TESTED THE HYPOTHESIS THAT THE MAGNITUDES OF THE PROBABILITIES COMPRISING THE LIKELIHOOD RATIO INFLUENCE THE DEGREE OF CONSERVATISM EVIDENCED IN A SUBJECTIVE PROBABILITY REVISION TASK. SS WERE 25 MALE UNDERGRADUATES. 2 SETS OF SAMPLING DISTRIBUTIONS WERE CONSTRUCTED SO THAT WHILE THE RATIOS WERE THE SAME IN EACH, THE PROBABILITIES CONSTITUTING THE RATIOS WERE LARGE FOR 1 SET AND SMALL FOR THE OTHER. THEN 2 GROUPS OF SS BASED SUBJECTIVE PROBABILITY REVISIONS ON THE SETS OF DISTRIBUTIONS. THE RATIOS HAD PRIMARY INFLUENCE ON REVISION. HOWEVER, FOR A GIVEN RATIO THE REVISIONS WERE MORE CONSERVATIVE IF THE COMPONENT PROBABILITIES WERE SMALL. THIS FACTOR MAY ACCOUNT FOR SOME OF THE CONSERVATISM REPORTED IN PREVIOUS EXPERIMENTS. (PsycINFO Database Record (c) 2006 APA, all rights reserved)