THE BASIS OF SUPERSTITIOUS BEHAVIOR: CHANCE CONTINGENCY,STIMULUS SUBSTITUTION,OR APPETITIVE BEHAVIOR?

This research examined three explanations for the "superstitious" behavior of pigeons under frequent fixed-time delivery of food: accidental response-reward contingency, stimulus substitution, and elicited species-typical appetitive behavior. The behavior observed in these studies consisted of occasional postfood locomotion away from the food hopper, and a predominant pattern of activity directed toward the hopper wall (wall-directed behavior), including approaching, stepping side to side, scratching with the feet, bumping with the breast, pendulum movements of the extended neck, and head bobbing, though not pecking. The consistency of these behavior patterns argued against explanation by accidental response contingencies, and the complexity of behavior was incompatible with the classic stimulus-substitution account. These studies also showed that: (1) response contingencies and prior stimulus experience can modify wall-directed behavior, but within definable limits; (2) pecking sometimes can be obtained in birds of specific strains, and by providing extended training; (3) placing the hopper in the floor at the center of a large chamber replaces wall-directed behavior with circling in a manner that resembles ground foraging for food. We conclude that superstitious behavior under periodic delivery of food probably develops from components of species-typical patterns of appetitive behavior related to feeding. These patterns are elicited by a combination of frequent food presentations and the supporting stimuli present in the environment.     

Incentive theory: IV. Magnitude of reward

Incentive theory is successfully applied to data from experiments in which the amount of food reward is varied. This is accomplished by assuming that incentive value is a negatively accelerated function of reward duration. The interaction of the magnitude of a reward with its delay is confirmed, and the causes and implications of this interaction are discussed.     

Testing the efficiency of Markov chain Monte Carlo with People using facial affect categories

Exploring how people represent natural categories is a key step toward developing a better understanding of how people learn, form memories, and make decisions. Much research on categorization has focused on artificial categories that are created in the laboratory, since studying natural categories defined on high-dimensional stimuli such as images is methodologically challenging. Recent work has produced methods for identifying these representations from observed behavior, such as reverse correlation (RC). We compare RC against an alternative method for inferring the structure of natural categories called Markov chain Monte Carlo with People (MCMCP). Based on an algorithm used in computer science and statistics, MCMCP provides a way to sample from the set of stimuli associated with a natural category. We apply MCMCP and RC to the problem of recovering natural categories that correspond to two kinds of facial affect (happy and sad) from realistic images of faces. Our results show that MCMCP requires fewer trials to obtain a higher quality estimate of people's mental representations of these two categories.     

A transfer of functions through derived arbitrary and nonarbitrary stimulus relations

During Experiments 1 and 2, subjects were trained in a series of related conditional discriminations in a matching-to-sample format (A1-B1, A1-C1 and A2-B2, A2-C2). A low-rate performance was then explicitly trained in the presence of B1, and a high-rate performance was explicitly trained in the presence of B2. The two types of schedule performance transferred to the C stimuli for all subjects in both experiments, in the absence of explicit reinforcement through equivalence (i.e., C1 = low rate and C2 = high rate). In Experiment 2, it was also shown that these discriminative functions transferred from the C1-C2 stimuli to two novel stimuli that were physically similar to the C stimuli (SC1 and SC2, respectively). For both these experiments, subjects demonstrated the predicted equivalence responding during matching-to-sample equivalence tests. In Experiments 3 and 4, the conditional discrimination training from the first two experiments was modified in that two further conditional discrimination tasks were trained (C1-D1 and C2-D2). However, for these tasks the D stimuli served only as positive comparisons, and ND1 and ND2 stimuli served as negative comparisons (i.e., C1 × ND1 and C2 × ND2). Subsequent to training, the negatively related stimuli (ND1 and ND2) did not become discriminative for the schedule performances explicitly trained in the presence of B1 and B2, respectively. Instead, the ND1 stimulus became discriminative for the schedule performance trained in the presence of B2, and ND2 became discriminative for the schedule performance trained in the presence of B1. All subjects from Experiment 4 showed that the novel stimulus SND1, which was physically similar to ND1, became discriminative for the same response pattern as that controlled by ND1. Similarly, SND2, which was physically similar to ND2, became discriminative for the same response pattern as that controlled by ND2. Subjects from both Experiments 3 and 4 also produced equivalence responding on matching-to-sample equivalence tests that corresponded perfectly to the derived performances shown on the transfer of discriminative control tests.     

Undermatching on concurrent variable-interval schedules and the power law.

The phenomenon of undermatching on concurrent variable-interval schedules is shown to be derivable by transforming the individual interreinforcement intervals of each variable-interval schedule and averaging the transformed values to produce an "estimate" of the rate of reinforcement the schedules deliver. If the transformation is based on a power function with a fractional exponent, such as is found in many studies of temporal control in animals, matching response rations to the ratios of these estimated rates of reinforcement yields undermatching. If the concurrent variable-interval schedules are arranged such that the individual intervals in each schedule have a constant proportionality (a procedure found in many commonly used variable-interval schedules) the slope of the line relating logarithms of response ratios and of programmed reinforcement ratios is identical to the exponent of the power transformation applied to the individual time intervals in the variable-interval schedules. In other cases this simple relation does not hold but the degree of undermatching is greater the lower the value of the exponent of the power function. This account of undermatching predicts values similar to those typically observed.     

Income and choice between different goods

In Experiment 1, 3 rats chose between two simultaneously operating variable-interval schedules, one of which provided saccharin water and the other, food. In five conditions, the absolute (and equal) reinforcement rates provided by the pair of equal-valued schedules were manipulated in the range of 36 to 240 per hour. Experiment 2 was identical to Experiment 1 except that these schedules operated successively, arranged by requiring the rat to stand on the side of the chamber correlated with each schedule. Food/saccharin choice ratios were inversely related to reinforcement rate in both experiments, although this effect was stronger in Experiment 2. When delivery rates were high, preference for food over saccharin often reversed as the session progressed. The results were interpretable in terms of economic accounts of choice (e.g., the minimum-needs hypothesis), as well as in terms of traditional psychological accounts (e.g., matching theory).     

Consumption-leisure tradeoffs in pigeons: Effects of changing marginal wage rates by varying amount of reinforcement

Pigeons' rates of responding and food reinforcement under simple random-ratio schedules were compared with those obtained under comparable ratio schedules in which free food deliveries were added, but the duration of each food delivery was halved. These ratio-with-free-food schedules were constructed so that, were the pigeon to maintain the same rate of responding as it had under the simple ratio schedule, total food obtained (earned plus free) would remain unchanged. However, any reduction in responding would reduce total food consumption below that under the simple ratio schedule. These “compensated wage decreases” led to decreases in responding and decreases in food consumption, as predicted by an economic model of labor supply. Moreover, the reductions in responding increased as the ratio value increased (i.e., as wage rates decreased). Pigeons, therefore, substituted leisure for consumption. The relationship between these procedures and negative-income-tax programs is noted.     

Optimality And Concurrent Variable-interval Variable-ratio Schedules

Despite claims to the contrary, all leading theories about operant choice may be seen as models of optimality. Although melioration is often contrasted with global maximization, both make the same core assumptions as other versions of optimality theory, including momentary maximizing, hill climbing, and the various versions of optimal foraging theory. The present experiment aimed to test melioration against more global optimality and to apply the visit-by-visit analysis suggested by foraging theory. Rats were exposed to concurrent schedules in which one alternative was always variable-ratio 10 and the other alternative was a variable-interval schedule. Although choice relations varied from rat to rat, the overall results roughly confirmed the matching law, a result often taken to support melioration. Pooling the data across sessions and across rats, however, resulted in no increment in unsystematic variance, lending support to the contention by Ziriax and Silberberg (1984) that the choice relation is partly constrained. When the data were analyzed at the level of visits, the results either disconfirmed predictions of melioration or showed regularities about which melioration is silent. Instead, performance tended toward a rough optimization, in which responding favored the variable ratio, but with relatively brief visits to the variable interval. There were no asymmetries in travel or variability that would indicate that different processes were involved in generating visits at the two different schedules. The findings point toward a more global optimality model than melioration and demonstrate the value of per-visit analysis in the study of concurrent performances.     

Responding of pigeons under variable-interval schedules of signaled-delayed reinforcement: effects of delay-signal duration.

Two experiments with pigeons examined the relation of the duration of a signal for delay ("delay signal") to rates of key pecking. The first employed a multiple schedule comprised of two components with equal variable-interval 60-s schedules of 27-s delayed food reinforcement. In one component, a short (0.5-s) delay signal, presented immediately following the key peck that began the delay, was increased in duration across phases; in the second component the delay signal initially was equal to the length of the programmed delay (27 s) and was decreased across phases. Response rates prior to delays were an increasing function of delay-signal duration. As the delay signal was decreased in duration, response rates were generally higher than those obtained under identical delay-signal durations as the signal was increased in duration. In Experiment 2 a single variable-interval 60-s schedule of 27-s delayed reinforcement was used. Delay-signal durations were again increased gradually across phases. As in Experiment 1, response rates increased as the delay-signal duration was increased. Following the phase during which the signal lasted the entire delay, shorter delay-signal-duration conditions were introduced abruptly, rather than gradually as in Experiment 1, to determine whether the gradual shortening of the delay signal accounted for the differences observed in response rates under identical delay-signal conditions in Experiment 1. Response rates obtained during the second exposures to the conditions with shorter signals were higher than those observed under identical conditions as the signal duration was increased, as in Experiment 1. In both experiments, rates and patterns of responding during delays varied greatly across subjects and were not systematically related to delay-signal durations. The effects of the delay signal may be related to the signal's role as a discriminative stimulus for adventitiously reinforced intradelay behavior, or the delay signal may have served as a conditioned reinforcer by virtue of the temporal relation between it and presentation of food.