“Counting” serially presented stimuli by human and nonhuman primates and pigeons

Much of Stewart Hulse's career was spent analyzing how animals can extract patterned information from sequences of stimuli. Yet an additional form of information contained in a sequence may be the number of times different elements occurred. Experiments that required numerical discrimination between different stimulus items presented in sequence are analyzed for primates and pigeons. It is shown that a model based on magnitude discrimination can account well for data from human and nonhuman primate experiments. Similar experiments carried out with pigeons showed a strong recency effect not found with primates. The pigeon data are modeled successfully, however, by assuming that representations of events decay as other events are presented.     

Qualities and locations of stimuli and responses affecting discrimination learning of chinchillas (Chinchilla laniger) and pigeons (Columba livia)

Chinchillas and pigeons were used as subjects in separate experiments to study interactions among stimulus and response characteristics in discrimination learning. Both the stimuli and the responses could differ with respect to their "quality" and their "location." Bright versus dim lights and upper versus lower lights served as the stimulus qualities and stimulus locations for the chinchillas, respectively. Red versus green lights and upper versus lower lights served as the stimulus qualities and stimulus locations for the pigeons, respectively. Respond versus no-respond and respond-left versus respond-right served as the response qualities and response locations, for both species, respectively. In both experiments, response-quality performance was superior when the discriminative stimuli differed in quality than when they differed in location, whereas response-location performance was superior when the discriminative stimuli differed in location than when they differed in quality. These results were interpreted within the framework provided by a general law of learning, that is, the "quality-location hypothesis."     

The discrimination of relative frequency by pigeons.

Five experiments addressed the issue of how pigeons learn to discriminate the relative frequency of stimuli. During a sampling period, three different stimuli (keylights) were presented serially, in mixed order, and with different frequencies. During a choice period, the stimuli were presented simultaneously, and reinforcement was arranged for choosing the stimulus that was presented the least number of times during the sample. The results showed that (a) the overall proportion of correct choices was always above chance levels; (b) the likelihood of a correct choice decreased with the serial position of the correct stimulus, a negative recency effect; (c) when the last three stimuli of the sample were constrained to be one of each kind, the negative recency effect decreased but errors became more likely when the correct stimulus occurred early in the sample, a negative primacy effect; (d) accurate performance generalized to new and larger samples; and (e) under some conditions the probability of a correct choice was independent of the serial position of the correct stimulus. The serial position curves suggest that in a least frequent discrimination task, two processes determine how the least frequent stimulus controls behavior: a passive decay process (the stimulus loses its effectiveness with time since its last occurrence), and a residual salience process (when the stimulus occurs in the first position it may decay to a higher asymptote than when it occurs in later positions.     

Timing light and tone signals in pigeons

Pigeons' ability to time light and tone stimuli was examined in four experiments. In Experiment 1, two groups of pigeons were trained to discriminate between 2- and 8-s durations of lights or tones and then were transferred to reversal or nonreversal discriminations in the alternate modality. Pigeons learned the light discrimination faster than the tone discrimination and showed immediate positive intermodal transfer from tone to light but not from light to tone. In Experiments 2-4, the peak procedure was used to study birds' timing of 15- and 30-s fixed-interval light and tone signals. Peak times on empty trials under baseline conditions closely approximated the length of fixed-interval signals. When pigeons were tested with time-outs and intermodal switches introduced midway through an empty trial, they tended to reset the timing mechanism and begin timing again from 0 s. With both estimation and production procedures, pigeons were less accurate when timing the tone stimuli than when timing the light stimuli. A comparison of these data with data from timing experiments with rats suggests several possible differences in timing processes between pigeons and rats.     

The effects of number of sample stimuli and number of choices in a detection task on measures of discriminability

Six pigeons were trained on a conditional discrimination task involving the discrimination of various intensities of yellow light. The research asked whether stimulus—response discriminability measures between any pair of stimuli would remain constant when a third or fourth sample and reinforced response were added. The numbers of different sample stimuli presented and different responses reinforced were two (Part 1), three (Parts 2 and 4), and four (Part 3). Across conditions within parts, the ratios of reinforcers obtainable for correct responses were varied over at least five levels. In Part 5, the numbers of sample stimuli and reinforced responses were varied among two, three, and four, and the reinforcer ratio between consecutive remaining samples was constant at 2:1. It was found that once a particular response had been reinforced, subjects continued to emit that response when the conditional stimulus for that response was no longer presented. Data analysis using a generalization-based detection model indicated that this model was able to describe the data effectively. Four findings were in accord with the theory. First, estimates of stimulus—response discriminability usually decreased as the arranged physical disparity between the sample stimuli decreased. Second, stimulus—response discriminability measures were independent of response—reinforcer discriminability measures, preserving parameter invariance between these measures. Third, stimulus—response discriminability measures for constant pairs of conditional stimuli did not change systematically as conditional stimulus—response alternatives were added. Fourth, log stimulus—response discriminability values between physically adjacent conditional stimuli summed to values that were not significantly different from estimates of the discriminability values for conditional stimuli that were spaced further apart.     

Olfactory discrimination, reversal learning, and stimulus control in rats.

Rats were trained to discriminate lights, tones, or odors and then given a series of discrimination reversals. Only rats trained with odors showed positive transfer on the first reversal and acquisition of a reversal set. Other experiments demonstrated that rats preferentially attend to odors when presented in compound with lights or tones; that odors exert more discriminative control than tones in tests using compound stimuli of competing sign; and that after pretraining on the positive stimulus, acquisition of an odor but not a light discrimination occurs with virtually no errors. These results demonstrate the importance of stimulus modality in the establishment of stimulus control and the need for more careful analysis of stimulus factors in cross-species comparisons of learning ability.     

An operant discrimination task allowing variability of reinforced response patterning

Five pigeons were trained to perform a discrimination task allowing variability of reinforced response patterning. The task consisted of moving a stimulus light within an 4×4 matrix of lights from the top left position to the bottom right position by pecking on two keys in succession in order to obtain a reinforcement. A peck on one key moved the light one position to the right and a peck on the other key moved it one position down. After preliminary training on alternating fixed-ratio 3 schedules of reinforcement, the birds could peck on either key in any order, but more than three responses on a key resulted in a blackout followed by the return of the stimulus light to the start position. Results indicate that initially the birds used a wide variety of response patterns to obtain reinforcement, but with continued practice, response patterns became more stereotyped.     

Correspondence as conditional stimulus control: insights from experiments with pigeons.

Correspondence between saying and doing, typically studied in young children and individuals with developmental disabilities, was examined as an instance of conditional stimulus control. In Experiment 1, 3 pigeons were exposed to a two‐component repeated‐trials procedure. In the first—sample or say—component, two response keys transilluminated by different colored lights were presented and the pigeon pecked one of the keys. After 1 s of darkness in the chamber, the second—choice or do—component was presented, in which the two keys again were transilluminated, one by the color selected in the first component and the second by another color. Selecting the color that matched that selected in the say component resulted in access to food. Selecting the other color produced a blackout of the chamber. After an intertrial interval (ITI), the next say component was programmed, and the procedure was repeated. Correspondence remained at chance levels through several manipulations of ITI duration and sample response requirement. When a correction procedure was added such that only the originally selected sample stimulus was re‐presented until a correct choice response occurred, reliable correspondence developed in 2 pigeons. This correspondence was eliminated by making reinforcement independent of correspondence and subsequently was reestablished when reinforcement again depended on correspondence. In Experiment 2, 3 other pigeons rapidly acquired correspondence under the final procedure used in Experiment 1. Increasing the time interval between the say and do components diminished correspondence. The results of the two experiments suggest how correspondence may be considered an instance of conditional stimulus control and that it is possible to construct a homologue of human say‐do correspondence with pigeons.     

Complex learning and information processing by pigeons: a critical analysis

Three models of conditional discrimination learning by pigeons are described: stimulus configuration learning, the multiple-rule model, and concept learning. A review of the literature reveals that true concept learning is not characteristic of the behavior of pigeons in matching-to-sample, oddity-from-sample, or symbolic matching studies. Instead, pigeons learn a set of sample-specific S^D rules. Transfer of the discrimination to novel stimuli, at least along the hue dimension, is predicted by a “coding hypothesis”, which holds that pigeons make a unique, but usually unobserved response, R₁, to each sample, and that the comparison stimulus chosen depends on which R₁ was emitted in the presence of the sample. Convincing evidence is found that pigeons do code sample hues, but there is little evidence that allows one to infer that the “coding event” must have behavioral properties. Parameters of the conditional discrimination paradigm are identified, and it is shown that by appropriate parametric manipulation, a variety of analogous tasks may be generated for both human and animal subjects. The tasks make possible the comparative study of complex learning, attention, memory, and information processing, with the added advantage that behavior processes may be compared systematically across tasks.     

On the form of stimulus generalization curves for visual intensity

Twelve pigeons were given successive discrimination training involving variable-interval reinforcement for key pecking in the presence of one intensity of monochromatic light randomly alternated with extinction for pecking during another intensity. All of the pigeons were then tested in extinction for generalization along the intensity dimension, and all showed a displacement of maximal responding from the positive stimulus in the direction opposite the negative stimulus. For six of the pigeons, for which the test included only three values beyond the positive stimulus, four showed peaked gradients but two did not, showing monotonic gradients with maximal responding to the most extreme test value. For another six pigeons tested over a wider range, all showed peaked gradients. Thus, when a sufficiently wide range of test values is employed, generalization gradients for visual intensity have the same peaked form as do gradients for qualitative visual dimensions such as wavelength or line angle.     

Some determiners of attention

Three experiments, using a total of 13 pigeons, examined the stimulus control acquired by the separate components of a compound visual stimulus transilluminating the pecking key. Experiment I measured the control acquired by components of compound discriminative stimuli used in discrimination training. Experiment II sought to demonstrate the effect of pretraining a single stimulus discrimination on control acquired by each component in a compound stimulus discrimination. It also investigated the effect of training the compound stimulus discrimination before the single stimulus discrimination. Experiment III sought a continuous stimulus control function when pretraining stimulus intensities were varied. The results suggest that the extent to which a bird "pays attention" to a stimulus, defined in terms of the degree of stimulus control acquired by that stimulus, is determined by how well it previously learned to discriminate that stimulus from other stimuli.     

The emergence of symmetry in a conditional discrimination task using different responses as propioceptive samples in pigeons

In Experiment 1, 10 pigeons were exposed to a successive symbolic matching-to-sample procedure in which the sample was generated by the pigeons' own behavior. Each trial began with both response keys illuminated white, one being the "correct" key and the other the "incorrect" key. The pigeons had no way of discriminating which key was correct and which incorrect, since these roles were assigned on a random basis with the same probability of 0.5 for each key. A fixed ratio of five responses was required on the correct key. However, each time the pigeon pecked the incorrect key, the correct key response counter reset. Five consecutive pecks on the correct key was the only way to end this component, and switch off both key lights. Two seconds later, these same keys were illuminated again, one green and the other red (comparison stimuli). Now, if the correct white key had been on the left, a peck at one color produced food, and if the correct white key had been on the right, a peck at the other color produced food. When the pigeons had learned this discrimination, they were exposed to several symmetry tests (simultaneous presentations of both keys illuminated the same color-i.e., both red or both green), in order to interchange the sample with the comparison stimuli. In Experiment 2, the importance of requiring discrimination between the samples and between the comparisons was analyzed. In Experiment 3, we compared the results of Experiment 1 with a slightly different experiment, which resulted in discrimination of key position, an exteroceptive stimulus. The results showed that symmetry emerged only when different responses were used as samples.     

CONCEPT LEARNING AND LEARNING STRATEGIES

Abstract —Concept teaming and learning strategies of pigeons were manipulated in a matching-Co-sample task. Groups of 4 pigeons responded either 0, J, 10, or 20 times to a sample stimulus, and then chose between a matching comparison stimulus and a nonmatching comparison stimulus. Tests with unfamiliar arrangements of the three training stimuli showed that learning was not by if-then rules. Tests with novel stimuli showed that as the number of sample responses increased, learning about the configural pattern of each display gave way to more teaming about the sample-comparison relationship and more concept learning. Pigeons making the most sample responses showed complete concept teaming.     

Visual dominance in the pigeon

In Experiment 1, three pigeons were trained to obtain grain by depressing one foot treadle in the presence of a 746-Hertz tone stimulus and by depressing a second foot treadle in the presence of a red light stimulus. Intertrial stimuli included white light and the absence of tone. The latencies to respond on auditory element trials were as fast, or faster, than on visual element trials, but pigeons always responded on the visual treadle when presented with a compound stimulus composed of the auditory and visual elements. In Experiment 2, pigeons were trained on the auditory-visual discrimination task using as trial stimuli increases in the intensity of auditory or visual intertrial stimuli. Again, pigeons showed visual dominance on subsequent compound stimulus test trials. In Experiment 3, on compound test trials, the onset of the visual stimulus was delayed relative to the onset of the auditory stimulus. Visual treadle responses generally occurred with delay intervals of less than 500 milliseconds, and auditory treadle responses generally occurred with delay intervals of greater than 500 milliseconds. The results are discussed in terms of Posner, Nissen, and Klein's (1976) theory of visual dominance in humans.     

Further Investigation of Responding Elicited by BC and C After A+, AB-, ABC+ Training

Two experiments with pigeons explored conditioned keypeck responding to new visual targets after visual compound discrimination training. In the first experiment, pigeons were trained with an A+, AB-, ABC+, AD-, ADC+ task, in which stimulus A signalled food, stimulus compounds AB and AD signalled no food, and stimulus compounds ABC and ADC signalled food. According to both an elemental model (Rescorla & Wagner, 1972) and a configural model (Pearce, 1987) of Pavlovian conditioning, test compounds BC and DC should elicit less responding than should C alone. However, the birds responded more to BC and DC than to C. In the second experiment, another set of pigeons was trained with an A+, AB-, ABC+, AD-, ADE+ task, in which stimulus A signalled food, stimulus compounds AB and AD signalled no food, and stimulus compounds ABC and ADE signalled food. Contrary to the prediction of the Rescorla-Wagner model, keypeck responding was not less on BC and DE trials than on C and E trials in testing. However, B and D attenuated responding to E and C, respectively, when presented in compounds BE and DC. The Pearce model was able to accommodate these results.     

Behavioral contrast without response-rate reduction

Behavioral contrast was obtained in two experiments, which both employed a standard free-operant successive discrimination (a multiple variable-interval extinction schedule), without the occurrence of reductions of response rate in the extinction component. In Experiment I, one group of four pigeons was trained on a multiple schedule in which one stimulus was associated with a variable-interval schedule and the second stimulus with response-independent reinforcement on a free variable-interval schedule. Though by the end of this training three pigeons were responding very little to the second stimulus, when this stimulus was associated with extinction all subjects showed a contrast effect. In Experiment II, eight pigeons were trained extensively to respond to a single stimulus on a variable-interval schedule, before a second stimulus associated with extinction was introduced. This second stimulus was dissimilar to the initial stimulus and five pigeons never responded in its presence. Nevertheless, all pigeons showed a contrast effect and there was no evidence that the effect was smaller in errorless subjects or smaller than in a subsequent discrimination where all subjects made many errors. Both experiments indicated that response reduction in one component of a multiple schedule is not a necessary condition for the occurrence of contrast.     

Effect of sample presentation time on long-delay matching in the pigeon

The effect of sample stimulus presentation time on long-delay matching in highly practiced pigeons was investigated. The birds were found capable of above chance matching performance at a delay of 60 sec provided the sample stimulus was presented for 4 sec or longer. Matching accuracy increased as a negatively accelerated function of sample stimulus presentation time and decreased as a negatively accelerated function of time since the termination of the sample. The rate of forgetting was found to be independent of sample stimulus presentation time. The data were inconsistent with a temporal discrimination interpretation of the effect of presentation time on delayed matching. The data were interpreted as supporting a simple trace strength and decay model of pigeon delayed matching.     

Memory codes in pigeon short-term memory: Effects of varying the number of sample and comparison stimuli

In a delayed conditional discrimination task, pigeons can remember either some aspect of the conditional stimulus (i.e., they can code retrospectively) or some aspect of the stimulus to which they will respond at the end of the delay (i.e., they can code prospectively). To determine the nature of the memory code, we varied the number of possible sample stimuli (two or four) and the number of possible comparison stimuli (two or four) factorially across groups. Birds in all four groups were initially trained on a zero-delay, conditional discrimination with lines (vertical and horizontal) and/or shapes (circle and triangle), and were then tested with longer delays between sample offset and comparison onset. Acquisition of the conditional discrimination was affected by both the number of sample and comparison stimuli: birds were slower to reach criterion the greater the number in either stimulus set. During delay testing, however, only the number of comparisons affected performance. Overall, retention was poorer with four comparisons than with two. These data provide evidence for prospective coding in pigeon short-term memory.     

Performance of pigeons on delayed simple and conditional discriminations under equivalent training procedures

A pair of experiments investigated the short-term memory of pigeons under delayed simple and conditional discriminations. Trial sequences in both discriminations consisted of a color as the sample stimulus, a memory interval, a line orientation as the test stimulus, and a trial outcome, which was either food reinforcement or blackout. Pecking rates during the test stimulus defined discrimination performance. In the simple discrimination, the sample provided the necessary information regarding the subsequent trial outcome. In the conditional discrimination, the sample and test stimuli conjointly provided this information. In Experiment 1, the two procedures were compared with independent groups of pigeons. In Experiment 2, the comparison was made within subjects. The simple discrimination was acquired more quickly and was performed better with a memory requirement. Introduction of long delays disrupted performance even at shorter delays in both discriminations. Postulation of prospective as well as retrospective mediating processes facilitates the interpretation of these results.     

Discrimination of intentional and random motion paths by pigeons

Twelve pigeons (Columba livia) were trained on a go/no-go schedule to discriminate between two kinds of movement patterns of dots, which to human observers appear to be "intentional" and "non-intentional" movements. In experiment 1, the intentional motion stimulus contained one dot (a "wolf") that moved systematically towards another dot as though stalking it, and three distractors ("sheep"). The non-intentional motion stimulus consisted of four distractors but no stalker. Birds showed some improvement of discrimination as the sessions progressed, but high levels of discrimination were not reached. In experiment 2, the same birds were tested with different stimuli. The same parameters were used but the number of intentionally moving dots in the intentional motion stimulus was altered, so that three wolves stalked one sheep. Despite the enhanced difference of movement patterns, the birds did not show any further improvement in discrimination. However, birds for which the non-intentional stimulus was associated with reward showed a decline in discrimination. These results indicated that pigeons can discriminate between stimuli that do and do not contain an element that human observer see as moving intentionally. However, as no feature-positive effect was found in experiment 1, it is assumed that pigeons did not perceive or discriminate these stimuli on the basis that the intentional stimuli contained a feature that the non-intentional stimuli lacked, though the convergence seen in experiment 2 may have been an effective feature for the pigeons. Pigeons seem to be able to recognise some form of multiple simultaneously goal-directed motions, compared to random motions, as a distinctive feature, but do not seem to use simple "intentional" motion paths of two geometrical figures, embedded in random motions, as a feature whose presence or absence differentiates motion displays. Electronic Publication