Abstract

Four experiments examined whether a word's meaning can be processed at the detection threshold at which people can only judge the presence of the word at a chance level or any other identification thres­holds. Previous lexical decision studies reporting evidence for preconscious perception seemed to have suffered from conceptual as well as methodological weakness. The present study explored possible thres­hold conditions for preconscious perception, along with a test of an alternative hypothesis for the previous findings, namely the fragment hypothesis. Our experiments showed that masked prime words at a 50% identification threshold facilitate the processing of semantically related probe words and that the facilitation effects are independent of the associative strengths between the prime and probe. The latter result seems to support Freud's hypothesis on the multiple activations of associative pathways re­lated to an unidentified stimulus.