Poster B59, Thursday, August 16, 3:05 – 4:50 pm, Room 2000AB
Picture naming in American Sign Language: an ERP study of the effects of iconicity and alignment
Meghan McGarry1, Megan Mott1, Katherine J. Midgley1, Phillip J. Holcomb1, Karen Emmorey1;1San Diego State University
Growing evidence suggests that iconicity may play a role in the recognition and/or production of signs in American Sign Language (ASL). Iconic signs have a motivated resemblance between their form and meaning, whereas non-iconic signs have a more arbitrary mapping between their form and meaning. The present study investigated the effects of iconicity and picture-alignment on sign production (see below for description of alignment). Specifically, we compared the production of iconic and non-iconic signs in a picture-naming task. In addition, for the iconic signs we compared an aligned picture condition in which a visually salient feature of the picture aligned with the iconic feature of the sign (e.g., the ASL sign BIRD depicts a bird’s beak and aligns with a picture of a bird with a prominent beak) and a non-aligned picture condition (e.g., a picture of a bird in flight where the beak is not visible). Deaf ASL signers named 176 pictures: 88 were named with non-iconic signs and 88 were named with iconic signs. Half of the 88 iconic signs were in the picture-aligned condition and half were in the non-aligned condition. The order of the pictures was counter-balanced across participants. Naming latencies did not differ for iconic and non-iconic signs, and picture-alignment had no effect on naming latencies for the iconic signs. EEG was also recorded, and Event-Related Potentials (ERPs) were time-locked to picture onset and averaged offline. We investigated whether iconicity and/or picture-alignment modulated the N400 component, a negative going component that peaks around 400ms after stimulus onset and is associated with both the processing of the form and the meaning of the stimulus. Our results show that the N400 amplitude for iconic signs was more negative than for non-iconic signs, with significant differences in mean amplitude beginning around 200ms and offsetting around 600ms. This finding suggests that the retrieval of iconic signs may activate additional perceptual or sensory-motor features compared to retrieval of non-iconic signs, resulting in increased negativity. The ERPs to iconic signs in the picture-aligned condition also showed reduced N400 negativity compared to iconic signs in the non-aligned condition, with significant differences beginning around 200ms and offsetting around 650ms. The reduced negativity for iconic signs in the aligned condition may constitute a unique priming effect that occurs when the visible features of an iconic sign and the to-be-named picture overlap. A right anterior scalp distribution was observed across the epoch for both the picture-aligned vs. nonaligned contrast and for contrast between iconic and non-iconic signs. Overall, the results indicate that retrieval of iconic signs in a picture-naming task involves more elaborate processing than retrieving non-iconic signs, resulting in increased negativity for iconic signs. This effect of iconicity may be akin to the increased N400 negativity observed for concrete words. In addition, the results indicate that the structural alignment between visual features of a to-be-named picture and an iconic sign facilitates lexical retrieval compared to when there is no visual feature overlap.
Topic Area: Signed Language and Gesture