in Group 0-S froze more under their shocked deprivation level than under their non-shock level, z = 2.75, whereas the animals in Group 24-S did not freeze differentially under the two levels, z = .82. A between-group comparison further revealed that, using the pooled data of the last three 24-hr deprivation sessions, the rats in Group 24-S froze more than those in Group O-S, z = 3.66, whereas a similar analysis on the 0-hr deprivation data revealed no significant difference between the groups, z = .48.

The results regarding Group 0-S replicate earlier findings (Davidson et al., 1992). In this group, freezing appeared to be under the control of the different food deprivation levels, with the deprivation cues that in the training phase had been present on shock sessions evoking more freezing than the cues that had been present during nonshock sessions. However, unlike earlier reports we failed to observe control by the deprivation levels in Group 24-S. The reason for this inconsistency is not clear, but it must be noted that in the previous experiments too, rats shocked under 0 hr of food deprivation seemed to discriminate better than rats shocked under 24 hr of food deprivation. Cues that arise from 24 hr of food deprivation in the present procedure appear to be especially potent to attenuate freezing in rats that are consistently shocked under a 0-hr deprivation level and not shocked under a 24-hr level. Instead, 0-hr deprivation cues do not seem to acquire controlling potentials so readily, regardless of whether the subjects are shocked or not shocked in the presence of these cues.

Experiment 2

Before examining possible modulatory properties of food deprivation intensity stimuli in occasion-setting procedures using only a single conditioning session (Experiments 3 and 4), it is necessary to examine the pattern of freezing after a single conditioning session under one of two deprivation levels, followed by tests under both levels. Such an assessment, which was performed in Experiment 2, is important for two reasons. First, prior studies using external contexts repeatedly have shown that, at least in aversive conditioning procedures, simple conditioning does not yield contextual control, that is, it is not context specific (see, e.g., Bouton, 1993). If food deprivation intensity stimuli are comparable to external contexts, the same should apply in the present experiment. Second, it is possible to establish potential unconditioned or nonassociative effects of the different food deprivation levels on conditioned freezing (performance effect) under the present conditions. This is of crucial importance with respect to a correct interpretation of the pattern of freezing observed in each of the other experiments.

Method

Subjects and Apparatus

Sixteen experimentally naive male Wistar rats with a mean body weight of 452 g (range: 405 - 519 g) served as the subjects. The US was a 1.0-mA, 0.5-s footshock. The apparatus used was the same as described in Experiment 1.

Procedure

Two groups of 8 subjects each were formed. The groups were matched on body weight.

Conditioning. All rats were placed in the conditioning box for 4.5 min. The shock US was presented 4 min after placement. The rats in Group 24-S had been deprived of food for 24 hr prior to this session. The rats in Group 0-S were under a 0-hr food deprivation level. Several minutes after the conditioning session, all animals were given access to food for 24 hr. On the following day, food was removed from the home cage for one half of the animals of each group.

Tests 1 and 2. On the next day, a two-session test phase was initiated. Each test consisted of placing the individual animals in the conditioning box for 4 min. The interval between the tests was 24 hr. One of the tests was conducted under a 24-hr food deprivation level, and the other under a 0-hr deprivation level. Test order was counterbalanced in each group.

Based on the data of the second test session (768 observations), the correlation between the observers' scorings was r = .93.

Results and Discussion

Whether an animal had been shocked under a 24-hr or a 0-hr food deprivation level did not have an effect on the pattern of freezing observed during the test sessions. In both groups, animals only froze more under the 0-hr level than under the 24-hr level if the first test was performed under the 0-hr level. If the first test was conducted under the 24-hr level, there was no difference in the level of freezing between tests. Pooled over test order, the rats in Group 0-S froze on a mean of 75.0% of the observations during the test under the 0-hr deprivation level and on a mean of 64.8% of the observations made during the test under the 24-hr level. There was no significant difference between these two percentages, z = 1.01. For the rats in Group 24-S, the corresponding mean percentages were, respectively, 72.7 and 60.5, and, again, the difference was not reliable, z = .84. There also was no significant difference between groups on either test, z = .11, for each comparison.

Pooled over groups, the right side of Figure 1 shows the mean freeing data obtained on the test under the 0-hr and the 24-hr level, separated on the basis of test order. The animals tested first under the 0-hr level froze more under that level than under the 24-hr deprivation level, z = 2.38, whereas those tested first under the 24-hr level did not freeze differentially under the two food deprivation levels, z = 1.35. Furthermore, animals tested first under the 0-hr level froze less under the 24-hr level than the animals that were first tested under 24 hr of food deprivation, z = 2.58.

The absence of significant between- and within-group differences indicates that conditioned responding completely generalized from the interoceptive "context" present during conditioning to another internal context. Thus, the deprivation cues did not appear to have an occasion-setting function under the present conditions. This corresponds with the absence of such a function in most simple aversive conditioning experiments manipulating external contexts. A retrieval function seems to be acquired especially in procedures in which a target stimulus is systematically reinforced in one context and nonreinforced in another (Bouton, 1993).

The effect of test order on the pattern of freezing can be accounted for by referring to two different processes. The first is simple extinction of conditioned freezing that occurs as a consequence of repeated nonreinforced exposure to a conditioning context. The second is an unconditioned, or unlearned freezing-attenuating, or general activity-enhancing effect of cues arising from 24 hr of food deprivation. Concerning the latter effect, hunger cues previously have been shown to increase spontaneous activity (Bolles, 1975; Mohanty & Mishra, 1984). The combined operation of these two processes may yield the pattern of freezing depicted in Figure 2. For the animals tested first under the 0-hr level, both processes worked for reduced freezing during the second (24-hr deprivation) test, relative to the first (0-hr) test, thereby yielding a significant difference. However, for the animals tested first under the 24-hr level, the two processes worked against finding a significant difference between tests: Extinction fostered reduced freezing on the second test, relative to the first, but the unconditioned effect of the 24-hr deprivation cues encouraged more freezing during the second test than during the first.

An explanation that takes into account the existence of an unlearned, activity-enhancing effect of 24-hr food deprivation cues may also be adopted to explain the absence of discrimination performance in Group 24-S of Experiment 1. Accordingly, the stimulus consequences of 24 hr of food deprivation did function as positive occasion setters in this group, thereby promoting enhanced freezing, but this effect was counteracted by the unlearned, response-attenuating effect of these same 24-hr deprivation cues.

A single shock US was also employed in the subsequent two experiments and the pattern of freezing observed during tests under the different deprivation levels in these experiments should be interpreted in the light of the results of Experiment 2.

Experiment 3

Contextual control of a conditioned response has also been observed in LI experiments. LI refers to the phenomenon that nonreinforced exposure to a stimulus impairs the acquisition of a conditioned response to that stimulus if it is subsequently paired with a US. A number of studies show that LI occurs only if the animal is conditioned in the preexposure context (e.g., Hall & Honey, 1989). Furthermore, it has been shown that if preexposure and conditioning are conducted in separate contexts, a subsequent test for responding to the stimulus reveals a less vigorous response in the preexposure context than it does in the conditioning context (Wright, Skala, & Peuser, 1986).

An occasion-setting account has been proposed to explain these LI data. The idea is that the conditioning context retrieves a stimulus-shock association, whereas the preexposure context retrieves a stimulus-no-shock association. In Experiment 3, we sought to assess whether stimuli arising from different levels of food deprivation can also acquire conditional control over conditioned freezing to a contextual stimulus in a LI procedure.

Method

Subjects and Apparatus

The subjects were 16 female Lewis rats, obtained from the University of Nijmegen. The animals weighed 163-194 g at the start of the experiment. Housing, maintenance, and apparatus were as described in Experiment 2.

Procedure

The rats were matched into two groups (n = 8) on the basis of body weight.

Preexposure. The rats in Group 0-LI/24-S were brought under 0 hr of food deprivation and were exposed to the future conditioning box for 20 min. A total of 7 such preexposure sessions was given. On the basis of pilot studies, these preexposure parameters were expected to yield LI. Group 24-LI/0-S received an identical treatment, except that the subjects were under 24 hr of food deprivation during each exposure session. The preexposure sessions for all animals were separated by 48 hr in order to allow the animals in Group 24-LI/0-S to consume food for 24 hr in the home cage after each preexposure session.

Conditioning. On the day following the last context preexposure session, all animals received one footshock US, 4 min after placement in the conditioning box. The animals in Groups 24-LI/0-S were under a 0-hr food deprivation level during this session; rats in Group 0-LI/24-S had been deprived of food for 24 hr.

Tests 1 and 2. On the day after the conditioning session, all rats were placed again in the conditioning box for 4 min. No events were planned and all subjects were under a 0-hr food deprivation level. On the next day, an identical session was performed, except that the rats were deprived of food for 24 hr. Thus, there was no counterbalancing of test order regarding deprivation level. On the basis of the results of Experiment 2 one must expect that, if only the two processes proposed in the discussion of that experiment (extinction and unconditioned effect of the 24-hr deprivation level) are operative in the present experiment, there is less freezing on the second test, which was performed under a 24-hr level, than on the first test, which was performed under a 0-hr deprivation level (see right panel of Figure 1, Test Order: 0 hr-24 hr). If, however, a different pattern of freezing is found, an additional process must have been at work under the present conditions.

The Test 1 and Test 2 recordings were scored by two observers. The correlation between the observers' freezing scores was r = .99 for Test 1 and .98 for Test 2.

Results and Discussion

The left side of Figure 2 shows the freezing data of the middle portion of each of the test sessions of Experiment 3 (between 80 and 160 s after placement). Within-group differences were most prominent during this period.

Wilcoxon tests revealed that Group 24-LI/0-S reliably froze less during Test 2 than during Test 1, z = 2.02, whereas in Group 0-LI/24-S, the level of freezing did not differ significantly between test sessions, z = 1.38. There were no significant between-group differences, zs [less than] .53.

The rats in Group 24-LI/0-S froze less during a test under a 24-hr food deprivation level than during a test under a 0-hr level. A similar pattern of responding was also obtained in Group 0-S in Experiment 1, which during training, also had received nonreinforced exposures to the training box under a 24-hr deprivation level and had received reinforced exposures under a 0-hr level. This correspondence may suggest that in each of the above-mentioned groups, the 24-hr deprivation cues were functioning as negative occasion setters, retrieving a conditioning box-no-shock association.

However, such a conclusion is somewhat premature. In Experiment 2, a similar response pattern was found for all animals tested in the same order as in the present experiment, although here, the 24-hr cues did not seem to function as occasion setters. Experiment 4 will further address the question whether or not 24-hr food deprivation cues can acquire negative occasion-setting properties, again using a procedure with only a single US presentation.

The null result concerning Group 0-LI/24-S can only be accounted for by assuming the operation of a process that counteracted the pattern of freezing normally observed after a single conditioning session, namely, less freezing during Test 2 conducted under a 24-hr level than during Test 1 performed under a 0-hr level (see Experiment 2). One possible process is positive occasion-setting by the 24-hr deprivation cues in this group that enhanced freezing, thereby eliminating a significant difference.

Experiment 4

Another procedure through which external contexts have been shown to become occasion setters is extinction. If a stimulus is followed by a US in Context A (conditioning) and then nonreinforced in Context B (extinction), renewed responding to the stimulus occurs when it is presented again in the original conditioning Context A (Bouton & Bolles, 1979). These, and other, results suggest that extinction is restricted to the extinction context. One possibility is that especially this context is capable of retrieving a stimulus-no-US relationship.

The purpose of Experiment 4 was to assess whether food deprivation intensity stimuli can also gain control over freezing to contextual stimuli in an extinction procedure in which rats only receive one conditioning trial.

Method

Subjects and Apparatus

Sixteen female Wistar rats, obtained from the University of Nijmegen, were used. The animals weighed 153-212 g at the start of the experiment and were housed and maintained as described for the other experiments. The apparatus was the same as employed in the previous experiment.

Procedure

Two groups of rats (n = 8) were matched on body weight.

Conditioning. The animals received one footshock US. The rats in Group 0-S/24Ext were shocked under a 0-hr food deprivation level; those in Group 24-S/0-Ext under a 24-hr deprivation level. Several minutes after termination of the conditioning session, all rats were given access to food for 24 hr. On the next day, food was removed from the home cage of each rat in Group 0-S/24-Ext.

Extinction. The extinction phase began on the following day. Two extinction sessions were performed during each of which all rats were merely individually placed in the conditioning box for 4 min. The animals in Group 0-S/24-Ext had been deprived of food for 24 hr prior to the first extinction session; the rats in Group 24-S/0-Ext were under the 0-hr deprivation level. After the first extinction session, the rats in Group 0-S/24-Ext were given access to food for 24 hr. Thereafter, all food was again removed for 24 hr to induce the appropriate deprivation level during the second extinction session.

Tests 1 and 2. Prior to the first test day, all animals had had food available for at least 24 hr, followed by a 24-hr period without food. Thus, all animals were under a 24-hr deprivation level. Test 1 was identical to each of the extinction sessions. Following this test, all animals were given access to food for 24 hr. A second test session, Test 2, was subsequently conducted. Except for the fact that the animals were under a 0-hr deprivation level, this test was identical to the first. Again, as in Experiment 3, tests were not counterbalanced with respect to deprivation level. The results of Experiment 2 suggest that the current test order is not likely to result in a major difference between test sessions if only extinction and unconditioned effects of the 24-hr deprivation cues determine freezing.

The Test 1 and 2 recordings were scored by two observers. The correlation coefficient computed between the observers' freezing scores was r = .98 for each test session.

Results and Discussion

The right panel of Figure 2 depicts the mean number of observations that were scored as freezing during the test sessions. Wilcoxon tests revealed that the rats in Group 0-S/24-Ext significantly froze less during Test 1 than during Test 2, z = 2.24, whereas the animals in Group 24-S/0-Ext did not freeze differentially on the two tests, z = .84. Furthermore, between-group comparisons revealed that during Test 1, the subjects in Group 0-S/24-Ext froze less than those in Group 24-S/0-Ext, z = 2.01. Groups did not differ in the level of freezing during Test 2, z = .63.

In Group 0-S/24-Ext, freezing under the 24-hr deprivation level was attenuated, relative to its own freezing level under the 0-hr deprivation condition, and relative to freezing in the 24-hr condition of Group 24-S/0-Ext. These findings suggest that in Group 0-S/24-Ext a process was at work that prevented finding the pattern of freezing observed during the tests of Experiment 2 (right side of Figure 1, right bars). It must be noted that Group 0-S from Experiment 1, and Group 24-LI/0-S from Experiment 3 can be considered to be closely related to Group 0-S/24-Ext from the present experiment. All of these groups received one or more shocks under 0 hr of food deprivation and nonreinforced exposure to the conditioning box under the 24-hr condition. Only the order of the reinforced and nonreinforced sessions was different. The similar effect of deprivation level on freezing suggests that in each of these groups, cues corresponding with 24 hr of food deprivation had acquired conditional control over freezing. They reduced freezing, relative to the level of freezing that appeared under the 0-hr deprivation level.

Also similar to what has been found in Experiments 1 and 3 regarding Group 24-S and Group 0-LI/24-S, respectively, Group 24-S/0-Ext did not display differential freezing under the different deprivation levels. As outlined above, this null result can be explained if it is accepted that a positive occasion-setting function of the 24-hr deprivation cues in this group was counteracted by nonassociative effects of these cues.

General Discussion

The results of the present experiments suggest that interoceptive stimuli that arise from 24 hr of food deprivation can attenuate freezing that was conditioned to external contextual cues under a 0-hr food deprivation level. The results further suggest that this attenuating potential is at least partly based upon an associative process which is activated in an explicit discrimination, latent inhibition, and extinction procedure (Experiments 1, 3, and 4, respectively). That 24-hr food deprivation cues can also enhance conditioned freezing through an associative process could not directly be observed in the present experiments (the groups shocked under the 24-hr deprivation condition did not show differential freezing) but was inferred from the results of Experiment 2. These suggested that 24-hr deprivation cues have an unconditioned or nonassociative effect on freezing in that they cause a reduction in the level of freezing, relative to a 0-hr food deprivation condition. As described in the separate discussions of Experiments 3 and 4, the joint operation of these associative and nonassociative processes may have caused the pattern of freezing observed in the different groups of Experiments 1, 3, and 4.

There are two possibilities regarding the nature of the proposed associative process. First, the 24-hr deprivation cues may have acquired a direct inhibitory association with shock, signaling the absence thereof. Second, these cues may have acted as retrieval cues for a training context-no-shock association (negative occasion setting).

Although the present experiments do not directly address this issue, some arguments can be given in favor of the second alternative. First, there appears to be a strong parallel between the outcome of the present experiments and studies on contextual occasion setting using external contexts. In both types of studies, contextual modulation appears when a common target stimulus is reinforced in one context and not reinforced in another. In this respect it must also be noted that occasion setting did not emerge under the conditions of Experiment 2, in which only a single reinforced session was performed, followed by two tests.(1) The outcome of this experiment also corresponds to findings from experiments manipulating external contexts after consistent reinforcement of a target stimulus.

Second, it is reasonable to assume that the deprivation intensity stimuli under the 24-hr condition (the "internal context") emerged gradually, as the time since the last meal increased. Consequently, to a large extent, the putative occasion-setting context was already present prior to placement of the rat in the "target" context. Furthermore, the internal context was also still present after the subject was removed from the training box ("termination" of the target stimulus). In other words: The target was embedded in the putative occasion-setting context. This is a condition that is conducive to the acquisition of occasion setting instead of direct, "simple" associations (e.g., Holland, 1986).

In the present experiments, the 0-hr food deprivation stimuli did not seem to have acquired a controlling property. A significant difference between groups was absent in each of Experiments 1, 3, and 4, when testing the animals under a 0-hr deprivation level. This failure to acquire a signaling potential may be caused by the 0-hr deprivation stimuli simply being not salient. The "internal state" corresponding with the 0-hr food deprivation regimen has been the "normal" state throughout the rats' lives (see also Rescorla, 1991, for suggestions that the salience of modulators influences the extent of modulatory control).

Collectively, the present experiments partly replicate earlier findings of a conditional-stimulus function acquired by food deprivation intensity stimuli. Furthermore, they provide an important extension because they also demonstrate that such a function occurs in latent inhibition and extinction procedures in which only a single US is presented. The latter suggests that deprivation intensity stimuli are even more potent conditional stimuli than hitherto assumed.

1 Occasion setting could, in principle, have emerged here during Test 2. Test 1 can be viewed as an extinction session identical to one of the extinction sessions performed in Experiment 4. Consequently, for the rats in, for instance, Group 0-S, Test I under a 24-hr deprivation level could be conceptualized as an "extinction" session. Theoretically, Test 2, performed under a 0-hr level, could then reveal "renewal" of conditioned responding because the test deprivation level corresponded with the deprivation level present during conditioning and not with the level present during "extinction." However, such an effect did not emerge in Experiment 2. Relevant between-group comparisons regarding Test 2 performance clearly failed to reveal significant differences. In other words: At least in Experiments 2, 3, and 4, the test manipulations were not sufficient to induce occasion setting on their own. However, even if the test manipulations should have contributed to occasion setting in Experiments 3 and 4, this would still support the main message of the present paper, namely, that an occasion-setting function is acquired very rapidly by food deprivation intensity stimuli.

References

BOLLES, R. C. (1975). Theory of motivation. New York: Harper & Row.

BOUTON, M. E. (1993). Context, time, and memory retrieval in the interference paradigms of Pavlovian learning. Psychological Bulletin, 114, 80-99.

BOUTON, M. E., & BOLLES, R. C. (1979). Contextual control of the extinction of conditioned fear. Learning and Motivation, 10, 445-466.

DAVIDSON, T. L. (1987). Learning about deprivation intensity stimuli. Behavioral Neuroscience, 101, 198-208.

DAVIDSON, T. L., FLYNN, F. W., & JARRARD, L. E. (1992). Potency of food deprivation intensity cues as discriminative stimuli. Journal of Experimental Psychology: Animal Behavior Processes, 18, 174-181.

HALL, G., & HONEY, R. C. (1989). Contextual effects in conditioning, latent inhibition, and habituation: Associative and retrieval functions of contextual cues. Journal of Experimental Psychology: Animal Behavior Processes, 15, 232-241.

HOLLAND, P. C. (1986). Temporal determinants of occasion setting in feature positive discriminations. Animal Learning and Behavior, 14, 111-120.

KOLPAKOV, V. G., BORODIN, P. M., & BARYKINA, N. N. (1977). Catatonic behaviour in the Norway rat. Behaviour, 62, 190-207.

MOHANTY, B., & MISHRA, P. K. (1984). Hunger drive and spontaneous activity of albino rats. Psychological Research Journal, 8, 18-25.

RESCORLA, R. A. (1991). Separate reinforcement can enhance the effect of modulators. Journal of Experimental Psychology: Animal Behavior Processes, 17, 259-269.

WRIGHT, D. C., SKALA, K. D., & PEUSER, K. A. (1986). Latent inhibition from context-dependent retrieval of conflicting information. Bulletin of the Psychonomic Society, 24, 152-154.
COPYRIGHT 1996 The Psychological Record
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1996 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Maes, J.H.R.; Vossen, J.M.H.
Publication:The Psychological Record
Date:Jun 22, 1996
Words:5685
Previous Article:Self-Control: Waiting Until Tomorrow for What You What Today.
Next Article:The effect of a change in body weight on running and responding reinforced by the opportunity to run.
Topics:


Related Articles
Emergent relations in the formation of stimulus classes by pigeons.
Emergence of conditional discriminations after constructed-response matching-to-sample training.
A traditional account of stimulus equivalence.
Unreinforced conditional selection within three-choice conditional discriminations.
US postexposure effect on conditioned flavor preference in the rat.
The operant-respondent distinction revisited: toward an understanding of stimulus equivalence.
RESPONSE PRIMACY IN FEAR CONDITIONING: DISENTANGLING THE CONTRIBUTIONS OF UCS VS. UCR INTENSITY.
Studying stimulus equivalence: defense of the two-choice procedure.
Derived conditional position discrimination in individuals without naming skills.
Control by contextual stimuli in novel second-order conditional discriminations.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters