Poster #9 - Robots are Humans? An Investigation of Young Children’s Animacy Judgments and Empathy

• In Event: Poster Session 02

Abstract

With the advancement of technology, there are more and more opportunities for children to use social robots and smart teaching materials and toys as their learning companions or teachers. Therefore, understanding how children perceive the impact of simulated social robots on human-computer interaction is an important issue. Empathy plays an important role in social interaction with others. Previous cognitive neuroscience research has shown that pain context tasks can induce young children’s empathic neural responses to the pain of others. Three- to five-year-old children can recognize the differences between living and non-living things. However, the technologies of biomimetic morphology for social robots are becoming more and more advanced, blurring the lines between living things and agents. Accordingly, the goals of this study are to examine young children how to empathize with anthropomorphic social robots and how to reason about the mind perception of living things and agents.

Seventy kindergarteners participated in this study. There were 35 children aged 5 (15 girls and 20 boys; M = 64.3 months) and 35 aged 6 (19 girls and 16 boys; M = 76.6 months). The experiment was conducted in two phases. In phase one, children were asked to look at pictures of human-, robot- and doll-hand in painful or non-painful situations, and then to judge the target’s painful feelings in a computerized two-alternative forced-choice procedure. In phase two, participants would answer simple questions about the target’s animacy judgments.

Using the model (Human, Robot, Doll) and age as independent variables, the effects of animacy and pain value (Painful minus Neutral) with a two-factor mixed design ANOVA were conducted in this study. The results showed a statistically significant difference in the animacy among the three targets. Holm's Test revealed that the animacy of humans was greater than robots and dolls, and there was no significant difference between the robots and dolls; there was a statistically significant difference in the pain value among the three targets. Holm's Test revealed that the pain value of humans was greater than robots and dolls, and there was no significant difference between the robots and dolls. There was no significant difference in the age.

In conclusion, children aged 5 -6 can distinguish the animacy of living and nonliving objects, and they sort robots and dolls in the same category that is different from humans. And young children are more empathetic toward humans than toward robots and dolls. However, it is worth noting that the ratio of the pain value of young children’s empathizing with robots and dolls is not less than .90, which reveals they also think that robots and dolls have painful feelings as humans in most conditions. To sum up, children of this age think robots and dolls are not humans according to animacy, but they empathize with robots and dolls. Young children seem to have more empathy for everything.

Keywords: young children, animacy, empathy for pain, human-robot interaction

Authors

• Pin-Han Jeffrey Wang, National Tsing Hua University
  Presenting Author

• Chia-Hui Chiu, University of Taipei

• Peter Kuan-Hao Cheng, National Tsing Hua University