The Comprehension of “Left” and “Right” in a Referential Communication Task

The Comprehension of “Left” and “Right” in a Referential Communication Task Stanka A. Fitneva (Fitneva@Queensu.ca) Department of Psychology, Queen’s University Kingston, ON K7L 3N6, Canada Yi Song (7ys4@Queensu.ca) Department of Psychology, Queen’s University Kingston, ON K7L 3N6, Canada Abstract The study examined the role of establishing local conventions in the interpretation of left and right. 32 adults participated in a referential communication game carrying commands like “Put the red cup to the left of the green.” The results suggest that local conventions strongly influence the interpretation of the two spatial terms but global expectations about the usage of the terms also play a role. Keywords: perspective taking; spatial cognition; eye tracking. Introduction Computing the meaning of spatial relation terms, especially of directional terms like left and right, is difficult and time consuming (e. g., Bryant & Tversky, 1991; Farrell, 1979; Franklin & Tversky, 1990; Maki & Braine, 1985; Maki, Grandy, & Hauge, 1979). Left and right are acquired late in development， after other conceptually similar terms like front/back and above/below (Fisher & Kamenzuli, 1987; Internicola & Weist, 2003). They take longer to produce and understand (Franklin & Tversky, 1990) and when other options are available, adults appear to avoid using left and right (Mainwaring, Tversky, Ogishi, & Schiano, 2003). The explanation of the left/right confusion has been historically attributed to the bilateral symmetry of the nervous system (i.e., the existence of projections into both hemispheres) and the conceptual complexity of the referential frames evoked by the terms (Corballis & Beale, 1976; Maki, Grandy & Hauge, 1979). Another explanation is the intrinsic ambiguity of left and right (Mainwaring et al., 2003; Schober, 1993). In face-to-face conversations, utterances like “Put the red cup to the left of the green” require establishing whose perspective should be considered: the speaker’s or the addressee’s? Language users rely on a set of conventions in interpreting the meaning of linguistic expressions (Lewis, 1979/1991). There are global conventions, which are shared by the broader linguistic community and represent general patterns of word use. These conventions are reflected in dictionaries. There are also local conventions which develop in the course of communication within a subgroup of the community (e.g., family members) or with a single other individual (e.g., Brennan & Clark, 1996; Garrod & Anderson, 1987). As an example of the distinction, consider the word “mother.” Its dictionary definition is “a female parent.” Spoken amongst family members, however, “mother” identifies a specific individual and thus not just a female parent but also possibly an accomplished musician and a great cook. In the present study, we examined the extent to which adults’ processing of left and right is influenced by local conventions established in conversation. The influence of local conventions in conversation is illustrated by Brennan and Clark (1996). In their study, when participants saw a dress shoe together a high-heel shoe and a sneaker, they labeled it a loafer rather than a shoe, which is a more common label. This was expected because participants had to distinguish the dress shoe from the other shoes. However, when afterwards participants saw the dress shoe in the context of unrelated objects, they still referred to it as a loafer. Here, a loafer is overinformative and a shoe would suffice. People’s tendency to follow and rely on the recently established conventions in conversation provides a compelling account of these findings. Research has consistently shown that speakers tend to assume the listener’s perspective in providing spatial descriptions (Mainwaring et al., 2003; Schober, 1993). For example, Schober (1993) presented participants with a modified version of the referential communication task (Glucksberg, Krauss, & Weisberg, 1966). They saw scenes depicting two identical objects (in addition to indicating the place from which the addressee is looking at the scene) and the participants’ task was to guide the addressee to picking one of them. There was a limited number of ways to describe the scenes. The participants could either adopt their own perspective, saying something like “it’s the one on my left,” or the listener’s perspective, saying something like “it’s the one on your right.” Schober (1993) found that most of the time, speakers took the addressee’s perspective. This finding was replicated by Mainwaring et al. (2003) who also showed that this usage is modulated by the speaker’s assumptions about their own and the addressee’s cognitive load. Schober (1993) and Mainwaring et al. (2003) suggested that speakers take the addressees’ perspective because it is easier to translate a visual scene into language (the speaker’s task) than language to a visual representation (the addressee’s task). Regardless of the explanation, the global convention for the usage and interpretation of spatial terms of reference in the referential communication task appears to involve the addressee’s perspective. Despite the prevalence of using left and right from the addressee’s perspective, speakers mix spatial perspectives