Spatial ability and handedness as potential predictors of space teleoperation performance

Abstract

NASA is concerned with finding performance predictors for space teleoperation tasks in order to improve training efficiency. Experiment 1 determined whether scores on tests of spatial skills could predict performance when selecting camera views for a simulated teleoperation task. The hypothesis was that subjects with high spatial ability would perform camera selection tasks more quickly and accurately than those with lower spatial ability, as measured by the Mental Rotation Test (MRT), Purdue Spatial Visualization Test (PSVT), and the Perspective Taking Ability (PTA) test. Performance was defined by task time, number of correct camera selections, preparation time, number of camera changes, and correct identification of clearance issues. Mixed regression and nonparametric tests showed that high-scoring subjects on the MRT and PTA spatial ability tests had significantly lower task times, higher camera selection scores, and fewer camera changes than subjects with lower scores, while High PSVT scorers had significantly lower preparation times. Experiment 2 determined whether spatial ability, joystick configuration, and handedness influenced performance of telerobotic fly-to tasks in a virtual ISS environment. 11 righthanded and 9 left-handed subjects completed 48 total trials, split between two hand controller configurations. Performance was defined by task time, percentage of translational and rotational multi-axis movement, percentage of bimanual movement, and number of discrete movements. High scorers for the MRT, PSVT, and PTA tests had lower Task Times, and High PSVT and PTA scorers made fewer Discrete Movements than Low scorers. High MRT and PTA scorers had a higher percentage of translational and rotational multi-axis movement, and High MRT scorers had a higher percentage of bimanual movement. The overall learning effect appears to be greater than the effect of switching between hand controller configurations. No significant effect of handedness was found. These results indicate that these spatial ability tests could predict performance on space teleoperation tasks, at least in the early phases of training. This research was supported by the National Space Biomedical Research Institute through NASA NCC 9- 58.