Poster #131 - Young children’s conception of robots: the role of morphology and goal-directedness

• In Event: Poster Session 01

Abstract

Prior work has yielded contradicting evidence regarding the age at which children consistently and correctly categorize things as living or non-living. The present study examined how children conceptualize robots in comparison to novel animals and artifacts. Using a modified version of Gottfried and Gelman’s (2005) Naïve Biology task, 3-year-olds were asked if robots, animals, or artifacts possess mechanical or biological parts. To gauge how much animacy children attributed to the animals, artifacts, and robots, children also responded to a set of interview questions. The interview questions examined how children perceive the psychological (i.e., talk, think and feel) and biological (i.e., is alive) aspects of robots, animals, and artifacts. To examine the role of morphology, we used two robots that were judged by adult raters to differ in their similarity to a human body: a humanoid robot (Nao) and a non-humanoid robot (Dash). To investigate the role of dynamic characteristics, children saw one robot behave in a goal-directed manner and one robot exhibit non-goal-directed behavior (e.g., moving towards a ball rather than away from it). The 3-year-olds in our study correctly attributed biological insides to the animal and mechanical insides to the artifact. However, 3-year-olds seemed confused about the biological status of both robots and assigned biological and mechanical insides equally. Regarding the Animacy Interview, 3-year-olds also performed at chance level for the psychological interview questions for robots, animals, and artifacts. Thus, children were unsure whether the robots, animals, and artifacts could communicate, think, or feel. Interestingly, children rated the robots (Dash = 61%; Nao = 59%) as more alive than the artifacts (artifacts = 48%) but less alive than the animals (animals = 71%). Thus, 3-year-olds were confused about robots overall, with roughly half of the children indicating the robot was alive and half attributing animate psychological abilities to the robots (the ability to feel, think, and talk). The present findings suggest that very young children are unclear about the nature of robots when making explicit judgments. Regardless of whether the robots engaged in goal-directed or non-goal-directed behavior, children conceptualized both robots in the same way. We speculate that 3-year-olds experienced difficulty categorizing the robots as they had characteristics of both living (e.g., morphological features such as eyes, autonomous movement) and non-living things (e.g., made of plastic and metal). A follow-up study with 5-year-olds will clarify when children begin to conceptualize robots as depictions of social agents.

Authors

• Elizabeth Jessica Goldman, Concordia University
  Presenting Author

• Anna-Elisabeth Baumann, Concordia University

• Jenna Beaudoin, Concordia University

• Diane Poulin-Dubois, Concordia University