«Простите, я никак не могу понять»: способы реагирования на непонимание во взаимодействии человека и робота
The article deals with the social mechanisms of human–robot interaction, related to the expression of non-understanding. The methodological framework of the study is ethnomethodological conversion analysis. Through a detailed analysis of conversations between callers and robot in one of Russian call centers it is shown that, contrary to the widespread view of non-understanding as a negative phenomenon consisting in the absence of understanding, non-understanding is a situational and concerted achievement of the participants to interaction. Two types of conditions for non-understanding between human and robot are distinguished in the article: institutional (related to the specific organizational context in which the interaction takes place and which is produced in it) and interactive (related to the organization of interaction). It is shown that non-understanding is a methodic phenomenon because participants to interaction use specific methods for its production, including both forms of expressing non-understanding and ways of its anticipation and identification. The methods used in conversations of callers with robots are: direct expression of non-understanding by robot, “robotization” of their speech by callers, and reformulation of their utterances by callers in response to robot’s non-understanding. In interaction with the robot humans, on the one hand, try to project and eliminate from the beginning the possible sources of non-understanding, and on the other hand, when non-understanding emerges, find a “problem” in their utterances and try to solve it by reformulating their words. The identified features of human–robot interaction allow not only to give a more accurate description of the human interaction with similar interactive agents, which are becoming increasingly common in everyday life, but also to show that non-understanding is not a failure in understanding, but a social achievement of the participants to interaction, whether humans or robots. Since non-understanding should be demonstrated and recognized in details and sequences of actions, it is necessary to consider it as an independent interactive phenomenon. This approach requires, in turn, to consider “human” and “robot” as an indexical phenomena, whose meaning is determined by the local order of interaction.