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

Poster B2, Wednesday, November 8, 3:00 – 4:15 pm, Harborview and Loch Raven Ballrooms

Sentence comprehension under conflict in aphasia

Malathi Thothathiri1, Edward Wlotko2;1The George Washington University, 2Moss Rehabilitation Research Institute

Sentence comprehension involves the integration of multiple (e.g., semantic and syntactic) cues. When these cues conflict, sentence interpretation may place increased demands on comprehension-relevant processes. We hypothesized that comprehension under conflict could rely on one or more of: (1) robust syntactic comprehension that can override other misleading cues; (2) short-term memory (STM) to store a sentence for re-parsing; and (3) executive function (EF) to resolve competition between interpretations. We conducted contrastive case studies of aphasic patients with different deficit combinations to elucidate the contribution of these processes to comprehension under conflict. Five patients with aphasia completed a battery of comprehension, STM, and EF tasks more than a year after their stroke (ages 53-72; 2 Male, 3 Female). Syntactic comprehension was assessed using a reversible-sentence-to-picture-matching task. For STM, we used rhyme probe, category probe, digit and word spans. For EF, we used the Stroop and Flanker tasks. We investigated how patients’ performance in these tasks related to performance on the critical conflict sentence comprehension task. In this task, sentences contained either congruent or incongruent syntactic and semantic cues (Congruent: The rabbit was chased by the fox; Incongruent: The fox was chased by the rabbit). In the incongruent case, syntactic structure led to an interpretation that conflicted with semantic expectations (foxes usually chase rabbits rather than vice versa). Participants were asked to match the sentence (e.g., The fox was chased by the rabbit) to one of 4 pictures, which included the target picture (rabbit chasing fox), the reversed picture (fox chasing rabbit), and 2 unrelated distractors (chance assuming intact lexical comprehension=50%). One patient (S3) performed at chance in reversible sentence comprehension (53.3%, chance=50%), suggesting an inability to use syntactic cues for assigning thematic roles. A second patient (S2) was 60% accurate in reversible sentence comprehension and had clear STM deficits (Rhyme: 1.29, Category: 3, Digit: 2, Word: 1.33). In the syntax-semantics conflict task, both patients overrelied on semantic cues, leading to worse performance on conflict (S3=42.5%; S2=47.5%) than non-conflict sentences (S3=77.5%; S2=85%). Thus, impairment in syntactic comprehension and/or STM was associated with inaccurate sentence comprehension under conflict. The remaining three patients showed relatively more intact reversible sentence comprehension (S1=100%; S4=80%; S5=67%) and STM (e.g., Word span: S1=3.41; S4=3.01; S5=3.71). These patients were also relatively more accurate in the syntax-semantics conflict task (conflict vs non-conflict sentences: S1=97.5% vs 95%; S4=67.5% vs 87.5%; S5=75% vs 80%). All three patients showed exaggerated reaction time (RT) costs in EF tasks (e.g., Stroop: S1=46%; S4=62.8%; S5=88.5%), suggesting decreased efficiency in resolving conflict. These patients also showed high RT costs for conflict relative to no-conflict sentences (S1=12.2%; S4=34.4%; S5=15%. Compared to S2=2.6%; S3=-29.8%). Taken together, our results suggest that (1) sentence comprehension under conflict requires intact syntactic comprehension, as expected; (2) STM could be important for scaffolding sentence comprehension, especially under conflict; and (3) inefficient conflict resolution during EF tasks may correspond to inefficient but not necessarily inaccurate comprehension under conflict. These findings highlight the differential roles of processes needed for accurate and efficient sentence comprehension.

Topic Area: Control, Selection, and Executive Processes

Back to Poster Schedule