Documents 

The primary goal of this dissertation was to understand to what extent are humans capable of accurately perceiving both their own rotations and the rotations of others and how do spatial manipulations of the actor and the perceiver affect the perception of the action. Four experiments were conducted. Experiment 1 examined the perceptual abilities of both expert gymnasts and novices in the accurate judgment of either a left or right rotation by an animated human figure (avatar) in a series of fixed picture plane orientations. Participants responded with either a verbal “left” or “right” and their mean accuracy and mean response time were recorded. Experiment 2 used the same stimulus but the participants were asked to report their answers by way of one of four arrow button combinations; right button corresponded to a right twist, left button corresponded to a right twist, up button corresponded to a right twist and down button corresponded to a right twist. Both mean accuracy and response time were recorded for each group. Experiment 3 mirrored Experiment 1 except the participants were randomly fixed in one of six picture plane orientations from 0° to 300° of rotation. Experiment 4 used the same apparatus as Experiment 3 where the participants were randomly placed in one of six picture plane orientations however the participants were rotated either left or right and were asked to verbally report their own twisting direction. Experiment 4 tested both novices and expert gymnasts. The general results suggest that the task of accurately determining the twisting direction of another human form is challenging and cognitively demanding. Under most conditions accuracy decreased and response time increased as the phase angle between the participant and the avatar approached 180°. In Experiment 1 experts and novices performed the same and were least accurate and took the longest to respond when the avatar was inverted. The results from Experiment 2 suggested a conflict of strategies between the constraints of the task and the inherent challenge of the task. Participants in Experiment 3 were generally uninfluenced by their own picture plane orientation and in almost all combinations of participant and avatar picture plane orientation they were more accurate and faster than the exclusively upright participants in
Experiment 1. The experts in Experiment 4 were flawless in their responses and significantly faster than in any other experiment. The novices were also relatively fast but the accuracy of the judgments on their own twisting directions was no better than the participants watching the avatar in Experiment 1. The findings suggest that the general task of determining human rotation as either a left or right turn is so challenging that a number of conflicting strategies may have been employed by the participants to help lessen the cognitive demands of the task. The data speaks to the specificity of expertise and outlines a potential discrepancy between expert observers and expert performers. The data suggests that the use of an internal reference frame during the spatial perception of biological motion may be consistent across conditions regardless of participant orientation. [editor summary]

The relationship between point-of-gaze and ball movements in three-ball juggling was examined as a function of expertise, pattern and tempo. Five intermediately skilled and five expert jugglers performed the standard and reverse cascade at three tempos, while point-of-gaze and ball movements were recorded simultaneously. Scaled to the size of the ball patterns, the experts made smaller point-of-gaze movements than the intermediates, especially in the horizontal direction and in the standard cascade. In both skill groups, point-of-gaze and ball movements were often 1 : 1 frequency locked in the horizontal direction, whereas in the vertical direction 1 : 2 frequency locking also occurred. In the latter direction, the 1 : 1 ratio prevailed in the intermediates and the 1 : 2 ratio in the experts. In addition, the incidence of the 1 : 1 ratio decreased and that of the 1 : 2 ratio increased with increasing tempo. Furthermore, in the vertical direction, increasing tempo resulted in a weaker 1 : 1 locking, whereas the strength of the 1 : 2 ratio remained unaffected by tempo. In the horizontal direction, the strength of the 1 : 1 locking was higher on average in the reverse cascade than in the standard cascade. We conclude that expertise in juggling is reflected by an overall reduction in the extent to which the balls are visually tracked, and that task constraints such as tempo and juggling pattern affect the visual search patterns of both expert and intermediate jugglers. [authors summary]