You are viewing the SNL 2017 Archive Website. For the latest information, see the Current Website.

Poster A8, Wednesday, November 8, 10:30 – 11:45 am, Harborview and Loch Raven Ballrooms

Tracking the time course of associative and categorical context effects in spoken word production

Andus Wing-Kuen Wong1, Ho-Ching Chiu1, Jie Wang2, Siu-San Wong1, Jinlu Cao2, Hsuan-Chih Chen2;1City University of Hong Kong, 2Chinese University of Hong Kong

A key issue in the production literature concerns the mechanism of lexical selection in speech production. The Lexical Selection by Competition (LSC) view assumes that lexical selection involves competition among multiple co-activated lexical units. Supporting evidence comes from picture-word interference (PWI) studies. In a PWI task, participants are asked to name individually presented picture and ignore a word distractor. Participants’ naming latencies were slower if the target (DOG) and distractor (cat) were from the same semantic category, relative to an unrelated control (pen). This categorical interference effect has been taken to support the LSC view as only the categorically related distractor, but not the unrelated distractor, would intensify the competition during lexical selection. Dissimilarly, the Response Exclusion Hypothesis (REH) assumes a late locus of the categorical interference effect. According to REH, longer time is required to exclude a categorially related response from the response buffer due to its response relevance. Furthermore, null or facilitation effects have been reported if the distractor was associatively related to the target (bone and DOG). The REH assumes that the associative effect arises from an early conceptual stage and the categorical interference from a late post-lexical stage. Conversely, the LSC view assumes a lexical locus for both associative and categorical effects. To verify the above two accounts, this study was conducted using the PWI paradigm with concurrent ERP recording. Thirty-four native Mandarin speakers participated. Each picture stimulus was paired with three word distractors, namely categorically related, associatively related, or phonologically related. Three corresponding unrelated conditions were constructed by re-pairing the targets and distractors in each condition. Participants’ naming latencies were submitted to linear mixed-effect modeling. The differences in naming latencies between related and unrelated conditions were significant for categorical distractors (b = -45.62, SE = 14.13, p = 0.002), marginally significant for phonological distractors (b = 21.72, SE = 12.68, p = 0.092), and not significant for associative distractors (b = -11.12, SE = 11.97, p = 0.357). The ERP signals were analyzed using a stimulus-locked approach (i.e., time-locked to the picture onset) and a response-locked approach (i.e., time-locked to the response onset). For the stimulus-locked analysis, mean ERP amplitude values were calculated in the time windows of 0-275, 275-450, and 450-600 ms post-target. For the response-locked analysis, mean ERP amplitudes were calculated within the 300 ms to 150 ms time window prior to the response onset and within the 150-ms time window preceding the response onset. Significant ERP effects were obtained in categorical and associative conditions in the time window between 275 and 450 ms post-target. Furthermore, significant ERP effects of target-distractor relatedness were observed only in the associative condition within the 300-ms pre-response period. The classical categorical interference was replicated in the present study. By combining stimulus-locked and response-locked approaches, this study showed relatively early ERP effects in categorical and associative conditions and late ERP effects only in the associative condition. These results are consistent with the LSC view that the categorical interference effect is due to competition during lexical selection.

Topic Area: Control, Selection, and Executive Processes

Back to Poster Schedule