Reversal Learning In Rats Research Articles

Gustatory stimulus control and reversal learning in rats

Behavioral indices of taste discrimination have been used to study gustatory physiology and psychophysics though they are often biased by taste and position preferences and offer limited stimulus control. In the present experiment five rats were trained to discriminate distilled water from 0.9% NaCl using a discretetrials go-no go procedure. When two rats were further trained on a series of 0.1% saccharin/NaCl discrimination reversals, correct responding occurred more rapidly following saccharin presentations independent of its stimulus property (S D or S Δ). The results are discussed in the context of experimental methods and the usefulness of the present technique for studies of gustatory physiology.

Effects of context manipulation on memory and reversal learning in rats with hippocampal lesions.

In Experiment 1, groups of hippocampal and control rats learned a simultaneous discrimination habit and were subsequently tested for its retention under variable contextual conditions. All groups recalled the discrimination response to an equally high level when testing conditions were constant throughout, but the hippocampal group showed impaired memory when contextual stimuli at recall testing did not conform to those of original learning. Experiment 2 indicated that the hippocampal impairment was not simply the result of introducing novel stimuli. In Experimental 3, animals were administered a reversal learning task with contextual stimuli varied between the two tests. The typically observed impairment of hippocampal animals on this task was reduced by contrasting contextual conditions. The data are seen to support a contex-tetrieval interpretation of hippocampal function.

Successive Visual Reversal Learning in Rats as a Function of Amount of Training per Reversal and Reduced Spatial Cues

An experiment is reported which examined the effects of varying amounts of training per reversal over successive brightness discrimination reversals in rats, in a situation which reduced visual cues to spatial position. The results were similar to those of studies showing that amount of training affects successive reversal improvement on spatial problems but no differences were observed as a result of varying the cues for spatial position. The latter finding failed to confirm the result of a recent experiment in which a learning criterion was employed rather than a fixed-trials procedure.

Effects of overtraining and pretrial administration of dextroamphetamine on reversal learning in rats.

Krivanek and McGaugh (1969) have demonstrated facilitative effects of pretrial administration of amphetamine compounds on brightness discrimination in mice. These investigators suggested that the drug affected time-dependent memory-storage processes but concluded that further research was necessary to determine their nature. The effect of d-amphetamine on the ability of rats to attend selectively to specific environmental cues was examined in the present study. According to a model of discrimination learning proposed by Sutherland and Mackintosh (1971), animals receiving postcriterion training acquire a reversal shift faster than nonovertrained animals because in the former case such additional training serves to increase the probability of attending to the problem's relevant dimension. If amphetamines facilitate attentive skills, then one would expect better performance in an amphetamine-injected group on tasks requiring selective attention to particular cues than in a group not given the drug. Ss were 20 naive, adult male, Long-Evans rats randomly distributed in groups of five across four conditions: Overtrained-Saline-injected, Overtrained4-amphetamine-injected, Nonovertrained-Saline-injected, and Nonovertrained4-amphetamine-injected. All Ss were given a light gray/dark gray discrimination task in a Grice box with one irrelevant dimension (stripes). Overtrained Ss received 100 postcriterion trials while nonovertrained Ss were given the reversal shift as soon as criterion (nine consecutively correct trials) was reached. Ten minutes prior to testing, half the Ss in each of these groups received Zmg/kg of .9% saline solution, the other half receiving an equal dosage of d-amphetamine. Ten noncorrected trials were administered daily with a 3- to 5-min. intertrial interval, correct responses reinforced with 5 45-mg Noyes pellets. Results of an analysis of variance on trials to criterion showed a significant difference between the overtrained and nonovertrained groups (F = 6.70, df = 1/16, 9 < .05), thereby lending further support to the generality of the overtraining reversal effect. No significant differences were obtained, however, between the learning scores of Ss injected with saline and those injected with d-amphetamine, nor were any interactions found. The heightened task complexity in the present experiment, together with a small subject pool, may provide one explanation for the lack of facilitation found by others under similar con

The effect of frustration and reduced spatial cues on visual discrimination and reversal learning in rats

Abstract Three experiments are described in which discrimination and reversal learning of a visual discrimination were examined in the presence and absence of visual cues to spatial position. In the first experiment rats learned a vertical-horizontal discrimination and reversal. The rate of learning of the initial discrimination and the reversal was not related to the presence or absence of the visual cues to position. In the second experiment rats learned a more difficult light grey-dark grey discrimination and reversal. Subjects for whom visual cues to position were reduced were slower to learn both the discrimination and the reversal. In the third experiment a period of frustration training, i.e. experience with an insoluble problem, preceded the discrimination learning. The frustration training interfered with the acquisition of the discrimination but the rate of learning was not related to the presence or absence of visual position cues.

Spatial reversal learning in rats and gerbils

Adult hooded rats and gerbils were trained to criterion on a spatial discrimination in a T-maze, followed by 15 daily 10-trial reversals. Results indicated that both species improved significantly over the 15 reversals, but there was a reliable difference between species. Specifically, gerbils made slightly more errors than did rats. Gerbils also were observed to explore in the maze more than rats. Present and prior findings suggest that gerbils perform less well than domestic rats on appetitive learning tasks. Other findings suggest that they are as good or better than rats at active avoidance learning. One possibility is that the gerbil’s greater tendency to explore and/or their higher activity level interferes with performance on appetitive tasks but enhances performance in active avoidance learning.

Infantile handling and successive spatial reversal learning in rats

The successive reversal learning methodology was used to compare the learning ability of rats which had received preweaning stimulation with those that had not been handled. The results showed that most of the male handled rats had completed seven reversals and that none from the other groups made it within 500 trials. Most of the female handled rats reached five reversals and most of the male nonhandled rats attained four reversals. The female nonhandled rats achieved only three reversals. It was suggested that the successive reversal learning methodology is likely to reveal group differences when the data are analyzed in terms of the number of reversals attained by the animals from different treatment groups within a fixed number of test trials. However, this methodology is less likely to reveal treatment effects when the data are analyzed for intratask group differences.

Spatial reversal learning in rats, pigeons, and goldfish

Four rats, four pigeons, and four goldfish were trained under comparable conditions for 40 trials per day on a spatial discrimination and a series of 30 daily reversals. All three groups of animals showed a significant reduction in errors per reversal over the course of the experiment, but the rats improved much more rapidly than the pigeons, who in turn improved more rapidly than the goldfish.

Performance on the hebb-williams maze as related to discrimination and reversal learning in rats

Performance on the Hebb-Williams Maze was studied in relation to visual form discrimination and reversal. Three groups, each of twenty hooded rats, were first tested on the Hebb-Williams Maze and then on discrimination and reversal in a T-maze with either vertical and horizontal black and white stripes or a black circle and triangle against white background as stimulus patterns. Animals in group I and II were tested for both discrimination and reversals. Hebb-Williams Maze scores were found to correlate significantly with performance in the first discrimination and reversal. The results are inconsistent with the assumption of Woods and his associates that Hebb-Williams Maze scores are merely reflective of locomotor behaviour.

Effects of intracranial reinforcement intensity and distributional variables on brightness reversal learning in rats.

Effects of intracranial reinforcement intensity and distributional variables on brightness reversal learning in rats.

Effects of rewarded and nonrewarded prereversal training on discrimination reversal learning in rats

Rats given nonrewarded experience with the former S+ after mastery of a visual discrimination learned the reversal of the discrimination with fewer errors than Ss given rewarded prereversal experience with the former S−. This finding was similar to those previously found with monkeys.

Reversal learning in rats as a function of the type of discrimination and the criterion of learning

Summary Rats were given ten T-maze reversals while criterion of learning (eight out of ten correct with last five consecutive; twelve consecutive correct; thirty consecutive correct) and type of discrimination (spatial v. non-spatial) were varied. Results showed no effects of overtraining and no significant decrease in errors over reversals in non-spatial group. Analysis of the situation was made in terms of simultaneous and successive non-spatial learning

Reversal learning in rats as a function of percentage of reinforcement and degree of learning.

An experiment was performed investigating the effect of percentage reinforcement and degree of overlearning on reversal learning in rats. 168 female albino rats, 75-80 days old, were run in a black-white discrimination task according to a 2 X 4 orthogonal design. Acquisition was under 1 of 2 percentages of reinforcement—100% and 50%. Upon reaching criterion the groups were divided into 4 subgroups receiving 0, 70, 140, 'or 280 trials overlearning. Following overlearning, reversal learning ,was begun. Acquisition level was also controlled. Results showed that following 50% reinforcement, there is an inverse relationship between the number of overlearning trials and ease of reversal learning. Following 100% reinforcement, no effect due to overlearning was found. Reversal was easier following 100% than following 50% reinforcement. It was concluded from the reversal results that at least 2 factors are required to explain the partial reinforcement effect and the overlearning reversal effect.