Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
948222 | Journal of Experimental Social Psychology | 2012 | 7 Pages |
To investigate the effect of a social audience on learning-by-teaching, we examined participants' solutions of the 4-ring Tower of Hanoi problem after they demonstrated the 3-ring problem to a social agent (a person) or a non-social agent (a computer). In and participants produced less optimal solutions of the 4-ring problem after demonstrating the 3-ring problem to a social agent. An analysis of pointing behavior demonstrated that social highlighting contributed substantially to this effect. Together, these findings indicate that more social highlighting may produce a cost, rather than a benefit, on how deeply the demonstrator encodes the problem solution. Experiment 3 clarified that these results were not simply caused by the disruptions inherent to social highlighting. Taken together, the results suggest that social highlighting does not come for free — producing the highlighting may lead to more shallow encoding of demonstrated actions.
► We examine learning from demonstrating to a social vs. non-social agent. ► Participants exhibited worse learning after demonstrating to a social agent. ► More social pointing, but not more looking, was linked to decreased learning. ► This effect was specific to social pointing, not just the associated interruptions. ► Social reasoning can sometimes generate a cost rather than benefit to learning.