Analyzing Action for Agents with Varying Cognitive Capacities

Abstract

in this report, we investigated participants' beliefs about how different agents represent action. in Experiment 1, participants divided actions into units for two hypothetical observers. Participants marked fewer units for a person than for either of two different machine agents, suggesting that motions were integrated into larger goal-based units for the human but not for the machines. An analysis of alignment between participant breakpoints and coded action features demonstrated that participants selected larger units aligned with changes in actors' goals when segmenting for humans, and that they selected smaller units aligned with actors' motions when segmenting for machines. in Experiment 2, one group of participants was presented a robot imbued with a minimal understanding of the object-directed and experience-dependent character of human action. Segmentation for this machine agent was similar to segmentation for a human, suggesting that human capacities can override a category-based distinction between humans' and machines' representations of action.

Digital Games, Design, and Learning

“…The relevance of camera viewpoint for learning was included because of the numerous reports that have shown that individuals who play first-person perspective “shooter” games, but not other games, demonstrate improvement on certain visual cognitive tasks (e.g., Feng, Spence, & Pratt, 2007 ; Green & Bavelier, 2006 , 2007 ). Anthropomorphism was included because of numerous findings suggesting that anthropomorphic attributes affect a range of perceptual, cognitive, and social tasks (e.g., Heider & Simmel, 1944 ; Killingsworth, Levin, & Saylor, 2011 ; Mahajan & Woodward, 2009 ).…”