Effect of mean and variability of event run length on two-choice learning.

Publisher Website
Google Scholar
Abstract
IN A 2 * 3 FACTORIAL DESIGN, 6 GROUPS OF 24 SS PREDICTED WHETHER E WOULD CALL OUT A 1 OR A 2 ON EACH OF 540 TRIALS. THE INDEPENDENT VARIABLES WERE MEAN LENGTH OF EVENT RUNS (4.5 AND 7.5) AND VARIABILITY OF RUN LENGTH (LOW, MEDIUM, AND HIGH). THE MAJOR FINDINGS WERE (1) MORE SS PREDICTED THAT RUNS WOULD END TOO SOON (EARLY ERRORS) IN THE LOW MEAN GROUPS THAN IN THE HIGH MEAN GROUPS, (2) THE EFFECT OF INCREASING MEAN WAS TO INCREASE THE NUMBER OF CONSECUTIVE EVENTS WHICH OCCUR BEFORE THE RESPONSE PROBABILITY CURVE BEGINS TO DECLINE, (3) GIVEN THE OCCURRENCE OF THE LAST EVENT IN THE LONGEST RUN FOR EACH GROUP, PREDICTIONS THAT THE RUN WOULD CONTINUE (LATE ERRORS) INCREASED AS VARIABILITY INCREASED, AND (4) REPETITION RESPONSES, PREDICTIONS OF THE EVENT WHICH OCCURRED ON THE PRECEDING TRIAL, INCREASED AS VARIABILITY INCREASED. IT WAS CONCLUDED THAT VARIABILITY OF EVENT RUNS IS A MAJOR FACTOR IN THE OCCURRENCE OF NEGATIVE AND POSITIVE RECENCY EFFECTS. (PsycINFO Database Record (c) 2006 APA, all rights reserved)

This publication has 0 references indexed in Scilit: