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

Poster E42, Friday, November 10, 10:00 – 11:15 am, Harborview and Loch Raven Ballrooms

Different mechanisms for lexical ambiguity resolution in individuals with ASD?

Emily Coderre1, Mariya Chernenok2, Trevor Brothers2, Barry Gordon1, Kerry Ledoux1;1Johns Hopkins University, 2University of California, Davis

Autism spectrum disorder (ASD) is characterized by language deficits, including difficulties determining the meaning of ambiguous words like homographs. The “weak-central coherence” theory attributes these difficulties to a tendency toward enhanced local processing at the expense of global integration. We investigated local and global processing during lexical ambiguity resolution in individuals with ASD using the N400 event-related potential. Sentences (adapted from Sitnikova et al., 2002) consisted of an initial clause biasing one meaning of a homograph. In a second clause, a critical word was congruent with the homograph’s dominant meaning but was either congruent or incongruent with the biased meaning. A sentence-final control word was congruent or incongruent with both local and global context to provide a baseline for N400 effects. A local processing bias would predict no N400 effect at critical words, since they always match the homograph’s dominant meaning. A global processing bias would predict enhanced N400 amplitudes when critical words are incongruent vs. congruent with the homograph’s biased meaning (i.e. an N400 effect). Weak central coherence would predict greater global processing in the TD group (an N400 effect for critical words) but greater local processing in the ASD group (no N400 effect). Both groups should show N400 effects at control words. Participants were 20 adults with ASD (ages 18-68, M = 34; 16 male) and 20 age- and sex-matched TD adults. Groups were matched on age and IQ (all p’s>0.24). Receptive vocabulary was lower in the ASD group (p<0.05) and was included as a covariate in analyses. Participants read 120 sentences for comprehension, adapted from Sitnikova et al. (2002), responding to relatedness probes after each sentence. Sentences were presented in rapid serial visual presentation format during concurrent EEG recording. Repeated-measure ANOVAs were conducted at nine scalp regions (representing frontal, central, and parietal sites and left, midline, and right lateralities) with factors of congruity, site, laterality, and group. At the control word, both groups showed centro-parietal N400 effects from 300-500 ms (all p’s<0.05). N400 effect magnitude was comparable between groups (all p’s>0.16), suggesting that participants successfully detected semantic anomalies. At the critical word, neither group showed N400 effects (all p’s>0.17), suggesting a lack of global processing for both groups. This finding contradicts our predictions and the results of the TD group in Sitnikova et al. (2002). This discrepancy could be due to age differences (our participants had a wider age range) or working memory differences (although neither study assessed working memory). Although neither group showed congruency effects, the TD group had larger N400 amplitudes overall compared to the ASD group (F(1,37)=6.33, p<0.05). Given the frequent anomalies in the stimuli, TD participants may have adopted a local processing strategy and/or kept both meanings of the homograph in mind to later evaluate congruity. As the ASD group showed smaller N400 responses overall, they may have processed semantic ambiguity in a fundamentally different way.

Topic Area: Meaning: Lexical Semantics

Back to Poster Schedule