Determining the Effect of Personality Types on Human-Agent Interactions

Hongying Du, Michael N. Huhns

2013 IEEE/WIC/ACM International Joint Conferences on Web Intelligence (WI) and Intelligent Agent Technologies (IAT) · 2013 · 11 citations · 4 references

Concepts

Abstract

Because of the large number of agents and robots beginning to affect everyday life of humans, it is important to understand how humans would treat agents in a mixed human-agent society. In this paper, we are trying to find answers to two questions: whether humans possess different attitudes towards other humans and agents, and whether the personality type of a human influences his/her decisions and how. To investigate these problems, first we use the Keirsey Temperament Sorter-II (KTS-II) to discover the personality types of our human participants. Then each participant plays the "Who Gets More Cake?" game three times, with a simulated human and an agent as opponents. The experimental results are shown in two aspects: the tendency aspect and the consistency aspect. It is shown that humans treat other humans and agents differently and humans with different KTS-II temperaments behave differently on the above two aspects. It is very possible that the Thinking-Feeling dichotomy of Myers-Briggs Type Indicator (MBTI) and the tendency results are not independent. Also, there is a correlation between the Extraversion-Introversion dichotomy and the consistency results.

References

4