Comparison of the postural and movement components during learning by dogs of an operant defensive reflex.

Chronic experiments on six dogs using a model of an operant defensive reflex associated with maintenance of flexion of the hindlimb of specified amplitude were performed to compare the characteristics of the postural and movement components during the learning process. Dogs were placed standing on four tension platforms. Signals were activated and data were recorded and stored using original PC programs. Original programs running on another PC were used to analyze the data. All dogs showed a series of characteristics for the appearance of a diagonal pattern of conditioned reflex posture rearrangement. During the period of complete formation of the "coordinated" program of the operant reflex (indicated by high performance criteria for execution of the operant task), the diagonal pattern of posture rearrangement was seen extremely rarely, in only occasional performances in calm dogs. It was only during the period of complete automatization of the movement habit that the diagonal pattern of postural rearrangement was seen consistently in all performances. By this time, there was a sharp increase in the tensogram amplitudes for all four limbs, which sharp increases in the correlation coefficients between individual tensogram performances. These results suggest that in the experimental conditions used here, the diagonal pattern of postural rearrangement appeared significantly later than the movement pattern needed for resolving the operant task. These data also lead to the conclusion that dogs can complete operant defensive limb movements associated with maintaining a specified flexor posture in the absence of preliminary rearrangement of the posture having the diagonal pattern.