Describing and analyzing error for one-dimensional performance tasks is fairly straightforward, but suggestions for describing and analyzing error for two-dimensional performance tasks (e.g., marksmanship) are quite problematic. Specifically, imposing an arbitrary axis onto the two-dimensional work space, along which traditional one-dimensional measures can be computed and analyzed, yields measures of accuracy, bias, and consistency that are entirely dependent upon the choice of axis. The present contribution offers new measures and methods for describing and analyzing data from two-dimensional performances. Unlike the resu1ts from previous suggestions, the approaches described herein yield results that are completely independent of the axes used to quantify the individual two-dimensional trials. These new approaches are strongly related to well-established methods for describing and analyzing error for one-dimensional tasks.

The end-state comfort effect has been observed in recent studies of grip selection in adults. The present study investigated whether young children also exhibit sensitivity to end-state comfort. The task was to pick up an overturned cup from a table, turn the cup right side up, and pour water into it. Two age groups (N = 20 per group) were studied: preschool children (2-3 years old), and kindergarten students (5-6 years old). Each child performed three videotaped trials of the task. Only 11 of the 40 children exhibited the end-state comfort effect, and there were no differences between age groups. Results revealed the emergence of five different performance patterns, none of which were consistent with sensitivity to end-state comfort. The findings have implications for the advance planning of manual control in young children.

The prediction emanating from memory drum theory (Henry & Rogers, 1960') that simple reaction time (SRT) increases as a response becomes more complex (i.e., increases in number of movement parts) was investigated. Experiments 1 (N = 20) and 3 (N = 16) indicated that SRT was longer for responses consisting of two and three parts than it was for a one-part response and this may be interpreted as support for the prediction. Failing to support the prediction, however, was the finding that SRT was essentially the same for responses consisting of two and three parts. This may not be too damaging to the theory because it could simply be reflecting an upper limit in terms of numbers of parts or response duration for causing an increase in SRT. Experiments 2 (N = 20) and 3 revealed an SRT effect between two responses that were supposed to be equal in complexity. At first, this finding appeared to be contrary to the prediction, but it may be interpreted as support for it because one of the responses defined as having one movement part could actually have had two

The question of whether changes seen in simple reaction time (SRT) as a function of response complexity (i.e., number of movement parts) should be considered as differences in the time needed to centrally program a motor response was addressed. Using a large-scale tapping response, 14 subjects contacted from one to five targets positioned in a straight line, while a second group of 14 subjects executed 90 degrees changes in direction in striking the targets. Results revealed that mean SRT and mean premotor time increased linearly as the number of movement parts increased, regardless of whether changes in movement direction had to be programmed, with the greatest increase occurring between one-, and two-part responses. Increases in motor time were not sufficient to account for the sizeable SRT effect. These findings support the position of increased central programming time for more complex responses, and also help establish some of the boundaries of the complexity effect.

No abstract

Two experiments examined the interaction of vision and articular proprioception in simple one-hand catching. In Experiment 1 (N = 18) skilled baseball and softball players used the left and right hands to catch slowly moving tennis balls, while Experiment 2 (N = 16) used novice catchers as subjects. In half the trials, sight of the catching hand was prevented by placing a screen alongside the subjects' face. Results of Experiment 1 revealed that the screen caused minimal disruption of the positioning phase of the catch, with moderate disruption of the grasping phase. However, for the unskilled subjects of Experiment 2, the screen caused considerable disruption of positioning. The data provide only minimal support for Smyth and Marriott' (1982) contention that limb position is inadequately specified by articular proprioception. It is argued that skill level serves as a mediator in the ability to use proprioception for limb positioning, but vision appears necessary to control the precise temporal organization of the grasp phase of one-hand catching.

During a unimanual grip selection task in which people pick up a lightweight dowel and place one end against targets at variable heights, the choice of hand grip (overhand vs. underhand) typically depends on the perception of how comfortable the arm will be at the end of the movement: an end-state comfort effect. The two experiments reported here extend this work to bimanual tasks. In each experiment, 26 right-handed participants used their left and right hands to simultaneously pick up two wooden dowels and place either the right or left end against a series of 14 targets ranging from 14 to 210 cm above the floor. These tasks were performed in systematic ascending and descending orders in Experiment 1 and in random order in Expiment 2. Results were generally consistent with predictions of end-state comfort in that, for the extreme highest and lowest targets, participants tended to select opposite grips with each hand. Taken together, our findings are consistent with the concept of constraint hierarchies within a posture-based motion-planning model.

Several features of the actual movement pathway in two rapid target-striking tasks were quantified by using high-speed cinematography, and whether the movement pathway is constrained as a function of the accuracy demands imposed by the size of the subtended angle was determined. Subjects (N = 16) first hit an 8-cm-diameter target located 10 cm to the left of a start position and then, depending on the condition, moved another 10 cm to hit either a 6-cm- or 1.5-cm-diameter target. Subtended angles were 17.1 and 4.3 degrees for the large and small second-target conditions, respectively. Fifty trials per condition were performed, the last 3 of which were filmed at 120 Hz. The vertical dimension of movement (peak height along the z-axis) was captured directly from the camera view, whereas the horizontal (y-axis) dimension, that is, the dimension orthogonal to the principal direction of motion, was captured through a mirror positioned above the target board. Reaction times and movement times were significantly longer in the small second-target condition, thus replicating the well-known response complexity effect. Kinematic analyses revealed that when the subtended angle was smaller, there was significantly less horizontal pathway deviation as well as significantly higher peak vertical displacement in the movement. Therefore, the accuracy demands imposed by a smaller subtended angle do constrain the actual movement pathway.