Audience‐Contingent Variation in Action Demonstrations for Humans and Computers

Abstract

People may exhibit two kinds of modifications when demonstrating action for others: modifications to facilitate bottom-up, or sensory-based processing; and modifications to facilitate top-down, or knowledge-based processing. The current study examined actors' production of such modifications in action demonstrations for audiences that differed in their capacity for intentional reasoning. Actors' demonstrations of complex actions for a non-anthropomorphic computer system and for people (adult and toddler) were compared. Evidence was found for greater highlighting of top-down modifications in the demonstrations for the human audiences versus the computer audience. Conversely, participants highlighted simple perceptual modifications for the computer audience, producing more punctuated and wider ranging motions. This study suggests that people consider differences in their audiences when demonstrating action.

Social audiences can disrupt learning by teaching

“…In a replication of Herberg et al (2008), participants in the Computer condition looked less frequently at the audience picture (M = 4.04, SD = 3.17) than participants in the Human condition (M = 9.00, SD = 3.98), t (42) = 4.54, p b 0.01. They also pointed less frequently for the computer audience (M = 1.86, SD = 2.70 versus M = 5.27, SD = 4.07), t (42) = 3.24, p b 0.01.…”

“…We employed a design partly based on research establishing social effects from the mere presence of a passive audience (Guerin, 1986;Zajonc & Sales, 1966) and similar to that used in Herberg et al (2008) in which participants taught a human or a computer (represented by a simple picture) to solve the Tower of Hanoi problem. Thus, our goal was to test the degree to which the mere presence of a social audience would affect learning, independent of any actual interaction with that audience.…”

“…Even the mere suggestion that social actors will later be exposed to persuasive statements produced by the subject can increase subjects' attitude change (Nel, Helmreich, & Aronson, 1969;Wicklund & Duval, 1971;Wood, 2000). More to the point, research has previously shown that demonstrating simple tasks to an imagined social audience (represented by a picture of a person) produces a range of social responses, including social highlighting behaviors (such as pointing to task-relevant goals) that do not occur for a nonsocial audience (a computer; Herberg, Saylor, Ratanaswasd, Levin, & Wilkes, 2008).…”

Human Tutorial Instruction in the Raw

“…In general, humans interact with computers differently than with other humans, and this applies to instruction as well [Herberg et al 2008]. Other studies report that existing learning algorithms do not conform to the kinds of instruction that humans provide Breazeal 2008a, 2008b;Thomaz and Cakmak 2009], making it hard to adapt them for learning from tutorial instruction.…”

Analyzing Action for Agents with Varying Cognitive Capacities

“…Previous research has documented that people make different attributions about the cognitive capacities of humans, nonhuman animals, and machines (e.g., Brand, Baldwin, & Ashburn, 2002;Eddy, Gallup, & Povinelli, 1993;Gray, Gray, & Wegner, 2007;Herberg, Saylor, Levin, Ratanaswasd, & Wilkes, 2008;Levin et al, 2006;Rasmussen, Rajecki, & Craft, 1993;Shechtman & Horowitz, 2003). However, it remains unclear how these explicit beliefs influence on-line expectations about other agents' action representations.…”