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.     

Retrospective coding in pigeons' delayed matching-to-sample

In this study we examined how coding processes in pigeons' delayed matching-to-sample were affected by the stimuli to be remembered. In Experiment 1, two groups of pigeons initially learned 0-delay matching-to-sample with identical comparison stimuli (vertical and horizontal lines) but with different sample stimuli (red and green hues or vertical and horizontal lines). Longer delays were then introduced between sample offset and comparison onset to assess whether pigeons were prospectively coding the same events (viz., the correct line comparisons) or retrospectively coding different events (viz., their respective sample stimuli). The hue-sample group matched more accurately and showed a slower rate of forgetting than the line-sample group. In Experiment 2, pigeons were trained with either hues or lines as both sample and comparison stimuli, or with hue samples and line comparisons or vice versa. Subsequent delay tests revealed that the hue-sample groups remembered more accurately and generally showed slower rates of forgetting than the line-sample groups. Comparison dimension had little or no effect on performance. Together, these data suggest that pigeons retrospectively code the samples in delayed matching-to-sample.     

Presence/absence-sample matching by pigeons: divergent retention functions may result from the similarity of behavior during the absence sample and the retention interval

Divergent choose-absence retention functions typically found in pigeons following presence/absence-sample matching have been attributed to the development of a single-code/default coding strategy. However, such effects may result from adventitious differential responding to the samples. In Experiment 1, retention functions were divergent only when differential sample responding could serve as the basis for comparison choice. In Experiment 2, when pecking did not occur during the retention interval, a choose-absence bias was found, but when pecking occurred during the retention interval, a choose-presence bias resulted. In Experiment 3, positive transfer was found when a stimulus associated with the absence of pecking replaced the absence sample but not when a stimulus associated with pecking replaced the presence sample. Thus, presence/absence-sample matching may not encourage the development of a single-code/default coding strategy in pigeons.     

Emergent identity matching after successive matching training. II: Reflexivity or transitivity

Three experiments evaluated whether the apparent reflexivity effect reported by Sweeney and Urcuioli (2010) for pigeons might, in fact, be transitivity. In Experiment 1, pigeons learned symmetrically reinforced hue-form (A-B) and form-hue (B-A) successive matching. Those also trained on form-form (B-B) matching responded more to hue comparisons that matched their preceding samples on subsequent hue-hue (A-A) probe trials. By contrast, most pigeons trained on just A-B and B-A matching did not show this effect; but some did--a finding consistent with transitivity. Experiment 2 showed that the latter pigeons also responded more to form comparisons that matched their preceding samples on form-form (B-B) probe trials. Experiment 3 tested the prediction that hue-hue matching versus hue-hue oddity, respectively, should emerge after symmetrically versus asymmetrically reinforced arbitrary matching relations if those relations are truly transitive. For the few pigeons showing an emergent effect, comparison response rates were higher when a probe-trial comparison matched its preceding sample independently of the baseline contingencies. These results indicate neither a reflexivity nor a transitivity effect but, rather, a possible identity bias.     

Associative symmetry in the pigeon after successive matching-to-sample training

If an organism is explicitly taught an A→B association, then might it also spontaneously learn the symmetrical B→A association? Little evidence attests to such “associative symmetry” in nonhuman animals. We report for the first time a clear case of associative symmetry in the pigeon. Experiment 1 used a successive go/no go matching‐to‐sample procedure, which showed all of the training and testing stimuli in one location and intermixed arbitrary and identity matching trials. We found symmetrical responding that was as robust during testing (B→A) as during training (A→B). In Experiment 2, we trained different pigeons using only arbitrary matching trials before symmetry testing. No symmetrical responding was found. In Experiment 3, we trained other pigeons with only arbitrary matching trials and then tested for symmetry. When these pigeons, too, did not exhibit symmetrical responding, we retrained them with intermixed identity and arbitrary matching trials. Less robust symmetrical responding was obtained here than in Experiment 1. Collectively, these results suggest that identity matching may have to be learned concurrently with arbitrary matching from the outset of training for symmetry to emerge.     

The development of emergent differential sample behavior in pigeons

Three experiments attempted to replicate Manabe, Kawashima, and Staddon's (1995) finding of emergent differential sample behavior in budgerigars that has been interpreted as evidence of functional equivalence class formation. In Experiments 1 and 2, pigeons initially learned two-sample/ two-alternative matching to sample in which comparison presentation was contingent on pecking one sample on a differential-reinforcement-of-low-rate (DRL) schedule and the other on a fixed-ratio (FR) schedule. Later, two new samples were added to the task. Comparison presentation on these trials occurred after the first sample peck following a predetermined interval (Experiment 1) or after completion of either the DRL or FR requirement, whichever occurred first (Experiment 2). Experiment 1 found no evidence for emergent spaced versus rapid responding to the new samples as they established conditional control over the familiar choices. By contrast, differential responding did emerge for some pigeons in Experiment 2, with responding to each new sample coinciding with the pattern explicitly conditioned to the original sample occasioning the same comparison choice. This emergent effect, however, disappeared for most pigeons with continued training. Experiment 3 systematically replicated Experiment 2 using differential peck location as the sample behavior. Differential location pecking emerged to the new samples for most pigeons and remained intact throughout training. Our findings demonstrate a viable pigeon analogue to the budgerigar emergent calling paradigm and are discussed in terms of equivalence- and non-equivalence-based processes.     

Categorical shape and color coding by pigeons

Categorical coding is the tendency to respond similarly to discriminated stimuli. Past research indicates that pigeons can categorize colors according to at least three spectral regions. Two present experiments assessed the categorical coding of shapes and the existence of a higher order color category (all colors). Pigeons were trained on two independent tasks (matching-to-sample, and oddity-from-sample). One task involved red and a plus sign, the other a circle and green. On test trials one of the two comparison stimuli from one task was replaced by one of the stimuli from the other task. Differential performance based on which of the two stimuli from the other task was introduced suggested categorical coding rules. In Experiment 1 evidence for the categorical coding of sample shapes was found. Categorical color coding was also found; however, it was the comparison stimuli rather than the samples that were categorically coded. Experiment 2 replicated the categorical shape sample effect and ruled out the possibility that the particular colors used were responsible for the categorical coding of comparison stimuli. Overall, the results indicate that pigeons can develop categorical rules involving shapes and colors and that the color categories can be hierarchical.     

Episodic-like memory in pigeons

It has been proposed that memory for personal experiences (episodic memory, rather than semantic memory) relies on the conscious review of past experience and thus is unique to humans. In an attempt to demonstrate episodic-like memory in animals, we first trained pigeons to respond to the (nonverbal) question "Did you just peck or did you just refrain from pecking?" by training them on a symbolic matching task with differential responding required to the two line-orientation samples and reinforcing the choice of a red comparison if they had pecked and the choice of a green comparison if they had not pecked. Then, in Experiment 1, after providing the conditions for (but not requiring) the pigeons to peck at one new stimulus (a yellow hue) but not at another (a blue hue), we tested them with the new hue stimuli and the red and green comparisons. In Experiment 2, we tested the pigeons with novel stimuli (a circle, which they spontaneously pecked, and a dark response key, which they did not peck) and the red and green comparisons. In both experiments, pigeons chose the comparison appropriate to the response made to the test stimulus. Thus, the pigeons demonstrated that they could remember specific details about their past experiences, a result consistent with the notion that they have the capacity for forming episodic-like memories.     

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.     

Mediational use of internal representations of food and no-food events by pigeons

Two experiments were conducted with pigeons to determine whether internal representations of food and no-food events can act as mediators in the formation of new associations. Specifically we asked if the representation of an anticipated event can replace the event itself in an established conditional discrimination. In Phase 1 of both experiments pigeons were differentially autoshaped to peck hue stimuli, only one of which was followed by access to food. In Phase 2 they were trained on a symbolic 0-delay matching-to-sample (DMTS) task with food and no-food samples and line-orientation comparisons. In Phase 3 the birds were tested on a symbolic DMTS task in which the hue stimuli from Phase 1 were substituted for the Phase 2 food-event samples. For the Pos group, the hue followed by food in Phase 1 was substituted for the food sample and the hue associated with no-food was substituted for the no-food sample so that the resulting sample/comparison pairings were consistent with the food-event mediator. For the Neg group the pairings were reversed so that the sample/comparison pairings were inconsistent with the food-event mediators. In Experiment 1, when the no-food event was the absence of an event, acquisition of the transfer task was significantly more rapid in the Pos than in the Neg condition. In Experiment 2, when the no-food event was the presentation of an empty food hopper, the Pos group transferred at a significantly higher level than the Neg group. The two experiments provide evidence that in pigeons, representations involving event anticipations can be substituted for, and are thus similar to, the events themselves.     

Evidence for detection of one duration sample and default responding to other duration samples by pigeons may result from an artifact of retention-test ambiguity

S. C. Gaitan and J. T. Wixted (2000) proposed that when pigeons are trained on a conditional discrimination to associate 1 duration sample with 1 comparison and 2 other duration samples with a 2nd comparison, they detect only the single duration, and on trials involving either of the 2 other duration samples, they respond to the other comparison by default. In 2 experiments, the authors show instead that pigeons lend to treat the retention intervals (such as those used by Gaitan and Wixted) as intertrial intervals, and thus, they tend to treat all trials with a delay as 0-s sample trials. The authors tested this hypothesis by showing that divergent retention functions do not appear when the retention interval is discriminably different from the intertrial interval.     

Quantitative analyses of matching-to-sample performance

Six pigeons performed a simultaneous matching-to-sample (MTS) task involving patterns of dots on a liquid-crystal display. Two samples and two comparisons differed in terms of the density of pixels visible through pecking keys mounted in front of the display. Selections of Comparison 1 after Sample 1, and of Comparison 2 after Sample 2, produced intermittent access to food, and errors always produced a time-out. The disparity between the samples and between the comparisons varied across sets of conditions. The ratio of food deliveries for the two correct responses varied over a wide range within each set of conditions, and one condition arranged extinction for correct responses following Sample 1. The quantitative models proposed by Davison and Tustin (1978), Alsop (1991), and Davison (1991) failed to predict performance in some extreme reinforcer-ratio conditions because comparison choice approached indifference (and strong position biases emerged) when the sample clearly signaled a low (or zero) rate of reinforcement. An alternative conceptualization of the reinforcement contingencies operating in MTS tasks is advanced and was supported by further analyses of the data. This model relates the differential responding between the comparisons following each sample to the differential reinforcement for correct responses following that sample.     

Perceptual learning in pigeons: Decreased ability to Discriminate samples mapped onto the same comparison in many-to-one matching

Humans often treat two stimuli that are associated with a common response as similar in other contexts. They do so presumably because those stimuli become conceptually or perceptually more similar to each other (perceptual learning). An analogous phenomenon may occur in pigeons when they are trained with a matching-to-sample procedure in which more than one sample is mapped onto the same comparison. In the present research, pigeons were trained to select one comparison following either of two samples (S₁ or S₂) and to select the other comparison following either of two different samples (S₃ or S₄). When the samples were then presented as positive and negative stimuli in a simple successive discrimination, samples that had been associated with the same comparison during original training (e.g., S₁ vs. S₂) were more difficult to discriminate than were samples that had been associated with different comparisons (e.g., S₁ vs. S₃). Thus, it appears that perceptual learning occurs in pigeons as well.     

Use of samples differing markedly in salience may encourage use of a single-code/default strategy in matching-to-sample in pigeons

To test the hypothesis that pigeons will only code the more salient sample when samples differ markedly in salience, pigeons were trained with samples consisting of a 2-s presentation of food (highly salient sample) and an 8-s presentation of keylight (less salient sample). During retention testing, pigeons tended to respond at longer delays as if an 8-s keylight sample had been presented. This finding is consistent with use of a single-code/default strategy in which only the 2-s food sample was coded and the comparison associated with an 8-s keylight sample was selected by default in the absence of memory for the salient 2-s food sample. Hence, a marked difference in sample salience appears to encourage use of a single-code/default strategy.     

Relative versus absolute stimulus control in the temporal bisection task

When subjects learn to associate two sample durations with two comparison keys, do they learn to associate the keys with the short and long samples (relational hypothesis), or with the specific sample durations (absolute hypothesis)? We exposed 16 pigeons to an ABA design in which phases A and B corresponded to tasks using samples of 1 s and 4 s, or 4 s and 16 s. Across phases, we varied the mapping between the samples and the keys. For group Relative, short and long samples were always associated with the same keys (e.g., Phase A: ‘1s→ Left, 4s→ Right''; Phase B: ‘4s→ Left, 16s→ Right''); for group Absolute, the 4-s sample was associated always with the same key (e.g., Phase A: ‘1s→ Left, 4s→ Right''; Phase B: ‘16s→ Left, 4s→ Right’). If temporal control is relational, group Relative should learn the new task faster than group Absolute, but if temporal control is absolute, the opposite should occur. We compared the results with the predictions of the Learning-to-Time (LeT) model, which accounts for temporal discrimination in terms of absolute stimulus control and stimulus generalization. The acquisition curves of the two groups were generally consistent with LeT and therefore more consistent with the absolute than the relative hypothesis.     

Some observations on control by spoken words in children's conditional discrimination and matching by exclusion

This study examined matching-to-sample procedures that might result in the emergence of conditional behavior never explicitly taught. Subjects were preschool children. Two pictures were displayed as comparisons on every trial, and samples were spoken words. In baseline training preceding each of three experiments, children learned to select pictures of a dog, a table, and a banana in response to their spoken English names. Thereafter, probe trials displayed novel comparisons with baseline comparisons: one novel comparison was displayed with the dog and another with the table. The three experiments differed primarily in the nature of the samples presented on probe trials. In Experiment 1, probe samples were novel words, "JAIJAI" and "BREEL." On the probes, each of seven subjects reliably selected the novel comparisons, apparently "excluding" the familiar ones. In Experiment 2, probe samples were from the subjects' baseline. On one probe, for example, the sample was "TABLE," and the subject had to choose either the dog or the novel picture. Exclusion was logically possible because the dog had always before been incorrect in the presence of "TABLE." Under these conditions, however, only two of nine children excluded reliably. In Experiment 3, probe samples were words that had never been samples on any matching-to-sample trial, but that had controlled the children's behavior in other settings. On one probe, for example, the sample was "PENCIL," and the subject had to choose either the dog or a novel picture. Subjects virtually always excluded the former and selected the latter. Unreliable exclusion in Experiment 2, therefore, apparently resulted because the probe samples had previously served also as samples on baseline trials. Spontaneous verbalizations recorded during probing provided further data consistent with this interpretation. The study helps to define variables controlling exclusion performances by showing that such performances are more likely to occur if the sample has no prior experimental history.     

Summation of symbols by pigeons (Columba livia): the importance of number and mass of reward items

In each of 3 experiments, different sets of 4 pigeons (Columba livia) were trained to discriminate between 2 visual symbols that covered wells containing food items that varied in number, mass, or both. In Experiment 1, the symbols were associated with 0, 1, 3, 5, 7, or 9 pieces of grain reward. The pigeons learned to choose the symbol corresponding to the larger reward, and on summation tests, they chose the pair of symbols that summed to the larger total reward. When number of food pellets was varied but mass of reward was held constant in Experiment 2, preference for the larger number symbols failed to appear. When number was held constant and mass was varied in Experiment 3, the pigeons showed a clear preference for the larger mass symbols on single-symbol and summation tests. These findings show that pigeons summate the value of symbols and are more likely to represent symbols by mass of food reward than by number of food items.     

Pigeons group time intervals according to their relative duration

In the present research, we asked whether pigeons tended to judge time intervals not only in terms of their absolute value but also relative to a duration from which they must be discriminated (i.e., longer or shorter). Pigeons were trained on two independent temporal discriminations. In one discrimination, sample durations of 2 and 8 sec were associated with, for example, red and green hue comparisons, respectively, and in the other discrimination, sample durations of 4 and 16 sec were associated with vertical and horizontal line comparisons, respectively. If pigeons are trained on a temporal discrimination and tested with intermediate durations, the subjective midpoint typically occurs close to the geometric mean of the two trained values. The 4- and 8-sec values were selected to be the geometric mean of the two values in the other discrimination. When a 4-sec test sample was presented with the comparisons from the 2- and 8-sec discrimination, the pigeons preferred the comparison associated with the shorter sample. Similarly, when an 8-sec test sample was presented with the comparisons from the 4-and 16-sec discrimination, the pigeons preferred the comparison associated with the longer sample. Thus, a relative grouping effect was found. That is, durations that should have produced indifferent choice were influenced by their relative durations (shorter than or longer than the alternative) during training.     

Transformation of the discriminative and eliciting functions of generalized relational stimuli

In three experiments, match-to-sample procedures were used with undergraduates to establish arbitrary relational functions for three abstract visual stimuli. In the presence of samples A, B, and C, participants were trained to select the smallest, middle, and largest member, respectively, of a series of three-comparison arrays. In Experiment 1, the B (choose middle) stimulus was then used to train a steady rate of keyboard pressing before the A (choose smallest) and the C (choose largest) stimuli were presented. Participants pressed slower to A and faster to C than to B. Then B was paired with mild shock in a Pavlovian procedure with skin conductance change as the dependent variable. When presented with A and C, 6 of 8 experimental participants showed smaller skin conductance changes to A and larger skin conductance changes to C than to B. In Experiment 2, A was then used as a sample in a match-to-sample procedure to establish an arbitrary size ranking among four same-sized colored circle comparisons. One of the middle circles was then used to establish a steady rate of pressing before the other circles were presented. Five of 6 participants responded slower to the "smaller" circle and faster to the "larger" circle than they did to the "middle" circle. In Experiment 3, A, B, and C were then presented on a series of test trials requiring participants to pick the comparison that was less than, greater than, or equal to the sample. Novel stimuli were included on some trials. Results indicated that the relational training procedures produced derived relations among the stimuli used in training and that these allowed correct inferences of relative size ranking among novel stimuli.     

Relational Learning in Pigeons?

Two experiments investigated whether discrimination learning and transposition by pigeons were facilitated by the opportunity to compare rectangles differing in luminance or stars differing in number of vertices. In Experiment 1, one group was trained with stimuli from the same dimension appearing simultaneously on each trial, but for a second group such stimuli appeared on separate trials. The opportunity to compare stimuli from the same dimension on a single trial facilitated the learning of the luminance discrimination but not of the stars discrimination. Such comparison also resulted in greater luminance but not greater stars transposition. Using a different training procedure, Experiment 2 confirmed that the opportunity for comparison facilitated a luminance discrimination. The results for the star discrimination are entirely consistent with 'absolute' theories of discrimination learning; but the results for the luminance discrimination suggest some kind of 'relational' learning. Given the difference between the star and luminance discriminations, a low-level, sensory theory of relational learning seems most consistent with the data.