THE ERROR-REACTION TIME-CORRELATION AS A PREDICTION OF CATEGORY VERIFICATION MODELS

Several category verification models predict a positive error-reaction time correlation. A speed emphasis category verification task was use d to test the hypothesis that disagreement among subjects' category me mbership judgments accounts for a significant part of that correlation . Two commonly investigated categories were used: vehicle and furnitur e. Results showed that controlling for intersubject disagreement signi ficantly reduced the correlation (alpha = .05). The exact way by which disagreement affected the correlation was also shown. Words that admi tted intersubject differences in category membership judgments produce d false negative and false positive errors, with the latter having a s ignificantly higher probability (alpha = .01). Consequences for genera l memory and categorization models are discussed.