Recognition memory for one-trial-unitized word pairs: evidence from event-related potentials

Abstract

Performance in tests of associative memory is generally thought to require recollection while familiarity cannot support memory for associations. However, recent research suggested that familiarity contributes to associative memory when the to-be-associated stimuli are unitized during encoding. Here, we investigated the electrophysiological correlates of retrieval of word pairs after two different encoding conditions. Semantically unrelated word pairs were presented as separate lexical units in a sentence frame (non-unitized word pairs) or together with a definition that allows to combine word pairs to a new concept (unitized word pairs). At test, participants discriminated between word pairs that appeared in the same pairing during study, recombined, or new pairs. Memory processes were examined by means of event-related potentials (ERPs). An early old/new effect with a parietal maximum was found for unitized word pairs while a qualitatively different late old/new effect was elicited by non-unitized word pairs, only. These findings suggest that one-trial-unitized word pairs are recognized differently from non-unitized word pairs. We will discuss the possibility that unitization leads to the engagement of specific forms of familiarity-conceptual fluency and absolute familiarity.