Abstract: Although reinforcement often leads to repetitive, even stereotyped responding, that is not a necessary outcome. When it depends on variations, reinforcement results in responding that is diverse, novel, indeed unpredictable, with distributions sometimes approaching those of a random process. This article reviews evidence for the powerful and precise control by reinforcement over behavioral variability, evidence obtained from human and animal-model studies, and implications of such control. For example, reinforcement of variability facilitates learning of complex new responses, aids problem solving, and may contribute to creativity. Depression and autism are characterized by abnormally repetitive behaviors, but individuals afflicted with such psychopathologies can learn to vary their behaviors when reinforced for so doing. And reinforced variability may help to solve a basic puzzle concerning the nature of voluntary action.

Abstract: For 30 years Gallup's (Science 167:86-87, 1970) mark test, which consists of confronting a mirror-experienced test animal with its own previously altered mirror image, usually a color mark on forehead, eyebrow or ear, has delivered valuable results about the distribution of visual self-recognition in non-human primates. Chimpanzees, bonobos, orangutans and, less frequently, gorillas can learn to correctly understand the reflection of their body in a mirror. However, the standard version of the mark test is good only for positively proving the existence of self-recognition. Conclusive statements about the lack of self-recognition are more difficult because of the methodological constraints of the test. This situation has led to a persistent controversy about the power of Gallup's original technique. We devised a new variant of the test which permits more unequivocal decisions about both the presence and absence of self-recognition. This new procedure was tested with marmoset monkeys (Callithrix jacchus), following extensive training with mirror-related tasks to facilitate performance in the standard mark test. The results show that a slightly altered mark test with a new marking substance (chocolate cream) can help to reliably discriminate between true negative results, indicating a real lack of ability to recognize oneself in a mirror, from false negative results that are due to methodological particularities of the standard test. Finally, an evolutionary hypothesis is put forward as to why many primates can use a mirror instrumentally – i.e. know how to use it for grasping at hidden objects – while failing in the decisive mark test.

Abstract: We present the results of two experiments that probe the ability of rhesus macaques to match visual arrays based on number. Three monkeys were first trained on a delayed match-to-sample paradigm (DMTS) to match stimuli on the basis of number and ignore continuous dimensions such as element size, cumulative surface area, and density. Monkeys were then tested in a numerical bisection experiment that required them to indicate whether a sample numerosity was closer to a small or large anchor value. Results indicated that, for two sets of anchor values with the same ratio, the probability of choosing the larger anchor value systematically increased with the sample number and the psychometric functions superimposed. A second experiment employed a numerical DMTS task in which the choice values contained an exact numerical match to the sample and a distracter that varied in number. Both accuracy and reaction time were modulated by the ratio between the correct numerical match and the distracter, as predicted by Weber's Law.

Abstract: Past evidence that pigeons may adopt a single-code/default strategy to solve duration sample discriminations may be attributable to the similarity between the intertrial interval (ITI) and the retention interval. The present experiments tested whether pigeons would adopt a single-code/default strategy when possible ITI-retention-interval ambiguity was eliminated and sample salience was increased. Previous studies of duration sample discriminations that have purported to show evidence for the use of a single-code/default coding strategy have used durations of 0, 2, and 10 sec (Zentall, Klein, and Singer, 2004). However, the results of Experiment 1 suggest that the use of a 0-sec sample may produce an artifact resulting in inadvertent present/absent sample matching. In Experiment 2, when pigeons were trained with three nonzero duration samples (2, 8, and 32 sec), clear evidence for the use of a single-code/default strategy was found.

Abstract: Clement, Feltus, Kaiser, and Zentall (2000) reported that pigeons prefer discriminative stimuli that require greater effort (more pecks) to obtain over those that require less effort. In the present experiment, we examined two variables associated with this phenomenon. First, we asked whether delay of reinforcement, presumably a relatively aversive event similar to effort, would produce similar effects. Second, we asked whether the stimulus preference produced by a prior relatively aversive event depends on its anticipation. Anticipation of delay was accomplished by signaling its occurrence. Results indicated that delays can produce preferences similar to those produced by increased effort, but only if the delays are signaled.