Effect of response rates on non-distracted and distracted conditional discrimination performance

Abstract

Two experiments were carried out in order to see if higher rates of responding during training result in higher stability of accuracy in the face of distraction. In the first experiment hens were exposed to a repeated acquisition procedure where they were taught to peck a number of 3 key combination response sequences (chains). In the first condition hens had the opportunity respond at any rate, in the second condition delays between each correct response were used to reduce response rate. Each session was made up of two periods. During the first period (training), hens learnt to complete chains. The second period was a test period where a stroboscope was used to distract behaviour. Accuracy during each condition and period was measured. The number of trials and rate of reinforcement were held constant. This procedure was repeated in a second experiment in order to test if habituation to the stroboscope was occurring. During both experiments the stroboscope was found to distract behaviour during the test period. During training, accuracy was higher for the no-delay condition than the delay condition. There were no large differences in accuracy between the no-delays condition and the delays condition during the test period, this was the same for both experiments. The second experiment showed that repeated presentation of the stroboscope resulted in higher accuracy meaning that habituation had occurred. The findings showed that when rate of reinforcement and number of practices were controlled, rate of responding alone during training did not change the stability of accuracy during distraction which was unexpected.