Repository for a Unity implementation of the "skittles" virtual throwing task originally proposed in (Müller & Sternad, 2004), and more recently used in Zhang et al. (2018). This environment is designed to be used with a computer mouse to facilitate remote data collection or prototyping.
This repository is associated with a published abstract submitted to COGSCI2022. Please use the following citation when referencing this repository:
Nalepka, P., Schell, G., Patil, G., & Richardson, M. J. (2022). A computer mouse-based throwing task to study perceptual-motor skill learning in humans and machines [Abstract]. Proceedings of the Annual Meeting of the Cognitive Science Society, 44.
Perceptual-motor tasks offer redundant solutions to achieve a goal. However, not all solutions are equally robust to error-producing noise or variability and thus, skill learning can be viewed as a search process to identify behaviors that are error-tolerant. Throwing a ball to hit a target is one such example of a complex perceptual-motor skill that has been studied in the laboratory via the virtual “skittles” task, a simplified 2D task involving throwing a tetherball around a pole to hit a target. We implemented the task as a Unity3D environment (code here: https://github.com/ShortFox/SkittlesTaskEnvironment/) which enables participants to complete the task with a computer mouse and replicated key findings from previous research. Our implementation allows for remote data collection and can serve as a pedagogical tool to teach concepts in skill acquisition. Future work will use this task to explore human versus machine skill acquisition by leveraging Unity’s MLAgents reinforcement learning package.
The task environment can be found here. This project was tested using Unity 2019.4.4f.
The task can be played online using Unity's WebGL here.
The solution manifold for the task environment is pictured below:
Note: 0° release angle equates to the 12 o'clock position (90° equates to the 3 o'clock position, etc.). Positive angular velocity represents clockwise movement.
- Müller, H., & Sternad, D. (2004). Decomposition of Variability in the Execution of Goal-Oriented Tasks: Three Components of Skill Improvement. Journal of Experimental Psychology: Human Perception and Performance, 30(1), 212–233. https://doi.org/10.1037/0096-1523.30.1.212
- Zhang, Z., Guo, D., Huber, M. E., Park, S.-W., & Sternad, D. (2018). Exploiting the geometry of the solution space to reduce sensitivity to neuromotor noise. PLOS Computational Biology, 14(2), e1006013. https://doi.org/10.1371/journal.pcbi.1006013
If you have any questions or would like to discuss this research, please contact Dr. Patrick Nalepka (ShortFox) at patrick.nalepka@mq.edu.au.