From action to abstraction: The sensorimotor grounding of metaphor comprehension in Parkinson’s disease.
Stacey Humphries1, Nate Klooster1, Eileen Cardillo1, Anjan Chatterjee1; 1University of Pennsylvania
Can embodied cognition theories account for the ability to represent abstract ideas? Metaphors are a linguistic vehicle through which we understand the abstract. Proponents of embodied cognition argue that metaphor comprehension involves simulating the literal sense of the word being used metaphorically. However, evidence for this claim is mixed: some studies find activation of sensorimotor regions when people comprehend metaphors, whilst other studies observe sensorimotor activation for literal but not metaphorical sentences. This discordance may be caused by methodological differences between studies in the novelty of the metaphors and control conditions used, and a lack of control over psycholinguistic features of the stimuli. We designed a highly controlled Registered Report study (peer-reviewed and accepted in principle by Cortex prior to data collection) in patients with motor disturbance (Parkinson’s disease) on an extensively normed set of action verb and sound verb metaphors. Embodied theories predict that PD patients will be impaired on action metaphors but not sound metaphors. Metaphor comprehension is assessed with a semantic meaning-selection task. Bayesian inferential methods are used to permit optional stopping and to determine whether the null or alternative hypothesis is more strongly supported by the data. Current data from 9 patients shows that responses to action metaphors are less accurate than those to sound metaphors (87.03% vs. 93.7%), and are slower (7241ms vs. 6837ms). Our study offers an example of a theoretically motivated, methodologically rigorous, registered study accepted provisionally before final results are obtained, in attempt to mitigate pervasive false positive reports in the literature.
Topic Area: LANGUAGE: Semantic