An approach to human cognition based on AI player creation for a picture-guessing card game Dixit

Abstract

This paper describes our study on AI player creation for a picture-guessing card game Dixit. In the game, the active player makes up a phrase (or sentence) that describes the image on one of her hand. The relation between the image and the phrase had better be moderate as she scores nothing if all other players or no player guesses correctly. We analyze the process of human thinking in each phase of the game and implement each processing by using widely available techniques or databases including Google Cloud Vision API and Wikipedia Entity Vector. We show the AI player can play the game with human players as we expected. We also present a new perspective on the study of human cognition based on this approach.