The purpose of this thesis was to explore whether there is a direct route to action from vision, in which the selection of action is constrained by perceptual information from objects as well as by more abstract semantic knowledge. A connectionist model, NAM (Naming and Action Model) was developed, incorporating a direct perceptual route to action along with an indirect semantic route. This was used to simulate data from normal observers and neuropsychological patients. In addition to this, new empirical data were collected to provide a better understanding of the types of visual information used in action selection. One set of new experiments provided evidence that two variables, object viewpoint and semantic priming, differentially affect action and semantic decisions to objects. A further set of experiments showed that action verification tasks (e. g., is this drinking? ) could be facilitated by congruent information about handgrip - provided either by a static or a moving hand. This indicates that a direct route to action is sensitive to linked information about handgrip and movement. Finally, the direct route to action was examined through a case study of an optic aphasic patient. This patient showed spared naming of verbs relative to objects, but the opposite pattern of sparing occurred in reading. This double dissociation, within a patient, suggests that verb retrieval for objects can operate non-semantically. These computational, psychological and neuropsychological data provide evidence for a direct route to action.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:403462 |
| Date | January 2003 |
| Creators | Yoon, Eun Young |
| Publisher | University of Birmingham |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0091 seconds