Are two heads better than one? Developing a model for predicting group benefits in joint visuospatial tasks
Final Report Abstract
In daily life, humans often perform visual tasks together such as, for instance, solving puzzles or looking for a misplaced key. In such tasks, humans tend to distribute the labor, enabling them to reach a higher performance compared to performing the same task alone – a group benefit. In this project, I extended earlier findings on group benefits in several ways. I devised a model, which accurately can predict group benefits. Moreover, I found that humans have preferences on how they distribute the labor and that theses preferences depend on the task environment. I also found that humans are willing to distribute the labor with an artificial agent but only if that agent is behaving in a human-like way and importantly, is also described to be behaving in a human-like way prior to any interactions. While earlier research primarily investigated dyadic cooperative tasks, I also investigated how labor divisions come about in larger groups (i.e., triads) and found that increasing the group size still leads to group benefits but also additional accuracy costs. I also investigated physiological correlates of group benefits using pupil size measurements and found that the coordination effort required for devising labor divisions correlates with changes in pupil sizes, suggesting that pupil sizes could be used as a measure of coordination effort. Finally, while all of the above studies used visual tasks, I also investigated the benefits of labor divisions in multisensory tasks, in which each participant in a dyad responds to stimuli from a different sensory modality. For such tasks, labor divisions are only beneficial for spatial localization tasks but not for motion, counting, or temporal tasks. Taken together, I find a variety of factors influencing group performance and that group benefits can be accurately predicted using statistical modeling. Given that humans often collaboratively distribute task demands in several professions, these findings could be used to make labor divisions more efficient, assemble more effective teams, and generally reduce the risk of errors.
Publications
- (2019). Neurophysiological correlates of collective perceptual decision-making. European Journal of Neuroscience
Baumgart, K. G., Byvshev, P., Sliby, A. N., Strube, A., König, P., & Wahn, B.
(See online at https://doi.org/10.1111/ejn.14545) - (2020) Audiovisual integration during joint action: No effects for motion discrimination and temporal order judgment tasks. Frontiers in Psychology. 11, 79
Wahn, B., Dosso, J., & Kingstone, A.
(See online at https://doi.org/10.3389/fpsyg.2020.00079) - (2020). Dyadic and triadic search: Benefits, costs, and predictors of group performance. Attention, Perception, & Psychophysics. 82, 2415–2433
Wahn, B., Czeszumski, A., Labusch, M., Kingstone, A., & König, P.
(See online at https://doi.org/10.3758/s13414-019-01915-0) - (2020). Labor division in joint tasks: Humans maximize use of their individual attentional capacities. Attention, Perception, & Psychophysics. 82, 3085 – 3095
Wahn, B. & Kingstone, A.
(See online at https://doi.org/10.3758/s13414-020-02012-3) - (2020). Performing a task jointly enhances the sound-induced flash illusion. [epub ahead of print] Quarterly Journal of Experimental Psychology
Wahn, B., Rohe, T., Gearhart, A., Kingstone, A. & Sinnett, S.
(See online at https://doi.org/10.1177%2F1747021820942687)
