The Effect of the Repetitive Utterances Complexity on User’s Desire to Continue Dialogue by a Chat-oriented Spoken Dialogue System

Abstract

In everyday conversation, it is common for participants to repeat all or part of the other’s words, and such repetition is often accompanied by elements such as backchannels. Repetitive utterances have an empathic effect other than confirming information. However, if the repetitive utterances of a dialogue system are monotonous, the user may get bored quickly, and if it is too complex, there is a concern that it may place a cognitive burden on the user. In this study, we define complexity as the number of elements and patterns associated with repeated words and examine the effect of the complexity of repetitive utterances on the user’s perceived empathy and desire to continue dialogue. The complexity of the repetitive utterances was divided into three conditions: low, moderate, and high, and templates of repetitive utterances were made according to each condition. We constructed a chat-oriented spoken dialogue system that automatically generates repetitive utterances. A dialogue experiment was conducted with 12 subjects. As a result, no significant difference was found between the 3 complexity conditions for the evaluation items of the user’s perceived empathy and desire to continue dialogue. On the other hand, considering the characteristics of the user’s negative attitudes towards robots and the anxiety towards robots, the results suggest that the stronger the user’s negative attitudes and anxiety towards robots, the greater the desire to continue the dialogue after exposure to high- complexity repetitive utterances.