80 college Ss, divided equally between 2-choice and 4-choice tasks on the Selective Mathometer, practiced 20-unit S-R sequences for 30 trials by the correction method. Stimuli were 20 random dot patterns, all Ss being run under different alternating-response permutations. Each 2-choice key was assigned to 10 different slides, each 4-choice key to 5 different slides. Half the 2-choice Ss practiced on adjacent keys, half on remote. Key spacing was not significant but response availability differences, as predicted, produced superior performance for the 2-choice task. Practice, availability, and their interaction were all significant, p < .001. Application of the equation Rp = a(i)(r(N)) to correct first choices over reinforced trials (N) resulted in typical Rp curve forms (fits > 98%) consistent with the rational parameters of asymptote (a), initial probability (i), and rate (r). Mean latencies (Rt) of correct responses and mean errors (R-) decreased as negatively-accelerated functions of N. All main effects and interactions for Rt and R- scores were significant, p <.001.
Download full-text PDF |
Source |
---|---|
http://dx.doi.org/10.1080/00222895.1972.10734920 | DOI Listing |
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!
© LitMetric 2025. All rights reserved.