Does segmental overlap help or hurt? Evidence from blocked cyclic naming in spoken and written production

Abstract

Past research has demonstrated interference effects when words are named in the context of multiple items that share a meaning. This interference has been explained within various incremental learning accounts of word production, which propose that each attempt at mapping semantic features to lexical items induces slight but persistent changes that result in cumulative interference. We examined whether similar interference-generating mechanisms operate during the mapping of lexical items to segments by examining the production of words in the context of others that share segments. Previous research has shown that initial-segment overlap amongst a set of target words produces facilitation, not interference. However, this initial-segment facilitation is likely due to strategic preparation, an external factor that may mask underlying interference. In the present study, we applied a novel manipulation in which the segmental overlap across target items was distributed unpredictably across word positions, in order to reduce strategic response preparation. This manipulation led to interference in both spoken (Exp. 1) and written (Exp. 2) production. We suggest that these findings are consistent with a competitive learning mechanism that applies across stages and modalities of word production.

Keywords: Orthography; Phonology; Psycholinguistics; Semantics; Similarity.

Figures

Figure 1

Results of Experiment 1: The effect of distributed segmental overlap on response times for spoken picture naming of words in blocks with high segmental overlap (homogeneous) versus low segmental overlap (heterogeneous). Error bars represent the between-subjects standard errors of the means, corrected for repeated measures using the Cousineau (2005) method. Panel A includes data from all items. Panel B depicts the secondary analysis that includes only items that share the initial segment with at least half of the other items in the homogeneous block.

Figure 2

Results of Experiment 2: The effect of distributed segmental overlap on response times for written picture naming of words in blocks with high segmental overlap (homogeneous) versus low segmental overlap (heterogeneous). Error bars represent the between-subjects standard errors of the means, corrected for repeated measures using the Cousineau (2005) method. Panel A includes data from all items. Panel B depicts the secondary analysis that includes only items that share the initial segment with at least half of the other items in the homogeneous block.