Cue interaction in human contingency judgment

摘要: Most studies of human contingency judgment have been based on the assumption that frequency information about one predictor is assessed in isolation other predictors. Recent evidence, however, suggests judged predictive strength cue influenced by strengths copresent cues. Two experiments demonstrate stimuli with same outcome contingencies may nonetheless different as result interaction. The first experiment, which a within-subject design was used, provides demonstration blocking. A stimulus presented compound strong rated less than another in-compound non-predictive cue. In second interactions conditioned inhibition were examined. gained negative presentations positive when not presented. This compared an otherwise analogous predictor. These results support use animal-conditioning models accounts learning.