Partial Reinforcement Effects in Avoidance Conditioning

Abstract

& Eckert, 1775). The purpose of the present study was to examine the effects of partial and continuous reinforcement in avoidance condirioning. Also, the magnitude of negative reinforcement was varied. A total of 30 male albino rats, 100 days old, were given 66 avoidance conditioning trials (G/day). Three groups of 10 subjects each ran in a commercially made straight alley manufactured by the Hunter Co., Iowa City, Iowa. The alley was constructed of Plexiglas with a grid floor and was 152.4 cm long, 15.24 cm high, and 10.16 cm wide. All groups received .5-mA shock in the start and run sections of the alley when a 5-sec. avoidance criterion was not met. The three groups differed in the amount of shock received in the goal box. Subjects partially reinforced with large shock received .5 mA in the goal box on 50% of the trials and no shock in the goal box on 50% of the trials. Subjects partially reinforced with small shock received .2 mA in the goal box on 50% of the trials and no shock in the goal box on 50% of the trials. The continuously reinforced group received no shock in the goal box during acquisition. On Day 12 all groups were shifted to an exrinction schedule where they received .5 mA in the goal box for 5 sec. During extinction the avoidance contingency was not in effect so that no shock was delivered regardless of how long the subject remained in che start and runway sections of the alley. The extinction phase continued for 2 days (6 trials/day). Repeated measures analyses were performed on total speeds (ft./sec.) for Day 1 and Day 2 of extinction. The finding of greatest interest was the significant Group x Trials interaction on Day 1 (F10.135 = 4.77, p < .0001). Tukey's Post hoc procedure indicated that as early as the second trial continuously reinforced Ss (M = ,329) and Ss partially reinforced with small shock (M = .381), while not significantly different from each other. were both inferior to the group of Ss partially reinforced with large shock (M = 1.018). The present experiment demonstrates that the partial reinforcement effect occurs in avoidance conditioning, given appropriate experimental contingencies. These data further suggest a commonality berween positive and negative reinforcement with respect to behavioral phenomena and theoretical accounts of those phenomena.