Abstract
Human vision can be divided into vision for perception and vision for action (Milner & Goodale, 1995). In vision for perception the features of an object are processed independently (Kyllingsbaek & Bundesen, 2007). Are features also processed independently in vision for action? 20 different white bars were created using 4 lengths (5-7 cm.) and 5 orientations (30°- 60°). They were presented at arms length for 40 ms. on a black computer screen one at a time on either the left or right side of a central fixation mark. A contrast adjustment procedure lowered the intensity of the bar until barely visible. Vision for perception and vision for action were evaluated in separate blocks by two types of response modes for specifying length and orientation of the bars: Symbolic specification: specifying length and orientation on a keyboard. Motoric specification: moving index and thumb to the endpoints of the bar (using about 500 ms.). Positions of the fingers were measured and specification of raw lengths and orientations calculated. The raw lengths and orientations were then categorized into 4 lengths and 5 orientations. 7 subjects completed 320 trials of both types of specifications in separate blocks. Correct/incorrect symbolic or motoric specifications of length and orientation were entered into a 2x2 chi2. This was used as a measure of independence. In all subjects the chi2 analysis of both types of specifications supported independent processing of the two features. Also, Monte Carlo simulations of specifications of the two features (based on the observed marginal probabilities of correct specification of each feature) showed no significant differences between the observed specifications and the corresponding predictions of the model assuming independent processing of features. The features of an object are processed independently in both "vision for perception" and "vision for action".
Meeting abstract presented at VSS 2012