| When people describe objects' locations, they must take their addressee's vantage point into account. They can use deictic perspectives, speaking from their own point of view ("to my right"), from the addressee's point of view ("in front of you"), or from a point of view true for both ("closest to both of us"). Or they can locate objects non-deictically, using descriptions true no matter where either person is ("in between" some other objects, "towards the center of the table", etc.). How do they choose?;In two studies, students described objects' locations for addressees whose vantage points were either the same as theirs or offset by 90, 180, or 270 degrees. The addressees were either imaginary, as in all previous research, or real--another student, behind a visual barrier, who could converse. In Study 1, students picked out one of two circles in a series of simple displays, and in Study 2, students described the locations of many objects in a complex display.;In both studies speakers with imaginary addressees consistently took their addressees' perspectives. Speakers in pairs varied more, sometimes taking egocentric perspectives exclusively. When speakers found they weren't being understood, they tried a different perspective, and kept going until they were understood. Speakers used fewer deictic expressions when their partners' vantage points were offset, and they used different descriptive strategies. When the roles of speaker and addressee changed, new speakers tried to be "fair," taking their own perspectives just as often as the old speakers had taken their own.;Instead of using perspectives consistently, as in earlier research with solo speakers, speakers here designed what they said for the particular needs of their addressees. They did this in principled ways. In general, speakers ground what they say, working together for as long as they need to make sure they understand each other. They try to reduce collaborative effort, even if this means making it harder on one person. In doing this they take into account that putting themselves in another's shoes is easiest when the other's vantage point is most similar to theirs. |
