Proceedings of the Annual Meeting of the Cognitive Science Society

Many theories of communication claim that perspective-taking is afundamental component of the successful design of utterances for aspecific audience. We investigated perspective-taking in aconstrained communication situation: Participants played a wordguessing game where each trial required them to communicate atarget word without context. In each game, pairs of participants tookturns giving and receiving clues to guess target words, bothreceiving feedback after each trial. In Experiment 1, none of themeasures of participants’ performance improved over rounds,suggesting either that participants were unable to improve theirperspective-taking or that the task was simply too demanding forother reasons. In Experiment 2, we tested whether this lack ofimprovement was due to overall difficulty rather than inability totake perspective. While the success rate in Experiment 2 didimprove over the course of the game, our analyses indicated that theimprovement was due to participants discovering a frequencyheuristic (using rarer clue words) rather than improved perspective-taking per se. The results of these two experiments show thatimproving perspective-taking adaptively is very difficult when thereis no context to ground either signal choice or interpretation.