The range of automatic spreading activation in word priming

Abstract

Four unmasked-prime and three masked-prime experiments were used to investigate how far automatic spreading activation (ASA) spreads in semantic memory. The results from the unmasked-prime experiments indicated a one-step spread from the prime's representation to those of its direct associates. These results were inconclusive as to whether ASA spreads, in two steps, to representations of words indirectly related to the prime via an intervening word association. This inconclusiveness was due to a possible postaccess process affecting unrelated prime—target pairs and possibly also affecting indirectly related pairs. In the masked-prime experiments such a process was prevented. Again, one-step ASA was supported, but the data unambiguously ruled out any further spread. Implications of this finding with respect to the underlying memory structure are discussed.