Abstract
Gibson (1979) conjectured that perception of affordances involves detecting meaningful possibilities for action. Is the meaning obtained when an affordance is perceived qualitatively different from other types of semantic knowledge? Pilot investigations in our lab have discovered that affordance primes are processed slower than semantic features and non-associates in a linguistic semantic-categorization task that presented words on a computer screen. The slower processing of affordance primes might be due to the fact that affordances are typically encountered through our senses, and not as linguistic information. Chainay and Humphreys (2002) found that action knowledge was processed faster when objects were presented as pictures rather than words. Sensory information (pictures over words) may therefore be more relevant to action. For the present study, we hypothesized that pictorial depictions of objects might be better suited for facilitating affordance-based priming than linguistic information such as reading words on a computer screen. We investigated the effects of affordance priming using a spatial categorization task. 81 object nouns were compiled from the McRae et al. (2005) norms. We used photographs of objects drawn from the database compiled by as visual stimuli (Brodeur, Dionne-Dostie, Montreuil, & Lepage, 2010). Affordances denoted possibilities for action in relation to objects (e.g. sit – chair), whereas semantic features indicated definitional characteristics (e.g. has legs – chair). Participants were presented with a prime and asked to respond by indicating whether the presented target object could fit inside of a shoebox (Bowers & Turner, 2003). We manipulated image quality at three levels of blur to assess differential effects of the fidelity of visual information. Results showed that blurry images were processed the slowest. Consistent with our hypothesis, affordances were processed faster than control stimuli across all levels of image quality, suggesting that the availability of affordance versus feature information may facilitate processing of visual objects.