The structure of lexical semantic representations as revealed by semantic verification

Poster Session B - Sunday, April 14, 2024, 8:00 – 10:00 am EDT, Sheraton Hall ABC

Anna Keresztesy¹, Jessica Smith¹, Eliza Reedy², Bradford Mahon¹; ¹Carnegie Mellon University, ²University of Pittsburgh

A long-standing issue is whether conceptual representations are organized in the brain by semantic category, sensorimotor modality of knowledge, or jointly by both dimensions. Semantic verification, which is widely used to study conceptual representations, requires participants to decide the validity of short descriptive attribute-object pairs (e.g. “a wallet holds money”, “a wallet is used for cleaning”). We report a series of experiments using a new semantic verification task with four matched categories (animals, fruits/vegetables, tools, and manipulatable objects) orthogonally crossed with six semantic attributes (function, behavior, visual surface, visual form, typical location, thematic association). In Experiment 1, participants were presented with an attribute, followed by an object name, and indicated (button push) whether the attribute was true/false of the object. In Experiment 2, with identical stimuli and timing of presentation, participants responded verbally by reading the object name aloud if the presented attribute was true, or saying ‘false’ otherwise. Experiment 3 normed all attributes and object names for reading time, allowing us to partial out trial-level variance in processing time in Experiments 1 and 2. We found that (putatively) preferred attribute types were not associated with faster verification times for their respective categories (e.g., function attributes for tool, surface/color attributes for fruit/vegetables), for either manual or verbal responses, after partialling out processing times for attributes and objects. These findings suggest that neither category nor modality holds priority in verification judgements, contrary to the long-proposed alignment between knowledge types and categories with this task.

Topic Area: LANGUAGE: Semantic