Value transfer in a simultaneous discrimination appears to result from within-event Pavlovian conditioning.

When pigeons acquire a simple simultaneous discrimination, some of the value acquired by the S+ transfers to the S–. The mechanism underlying this transfer of value was examined in three experiments. In Experiment 1, pigeons trained on two simultaneous discriminations (A?+?B– and C?+?D–) showed a preference for B over D. This preference was reduced, however, following the devaluation of A. In Experiment 2, when after the same original training, value was given to D, the pigeons' preference for C did not significantly increase. In Experiment 3, when both discriminations involved partial reinforcement (S±), A?+?C– training resulted in a preference for B over D, whereas B?+?D– training resulted in a preference for A over C. Thus, simultaneous discrimination training appears to result in bidirectional within-event conditioning involving the S+ and S–. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Temporal context and conditioned reinforcement value.

The effectiveness of a stimulus as a conditioned reinforcer depends on the temporal context of reinforcement, that is, the overall rate of reinforcement in the situation. The dominant view has been that context determines the learned value of a stimulus directly, according to delay-reduction theory. By contrast, the contextual choice model (CCM) maintains that value is independent of context and incorporates the effects of context on choice in the framework of the matching law. The authors report 2 experiments with pigeons as subjects that use transfer tests to assess the value of stimuli in the concurrent-chains procedure. Results strongly support the assumption of CCM that pigeons learn the temporal relations between events independently of context but that context modulates the expression of that learning as choice. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Can pigeons really make transitive inferences?

L. von Fersen et al (see record 1991-29523-001) claimed that pigeons are able to make transitive inferences that are logically comparable to the kinds of transitive inferences that have been examined in children. Without putting their methodology or their results into question, analysis of the reinforcement conditions they used indicates that what was actually examined was not transitive inference and that this study does not provide evidence that pigeons are in fact capable of transitive inference. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Observing response in goldfish (Carassius auratus).

In three experiments we established that goldfish make observing responses. In Experiment 1 fish were rewarded according to a mixed schedule of reinforcement with variable-interval (VI) and extinction (EXT) segments alternating randomly, each in the presence of the same exteroceptive stimulus. By striking a second target the fish in one group could produce a stimulus associated with food (S+). Fish in a second group could produce a stimulus associated with extinction (S–). Although few observing responses were made, fish apparently found aversive the S–. In Experiment 2 fish could respond on a multiple schedule of reinforcement with VI and EXT segments alternating randomly, each in the presence of a different exteroceptive stimulus or on a mixed schedule. Fish preferred the multiple side during VI segments but not during EXT segments until the predictive value of S+ and S– was reduced. In Experiment 3 fish were required to respond to produce S+ and S–. Again, fish preferred the multiple side during VI segments only. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Pedunculopontine tegmental nucleus lesions impair stimulus–reward learning in autoshaping and conditioned reinforcement paradigms.

The role of the pedunculopontine tegmental nucleus (PPTg) in stimulus–reward learning was assessed by testing the effects of PPTg lesions on performance in visual autoshaping and conditioned reinforcement (CRf) paradigms. Rats with PPTg lesions were unable to learn an association between a conditioned stimulus (CS) and a primary reward in either paradigm. In the autoshaping experiment, PPTg-lesioned rats approached the CS+ and CS– with equal frequency, and the latencies to respond to the two stimuli did not differ. PPTg lesions also disrupted discriminated approaches to an appetitive CS in the CRf paradigm and completely abolished the acquisition of responding with CRf. These data are discussed in the context of possible cognitive function of the PPTg, particularly in terms of lesion-induced disruptions of attentional processes that are mediated by the thalamus. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Pigeons' inferences are transitive and the outcome of elementary conditioning principles: A response.

Contrary to H. Markovits and C. Dumas (see record 1992-37986-001), this article maintains that, although semantically questionable, the transitive-inference performance in pigeons demonstrated by L. von Fersen et al (see record 1991-29523-001) was impeccably transitive. Von Fersen et al proposed a local rule to account for performance. P. A. Couvillon and M. E. Bitterman (see record 1992-37970-001) provided a rationale for such a rule by pointing out that the equally reinforced central stimuli, B, C, and D, are unequally unreinforced. This article shows that many models that recognize an effect of nonreinforcement on stimulus value give similar results. Therefore, Couvillon and Bitterman's argument that nothing beyond conventional conditioning principles is necessary to account for the transitive-inference effect in pigeons is supported. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Spatial peak shift and generalization in pigeons.

How pigeons generalize across spatial locations was examined in the 4 experiments reported in this article. During training, a square was presented at a fixed height at 1 of 2 horizontal locations on a monitor screen. One location (S+ ) signaled reward, whereas the other one (S–) signaled no reward. The birds were then tested occasionally with a range of locations. After training with S+ only, the generalization gradient peaked at S+ and was approximately Gaussian in shape. After training with equal numbers of S+ and S– trials, response rates were higher on the S+ side of the distribution. This asymmetry diminished over testing. When the S+ and S– were close together, the peak of responding was shifted on initial generalization tests. Generalization gradients along the orthogonal vertical dimension were approximately exponential in shape. This is the first demonstration of generalization and peak shift in the spatial domain. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Amygdaloid complex lesions differentially affect retention of tasks using appetitive and aversive reinforcement.

Hypothesized that the amygdaloid complex (AC) is involved in the formation of stimulus–reward associations. A series of experiments (1A–2C) directly compared the effects of lesions (produced by injection of the excitotoxin N-methyl-{D}-aspartic acid) on 1-trial appetitive (APP) and 1-trial aversive (AV) learning in rats. Exps 1A and 1B, which used different degrees of reinforcement, showed no effect of lesions on the APP task, whereas acquisition of the AV task was significantly impaired. This impairment depended on the nature of the AV reinforcement used: Impairment was seen when a highly AV stimulus (footshock) was used but not when a less AV stimulus (0.2% quinine solution) was used. Control experiments showed that the effect of lesions was not due to reduced sensitivity to the footshock. In Exp 2, a novel odor conditioning task examined further the effect of AC lesions on acquisition of APP and AV stimulus–reinforcement associations. As in Exp 1, AC lesions impaired learning of the AV association but did not significantly influence the APP association. Although the AC may be involved in some APP rewarded learning, the differential effect of AC lesions on AV rewarded learning suggests a role in learning beyond the formation of stimulus–reinforcement associations. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Effects of contingency violations on the extinction of a conditioned fear inhibitor and a conditioned fear excitor.

In 4 experiments, 192 male Holtzman and Sprague-Dawley rats were used in a conditioned-suppression paradigm to assess the effects of contingency variations on responding to a conditioned stimulus (CS) inhibitor (CS–) and a conditioned stimulus excitor (CS+). In Exp I, various unconditioned stimulus/stimuli (UCS) frequencies were equated across the presence and absence of a CS– in the context of either background cues (continuous-trial procedure) or an explicit neutral event (discrete-trial procedure). With both procedures, a CS-alone treatment enhanced inhibition, whereas treatments involving 50 or 100% reinforcement for the CS– eliminated inhibition without conditioning excitation to that CS. The latter outcome also occurred in Exp II, with discrete-trial training equating considerably reduced UCS frequencies for the presence and absence of the CS–. In further evidence that inhibition was eliminated without conditioning excitation to the CS–, Exp III showed that a novel CS did not acquire excitation when 25, 50, or 100% reinforcement was equated across the presence and absence of that CS in the context of a discrete-trial event. Using the procedures of Exp I, Exp IV showed that a CS+ was extinguished by a CS-alone treatment but was substantially maintained by treatments involving 50 or 100% uncorrelated reinforcement. These effects for a CS+ and a CS– implicate CS–UCS contiguity, rather than contingency, as the factor determining the extinction of a CS. (33 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Picture fragment completion: Priming in the pigeon.

It has been suggested that the system behind implicit memory in humans is evolutionarily old and that animals should readily show priming. In Experiment 1, a picture fragment completion test was used to test priming in pigeons. After pecking a warning stimulus, pigeons were shown 2 partially obscured pictures from different categories and were always reinforced for choosing a picture from one of the categories. On control trials, the warning stimulus was a picture of some object (not from the S+ or S– category), on study trials the warning stimulus was a picture to be categorized on the next trial, and on test trials the warning stimulus was a randomly chosen picture and the S+ picture was the warning stimulus seen on the previous trial. Categorization was better on study and test trials than on control trials. Experiment 2 ruled out the possibility that the priming effect was caused by the pigeons' responding to familiarity by using warning stimuli from both S+ and S– categories. Experiment 3 investigated the time course of the priming effect. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

On the acquisition of dimensional stimulus control by the pigeon.

Trained 80 domestic pigeons to peck a white illuminated key for variable interval (VI) reinforcement. Ss were tested for angularity generalization following 0, 5, 10, or 20 min. of VI-reinforced exposure to a given line angle. For 40 Ss, the line angle was 90–; for the others, 30–. The 0-min Ss showed a systematic preference for 90–; however, 5 min. of training (with either SD value) produced nondifferential responding during generalization testing. Both 10 and 20 min. of training produced reliable gradients peaking at the SD values, with the 20-min gradient sharper but not reliably so. Furthermore, the 2 20-min gradients, peaking at 90– and 30–, were equal in slope. The various groups did not differ reliably in absolute level of responding during generalization testing, suggesting that response strength was fully established prior to the introduction of line training. Under this condition, dimensional stimulus control was acquired very rapidly and initial stimulus preferences were as rapidly overcome. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Stimulus control in instrumental discrimination learning and reinforcement schedule situations.

Investigated ways in which control is exercised over responding by cues associated with reinforcement (RC) as well as nonreinforcement (NC) in discrimination tasks. In Phase 1 of Exp I, with 20 male albino Holtzman rats, either RC or NC was reinforced. In Phase 2, RC and RN were employed as the brighter stimulus (S+) and the less bright stimulus (S–) cues. Results show that differential responding developed less rapidly when the cue reinforced in Phase 1 was the S– rather than the S+ cue in Phase 2, regardless of whether that cue was RC or NC. In Exp II, using 16 male albino Holtzman rats, the S+ cue was a compound of a hedonic cue and a brightness cue, as was the S– cue. Differential responding developed less rapidly when the hedonic cue reinforced in S– was the same as that reinforced in S+, again regardless of whether the cue was RC or NC. Results show that both cues regulated responding and that neither was frustrating to Ss. (20 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Memory mechanisms in pigeons: Evidence of base-rate neglect.

In delayed matching to sample, once acquired, pigeons presumably choose comparisons according to their memory for (the strength of) the sample. When memory for the sample is sufficiently weak, comparison choice should depend on the history of reinforcement associated with each of the comparison stimuli. In the present research, pigeons acquired two matching tasks in which Sample S1 was associated with one comparison from each task, C1 and C3, whereas Sample S2 was associated with Comparison C2, and Sample S3 was associated with Comparison C4. As the retention interval increased, the pigeons showed a bias to choose the comparison (C1 or C3) associated with the more frequently occurring sample (S1). Thus, pigeons were sensitive also to the (irrelevant) likelihood that each of the samples was presented. The results suggest that pigeons may allow their reference memory for the overall sample frequency to influence comparison choice, independent of the comparison stimuli present. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Reactivation of a visual discrimination in early infancy.

16 78–92 day old infants learned to produce movement in 2 successively presented overhead crib mobiles by footkicking in a conjugate reinforcement paradigm. The mobiles differed in both color and pattern displays on the suspended components. Following acquisition, a discrimination was introduced whereby responding was reinforced in the presence of one mobile (the positive stimulus; S+) but not in the presence of the other (the negative stimulus; S–). During a cued-recall retention test administered 21 days after the completion of discrimination training, Ss evidenced no retention of either initial or discrimination training. Nevertheless, a brief reactivation treatment with S+ 24 hrs prior to the long-term retention test alleviated forgetting and restored the S+/S– discrimination. Reactivation with S– was no more effective than no reactivation treatment at all. Results confirm the efficacy of reminder procedures with young infants and demonstrate the specificity with which the reminder alleviates forgetting. (39 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Context as a conditional cue in operant discrimination reversal learning.

Five experiments with 68 pigeons investigated the conditions under which contextual stimuli gain conditional control in the discrimination reversal paradigm. In Exp I, Ss learned an operant discrimination in which the positive stimulus (S+) was 555 nm and the negative stimulus (S–) was 576 nm in one context (houselight off plus white noise [HLFN]) and then learned the reversal (S+ 576 nm; S– 555 nm) in another context (houselight on plus tone [HLNT]). Subsequent wavelength generalization testing revealed responding appropriate to each context: The gradients peaked at 555 nm in HLFN and at 576 nm in HLNT. In Exp II, separate groups experienced both visual and auditory context cues, only visual ones, or only auditory ones. The visual cues worked as well as the compound, but the auditory cues gained no conditional control. In Exp III, houselight illumination replaced by white light directly added to the colors serving as discriminative cues. Results suggest that houselight illumination does not gain conditional control by altering the brightness and saturation of the key colors. In Exp IV, HLNT and HLFN gained conditional control over discriminations based on different angles of a white line, but background key color did not. In Exp V, conditional control over a color discrimination was established by contexts consisting of black and white striped walls vs plain walls. Findings suggest that pigeons use diffuse visual cues to identify the place where food-reinforced learning has taken place. (16 ref) (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

Temporal inferences by young children.

48 4–5 yr olds were taught to perform temporal operations with pairs of toys chosen from a set and tested with novel pairs of toys from the same set. For example, they might be taught to place Toy A through a trap door followed by Toy B, and Toy B followed by Toy C. Then they might be given the novel pair of Toys A and C to determine whether or not they make the transitive inference of placing A in the box before C. Results of 6 experiments show that Ss with little initial training made transitive temporal inferences of greater complexity than the transitive inference described above. Inferential strategies other than temporal transitivity, which were used by preoperational Ss when logical transitive inferences were impossible, were identified. These are discussed as the use of internal vs external serial boundaries. It is demonstrated how the methodology used in this study can be transformed into a test of children's linguistic competence. (25 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Associative processes in differential outcome discriminations.

This study assessed the role of stimulus–outcome (S–O) and response–outcome (R–O) associations in differential outcome performances. In Experiment 1, pigeons learned one-to-many matching-to-sample with differential R–O associations but with either differential or nondifferential sample-outcome (S–O) associations. Later, both groups showed strong transfer of matching to novel samples that were differentially associated with the same outcomes used in training. Experiment 2 reversed the training conditions for each group. The switch from differential to nondifferential S–O associations produced a large drop in matching accuracy, whereas the opposite switch had only a small effect. Following recovery, both groups again showed positive transfer to novel samples, although transfer was stronger following differential S–O training. These data support a 2-process ("outcome expectancy") account of differential outcome performances but also indicate a contribution of bidirectional R–O associations. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Representation of serial order in pigeons (Columba livia).

In Experiment 1, 2 groups of pigeons were trained to respond to either a 4-item (A→B→C→D) or 5-item (A→B→C→D→E) list. After learning their respective list, half of the subjects were trained on a positive pair with reinforcement provided when pairs were responded to in the order true to that of the original sequence (4-item: B→C; 5-item: B→D). The remaining subjects were trained on a negative pair with reinforcement provided for responding to the pairs in the order opposite to that learned in the original sequence (4-item: C→B; 5-item: D→B). Subjects in the positive pair condition learned their respective pair faster than did subjects in the negative pair condition. In Experiment 2, after reaching criterion on a 4-item list, subjects received 16 BC probe trials spread across 4 sessions of training. Subjects performed significantly above chance on the probe trials. The performance of our subjects in Experiments 1 and 2 demonstrates that, similar to monkeys, pigeons form a representation of the lists that they learn. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The development of transitive inference: A review of recent approaches.

Reviews research on the development of transitive inference in children. Two principal approaches, one associated with Piaget's stage theory, the other associated with an information-processing theory, are compared. The 2 approaches differ with respect to 4 aspects of method: choice of task, response required, initial training, and method of feedback. The divergent conclusions of the approaches regarding the age at which transitive inference emerges are discussed in relation to these points of methodological divergence. Several issues in the area of transitive inference are viewed as prototypical of diagnostic problems encountered in cognitive development generally. (33 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

The signal-generated partial reinforcement extinction effect.

The five experiments reported here distinguish between memory of the reinforcement event just experienced on a preceding trial, called event-generated memory, and a reinforcement event memory activated by a signal associated with the memory, called signal-generated memory. In each experiment, a group of rats received discrimination training in Phase 1 to establish some stimulus as a signal for nonreward (N) and then consistent reinforcement (CRF) in Phase 2, in which the signal for N was presented on selected rewarded (R) trials. I hypothesized that this treatment would activate a memory of nonreward, S{N} and that this signal-generated S{N} would become a signal for reward much like event-generated S{N} does during conventional partial reinforcement (PRF) training. Rats that received this training had substantially higher resistance to extinction than did CRF and nonsignaled control groups, an effect named the signal-generated partial reinforcement extinction effect (PREE). The effect was shown to be due to presenting a signal associated with N in the CRF context and not to intrinsic properties of the signal or other characteristics of the training experience; the effect was also shown to be as large as the conventional PREE. (PsycINFO Database Record (c) 2010 APA, all rights reserved)