Purchase this article with an account.
Matthew Keough, Ada Le, Jun Li, Matthias Niemeier; Differences between action and perception in learning object categories. Journal of Vision 2009;9(8):1111. doi: 10.1167/9.8.1111.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Understanding is grasping in many languages. The universality of this metaphor suggests that humans rely on representations of body movements when they think about concepts or categories. Given this, learning to sort objects into categories might improve if the objects are grasped. But the neural control of visually guided grasps involves areas in the parietal cortex, whereas areas in the occipito-temporal cortex are associated with recognizing objects and assigning them to categories. Therefore, learning object categories might be impaired if objects are grasped. As a third possibility, learning categories might involve ‘amodal’ neural mechanisms remote from perception and motor control. So then learning with or without grasps should not differ. To test these predictions, we asked participants to learn to sort objects into categories depending in the combination of two object chracteristics. People learned less well when they grasped the objects in different ways to indicate categories than when they pressed buttons, even when we varied button locations to manipulate processing load for button reaches. These results suggest that the language intuition is false, as sensoriotor control of grasps hinders category learning. It suggests that the functional specialization of action and perception permeates high-level functions of human concept knowledge.
This PDF is available to Subscribers Only