Automatic vs. Controlled Processing of Metaphors: An ERP Study

Student Author(s)

Nathan Axdorff

Faculty Mentor(s)

Dr. Gwenda Schmidt

Document Type

Poster

Event Date

4-12-2013

Abstract

The study examined automatic versus controlled processing of metaphor and divided visual field (DVF) presentation. Past research has shown that the right hemisphere processes broader semantic connections in language such as metaphors (Beeman, 2005). A negative-going event related potential (ERP) component 400 ms after the stimulus (N400) was presented was examined. When presenting certain figurative stimuli in either visual field the different hemispheres should have varying N400 peaks. We examined these components in sixteen right-handed, native English speaking undergraduates to compare results with a future study to be done with right hemisphere stroke patients and people with autism. We used a controlled stimulus set of 300 sentences that was divided into three figurativeness conditions (literal, metaphor and anomalous) (Cardillo et al., 2010) and counterbalanced to present the final word of each sentence to the left or right visual field. Stimulus onset asynchrony (SOA) was a between subjects factor that included long (1500 ms) and short (200 ms) SOA. Participants had to indicate after reading each sentence if they believed the sentence made sense. Electroencephalographic recordings at 16 electrode sites were used to assess the N400 amplitude, from 425 to 525 ms, post-stimulus which indicated the ERPs across the scalp in response to the presentation of the final word. Participants also completed the Autism Spectrum Quotient and the Systemizing Quotient-Revised (Baron-Cohen et al., 2001 & Wheelwright et al., 2006) for future comparison to the autistic subjects. A 3 (figurativeness: metaphor, literal, anomalous) x 2 (DVF: left, right) x 16 (electrodes sites) x 2 (SOA: long, short) mixed ANOVA revealed that short SOA elicited a larger N400 than long SOA which was marginally significant, F(1, 14)= 4.0, p= .065. Findings demonstrate some evidence that automatic processing (short SOA) elicits a larger N400 area under the curve than controlled processing (long SOA).

This document is currently not available here.

Share

COinS