OF THREAD AND MUM: FAMILARITY AND RECOLLECTIVE MECHANISMS IN THE PROCESSING OF AMBIGUOUS WORDS

Abstract

Performance on lexically ambiguous words (e.g., thread) is better than unambiguous words (e.g., ant) - an ambiguity advantage - when they are categorised as a word in the lexical decision task (LDT). However, an ambiguity disadvantage occurs when words are categorised with classes of things that they have shared properties with, in the semantic categorisation task (SCT; Hino et al., 2002; Piercey & Joordens, 2000). The role of relatedness of ambiguous words’ definitions as well as familiarity-based and recollection-based memory retrieval processing strategies, in producing this different outcome, is examined in the present research. Following the norming of relatedness of ambiguous words’ definitions in four pre-tests, two virtual LDT experiments and two SCTs were conducted to ascertain the data trends from previous literature. Finally, five memory tasks with outcomes that leverage the spectrum of familiarity-based to recollection-based processing were deployed. The findings suggest that relationships between definitions and processing strategies used during tasks affect performance on ambiguous words.