Effects of Lexical Competition on Immediate Memory Span for Spoken Words

Abstract

Current theories and models of the structural organization of verbal short-term memory are primarily based on evidence obtained from manipulations of features inherent m the short-term traces of the presented stimuli, such as phonological similarity. In the present study, we investigated whether properties of the stimuli that are not inherent in the short-term traces of spoken words would affect performance in an immediate memory span task. We studied the lexical neighbourhood properties of the stimulus items, which are based on the structure and organization of words in the mental lexicon. The experiments manipulated lexical competition by varying the phonological neighbourhood structure (i.e., neighbourhood density and neighbourhood frequency) of the words on a test list while controlling for word frequency and intra-set phonological similarity (family size). Immediate memory span for spoken words was measured under repeated and nonrepeated sampling procedures. The results demonstrated that lexical competition only emerged when a nonrepeated sampling procedure was used and the participants had to access new words from their lexicons. These findings were not dependent on individual differences in short-term memory capacity. Additional results showed that the lexical competition effects did not interact with proactive interference. Analyses of error patterns indicated that itemtype errors, but not positional errors, were influenced by the lexical attributes of the stimulus items. These results complement and extend previous findings that have argued for separate contributions of long-term knowledge and short-term memory rehearsal processes in immediate verbal serial recall tasks.One of the major assumptions of current verbal short-term memory (STM) models is the proposal that words are coded and represented in a phonological form in immediate memory. For example, in Baddeley and Hitch's (1974) well-known "working" memory model, they propose a specific mechanism called the "phonological (or articulatory) loop" to support the retention of verbal lists on a temporary basis. The impetus for positing specific mechanisms to handle phonological coding of words stems from a number of robust findings in the STM literature. These include the phonological similarity effect (e.g., Baddeley, 1966;Conrad, 1964), the word length effect (Baddeley, Thomson, & Buchanan, 1975), the unattended speech effect (Salamé & Baddeley, 1982, 1987, and the articulatory suppression effect (Baddeley, Lewis, &Vallar, 1984).The Interpretations given to these diverse findings on immediate memory span have not gone unchallenged (see, e.g., Caplan, Rochon, & Waters, 1992;Caplan & Waters, 1994; © 2003 The Experimental Psychology Society Requests for reprints should be sent to Winston D. Goh, National University of Singapore, Department of Social, Work & Psychology, 11 Law Link, Singapore 117570, Singapore. winston_goh@nus.edu.sg. NIH Public AccessAuthor Manuscript Q J Exp Psychol A. Author manuscript; available in PMC 2012 September 04. ...

The roles of semantic similarity and proactive interference in the word length effect

“…Many item errors tended to be semantic and/or phonological substitutions-for example, oak for oat. This is consistent with previous findings regarding neighborhood errors in STM (Goh & Pisoni, 2003;Roodenrys, Hulme, Lethbridge, Hinton, & Nimmo, 2002). It is harder to make such substitutions for longer words, given that the number of matching competitors should be inversely proportional to word length.…”

Orthographic distinctiveness and memory for order

“…The results from these studies suggest that orthographic distinctiveness may effect encoding in serial recall tasks as it does in recognition tasks; however, direct attempts to examine the effects of orthographic distinctiveness on serial recall have not always found consistent results (cf. Goh & Pisoni, 2003;Jalbert et al, 2011).…”

Lexical learning in bilingual adults: The relative importance of short-term memory for serial order and phonological knowledge

“…A number of studies show that long-term phonological knowledge influences STM performance, as reflected by better recall for words of high versus low lexical frequency or for nonwords containing high versus low probability phonotactic patterns (e.g., Gathercole, Frankish, Pickering & Peaker, 1999;Majerus, Van der Linden, Mulder, Meulemans, & Peters, 2004;Thorn & Gathercole, 1999;Thorn & Frankish, 2005; see also Goh &Pisoni, 2003, andRoodenrys &Hinton, 2002, for related findings). Furthermore, although phonological STM capacity at Age 4 predicts vocabulary knowledge at Age 5, this relationship later reverses, with vocabulary knowledge at Age 5 predicting phonological STM capacity at Age 6 (Gathercole et al, 1992).…”