C. Stachniss, K. Schill, and D. Uttal (Eds.): Spatial Cognition 2012, LNAI 7463, pp. 182–195, 2012. © Springer-Verlag Berlin Heidelberg 2012 Collaborating in Spatial Tasks: Partners Adapt the Perspective of Their Descriptions, Coordination Strategies, and Memory Representations Alexia Galati and Marios N. Avraamides Department of Psychology, University of Cyprus P.O. Box 20537, 1678 Nicosia, Cyprus {galati,mariosav}@ucy.ac.cy Abstract. The partner’s viewpoint influences spatial descriptions and, when strongly emphasized, spatial memories as well. We examined whether partner- specific information affects the representations people spontaneously construct, the description strategies they spontaneously select, and the representations their collaborating partner constructs based on these descriptions. Directors described to a misaligned Matcher arrays learned while either knowing the Matcher’s viewpoint or not. Knowing the Matcher’s viewpoint led to distinctive processing in spatial judgments and a rotational bias in array drawings. Directors’ descriptions reflected strategic choices, suggesting that partners considered each other’s computational demands. Such strategies were effective as reflected by the number of conversational turns partners took to coordinate. Matchers represented both partners’ viewpoints in memory, with the Directors’ descriptions predicting the facilitated perspective. Thus, partners behave contingently in spatial tasks to optimize their coordination: the availability of the partner’s viewpoint influences one’s memory and description strategies, which in turn influence the partner’s memory. Keywords: perspective-taking, coordination, spatial memory, dialogue. 1 Introduction People routinely share spatial information to coordinate in a variety of tasks, from giving directions to a visitor in an unfamiliar environment to jointly moving a piece of furniture across rooms. The selection of a perspective when producing or interpreting spatial descriptions has been systematically investigated, with findings identifying some of the cognitive, contextual, and communicative constraints influencing this selection process. However, this work usually focuses either on people’s linguistic choices without directly examining the representations that support perspective-taking (e.g., Schober, 1993, 1995, 1999) or focuses on processing in noninteractive tasks (e.g., Carlson-Radvansky & Irwin, 1994; Carlson-Radvansky & Logan, 1997; Mou et al., 2004a) or in tasks where the interaction between (presumably) collaborating partners is constrained (e.g., Duran et al., 2011; Shelton & McNamara, 2004). Thus, it’s not yet well understood how the perspectives that people spontaneously select, both for organizing spatial information in memory and for their descriptions, are