Campbell, Fiona Margaret; (2000) Imitative and Nonimitative Social Learning In a Two - Object / Two - Action Procedure. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text Imitative and nonimitative social learning in a two-object, two-action procedure.pdf Download (7MB) | Preview

Abstract

It is widely believed that observation of a conspecific performing an action on an object for food can facilitate acquisition of the observed response via imitative, in addition to, nonimitative social learning. Imitative social learning consists of response learning by observation (Heyes, 1993; Heyes & Ray, in press). It allows animals to learn responses, actions, or patterns of behaviour; how to execute them and what are their consequences (observational learning or imitation). Nonimitative social learning, on the other hand, consists of stimulus learning by observation (Heyes, 1993; Heyes & Ray, in press). It is the means by which animals acquire information about stimuli, objects, or events in the environment; their presence or location (stimulus enhancement), dynamic properties (emulation learning), and/or value (observational conditioning). The experiments reported in this thesis used a two-object/two-action procedure in an attempt to distinguish these two forms of learning (Ray, 1997). In Experiment 1, naive rats observed from one side of an operant chamber while demonstrators manipulated either a left or a right lever by lifting up or pressing down. When subsequently allowed access to the levers on test and rewarded for all responses, regardless of location and direction, observer rats showed a reliable tendency to manipulate the same lever in the same direction as their demonstrator. Unfortunately, these effects were not particularly robust. Numerous attempts to replicate Experiment 1 yielded either the location effect, the direction effect, or no effect of conspecific observation. Only in the reported experiment were both effects obtained simultaneously. Despite this problem, these results are still consistent with the hypothesis that rats can acquire information about both a stimulus and a response through conspecific observation. However, follow-up studies failed to support this impression. Instead, they indicated that although a number of different of social influences may act upon the rat's behaviour in a two-object/two-action procedure, response learning by observation may not be one of them. In Experiments 4-7, whether rats were exposed to the delivery of food following each of their demonstrator's responses was manipulated in order to examine the role played by demonstrator reinforcement in lever choice. These experiments confirmed previous findings by showing that rats are exposed to levers as a result of observing them pressed (Heyes, Ray, Mitchell, & Nokes, 1999), and, in addition, revealed that reinforcement of demonstrators' responses increased the probability that rats would approach and contact the lever which their demonstrator operated. Once in the vicinity of the lever, rats encountered odour cues deposited by demonstrators during the course of instrumental responding. These cues were found to be sufficient to bias rats' responses in favour of their demonstrator's direction (Experiments 2 and 3). A two-object/two-action procedure was also used in Experiment 8, where naive starlings observed demonstrators displacing either a red or a black plug from a hole in the lid of a plastic box by lifting up or pressing down. When presented with a sealed box on test, observer birds displaced the same plug in the same direction as their demonstrator. In contrast to rats, starlings showed strong effects of conspecific observation that could not be accounted for by demonstrator-deposited odour cues. Therefore, it is possible that this paradigm may be well-suited for the task of analysing both the psychological mechanisms of, and distinctive conditions favouring, imitative and nonimitative social learning.

Download activity - last month

Export as JSONCSVXML

Download activity - last 12 months

Export as JSONCSVXML

Downloads by country - last 12 months

Export as CSVXMLJSON

Archive Staff Only

View Item