Targeted cognitive training of spatial skills: Perspective taking in robot teleoperation
Spatial skills are critical for robot teleoperation. For example, in order to make a judgment of relative direction when operating a robot remotely, one must take different perspectives and make decisions based on available spatial information. Training spatial skills is thus critical for robot teleoperation, yet, current training programs focus primarily on psycho-motoric skills of the task, and less on the essential cognitive aspects of spatial skills. This work addresses this need by considering previous findings on relative direction judgments in training robot teleoperation. We developed and tested a basic training paradigm of perspective taking skill targeting the cognitive skill rather than psycho-motoric skill. An experiment tested a basic training paradigm using a stationary robot, with a training group receiving perspective taking training and a control group without training, and both tested on a transfer test with the robot. The results show that participants who went through a targeted cognitive skill training reached mastery level during the training, and performed better than the control group in an analogue transfer of learning test. Moreover, results reveal that the training facilitated participants with initial poor perspective taking skills reach the level of the high-skilled participants in transfer test performance. The study validates the possibility to target only cognitive aspects of spatial skills and result in better robot teleoperation.
|Cognitive Training, Spatial Information Processing, Targeted Training, Teleoperation, Training Spatial Skills Visual Perspective Taking|
|13th International Conference on Spatial Information Theory, COSIT 2017|
|Organisation||Department of Psychology|
Luko, L. (Liel), & Parush, A. (2017). Targeted cognitive training of spatial skills: Perspective taking in robot teleoperation. In Leibniz International Proceedings in Informatics, LIPIcs. doi:10.4230/LIPIcs.COSIT.2017.12