Abstract
Recent theories of semantic memory have proposed that concepts are grounded in sensorimotor activity and mediated by the context from which the knowledge is drawn (Barsalou, 1999, 2003, 2008). Conceptual knowledge draws upon information from all modalities and therefore includes knowledge of associated object actions linked with both function and general movement (Bub, Masson, & Cree, 2008). The following experiment examined the conditions under which action information exerts an influence on experimental tasks particularly when taxonomic information is present. The experiment used a forced-choice triad task giving participants the choice of selecting between items that shared either a taxonomic or an action based relation with the target. The results showed that when the objects were presented as images on a white background (context-lean condition), participants were more likely to select the taxonomically related item. In contrast, when the same triads were presented as images being used in a functional scene (context-rich condition) they were more likely to select the action-related item. The results show that action knowledge is not automatic but is context-dependent. In line with views on embodied semantics, action-related information is drawn upon when objects are viewed and this influences task performance despite being unnecessary for the task.
Original language | English |
---|---|
Pages | 2925-2930 |
Number of pages | 6 |
Publication status | Published - 2014 |
Event | The 36th Annual Conference of the Cognitive Science Society - Quebec City, Canada Duration: 23 May 2014 → 5 Aug 2014 |
Conference
Conference | The 36th Annual Conference of the Cognitive Science Society |
---|---|
Abbreviated title | CogSci Conference |
Country/Territory | Canada |
City | Quebec City |
Period | 23/05/14 → 5/08/14 |