Spatial cognition through the lens of spatial language

Cogn Process (2009) 10 (Suppl 2):S212–S214 DOI 10.1007/s10339-009-0291-x EXTENDED ABSTRACT Spatial cognition through the lens of spatial language Michele I. Feist Ó Marta Olivetti Belardinelli and Springer-Verlag 2009 Introduction Much recent research in spatial cognition has focused on the modeling and representation of geometrically tractable aspects of inter-object relations (e.g., Freksa 1992; Hois and Kutz 2008; Kurata 2008), providing elegant and precise characterizations of a variety of possible spatial relations. While this information is surely part of human spatial cognition, however, it is insufficient by itself to account for human conceptual representations of space (Klippel et al. 2008). Language has been called a ‘‘window into the mind’’; yet, what can we learn about spatial cognition from spatial language? Research into the meanings of locative spatial terms indicates that they are semantically quite complex (e.g., Coventry and Garrod 2004; Feist 2000, 2008; Levinson et al. 2003), suggesting that similar complexity may underlie spatial cognition more generally. Levinson (1996, p. 374) has argued that ‘‘[e]very linguistic distinction must be supported by the relevant conceptual distinctions, perceptual acuities and mental algorithms.’’ As such, those aspects of the spatial world that must be attended in order to fluently speak must similarly form a part of human spatial cognition. If this were not the case, as Levinson argues, it would be impossible to use the linguistic system. In this paper, I will explore a view of spatial cognition through the language used to describe static spatial scenes. Based on cross-linguistic evidence about the factors important to related terms in 24 languages and on M. I. Feist (&) Institute of Cognitive Science, University of Louisiana at Lafayette, Lafayette, LA, USA e-mail: [email protected] 123 experimental evidence about the use of both topological (e.g., in and on) and projective (e.g., left, right, front, and back) prepositions in English, I will argue that spatial cognition may, like spatial language, be influenced by three types of information about a spatial scene and the objects in it: geometric, functional, and qualitative physical information. Cross-linguistic evidence and potential universals One source of evidence regarding potentially universal elements of spatial cognition comes from the comparison of spatial terms across a variety of unrelated languages. Studies of the lexical semantics of spatial terms across languages have provided evidence of considerable variation in the details of how spatial relations are linguistically categorized in addition to compelling similarities (Bowerman and Pederson 1996; Feist 2000, 2008; Levinson et al. 2003). For example, Feist (2008) noted two sources of variation in the spatial vocabularies of the 24 languages that she studied. First, she found that there are two kinds of spatial terms evident across languages: specific spatial terms, which occur in limited contexts and encode relatively specific locational information, and general spatial terms, which occur in virtually all contexts and concomitantly encode little specific locational information. Second, she observed that the sets of contexts within which each of the specific spatial terms that she collected was used were highly variable across languages, echoing the referential variability that had previously been noted in the literature (e.g., Bowerman and Pederson 1996; Feist 2000; Levinson et al. 2003). Despite these differences, Feist (2008) noted a high-level similarity: geometric, functional, and qualitative physical attributes of spatial Cogn Process (2009) 10 (Suppl 2):S212–S214 scenes characterize the meanings of both specific spatial terms (e.g., English in and on) and general spatial terms (e.g., Japanese ni), across the range of languages sampled. Because these aspects of spatial scenes were found to be encoded across languages, she argued that it is possible that they participate in our conception of space more generally. Experimental tests of the influences of the factors While the cross-linguistic evidence is compelling, it is to be expected that geometric, functional, and qualitative physical factors will influence speakers’ use of spatial terms in a controlled experimental task if these factors are in fact part of human spatial conception. In order to test for this possibility, Feist and Gentner (2003) and Feist (2000) created scenes that orthogonally varied factors of each of these types, then asked English-speaking participants to choose either in or on to describe the resulting scenes. In each scene, a single Figure object (or locatum) was placed in contact with an external surface of the Ground, or reference, object. Based on the reasoning that the animacy of the Figure and Ground would influence speakers’ perceptions of the relative stability of the scene, speakers’ expectations regarding this qualitative physical attribute of the scene were manipulated via changes in the animacy of the Figure and in the animacy of the Ground, with animate Grounds and inanimate Figures expected to lead to greater perceived stability. Variations in functional information were achieved via changes in the noun used to label the inanimate Ground (which was ambiguously drawn to maximize participants’ reliance on the noun for functional information), following findings in linguistics that noun choices are influenced by information regarding an object’s function (Labov 1973; cf. Coventry et al. 1994). Finally, geometry was manipulated via the concavity of the Ground, with greater concavity suggesting an interior and, thus, expected to be more compatible with in. Feist and Gentner (2003) and Feist (2000) found effects of each of these manipulations on English speakers’ choice between in and on, in addition to interactions between them. This suggests that all three kinds of factors participate in the semantics of spatial terms, and, by extension, potentially in spatial cognition more generally. Further evidence for the influences of multiple aspects of inter-object relations on the language used to describe those relations comes from ongoing work by Robinette and colleagues (in preparation; Robinette et al. 2008) on the uses of projective spatial terms in English (left, right, front, and back). Drawing on the methods employed by Feist and Gentner (2003) and Feist (2000), these studies examine the factors influencing the choice between an object-centered (intrinsic; Levinson 1996) frame of reference and a viewer- S213 centered (relative; Levinson 1996) one to describe a static pictured scene. In addition to the presence of a salient, distinguishable front on the Ground object (Robinette et al. 2008), Robinette (in preparation) has observed that the existence of a functional relation between the Figure and Ground increases the likelihood that an object-centered frame of reference will be adopted, providing evidence for the influence of both geometric and functional attributes on the uses of English projective spatial terms much like those found for English topological terms. Discussion Taken together, the studies reviewed above suggest a multi-componential view of spatial language incorporating geometric, functional, and qualitative physical attributes of spatial scenes (Feist 2000, 2008). Not only are influences of these three families of attributes evident across languages despite considerable semantic variation in spatial vocabularies, but also these attributes have also been found to influence speakers’ spatial term choice in controlled experimental tasks. Given that spatial language is multi-componential, it follows that spatial cognition may be similarly influenced by each of these three aspects of the relations between objects in the world, as the lack of such influences on spatial cognition would render the associated linguistic descriptions impossible (Levinson 1996). This suggests that the geometrically tractable aspects of inter-object relations studied in the field of spatial cognition (e.g., Freksa 1992; Hois and Kutz 2008; Kurata 2008), while critically important, are only part of the story. The rest of the story may lie in those extra-geometric factors, including functional and qualitative physical information, which, together with geometric information, may form the basis for a deeper understanding of human spatial cognition. References Bowerman M, Pederson E (1996) Cross-linguistic perspectives on topological spatial relationships. Paper presented at the 91st annual meeting of the American Anthropological Association, San Francisco Coventry KR, Garrod SC (2004) Saying, seeing and acting: the psychological semantics of spatial prepositions. Psychology Press, London Coventry K, Carmichael R, Garrod SC (1994) Spatial prepositions, object-specific function, and task requirements. J Semant 11:289–309 Feist MI (2000) On in and on: an investigation into the linguistic encoding of spatial scenes. Northwestern University, Evanston 123 S214 Feist MI (2008) Space between languages. Cogn Sci 32:1177–1199 Feist MI, Gentner D (2003) Factors involved in the use of in and on. In: Alterman R, Kirsh D (eds) Proceedings of the twenty-fifth annual meeting of the Cognitive Science Society. Lawrence Erlbaum Associates, Mahwah, pp 390–395 Freksa C (1992) Using orientation information for qualitative spatial reasoning. In: Frank AU, Campari I, Formentini U (eds) Theories and methods of spatio-temporal reasoning in geographic space, LNCS 639. Springer, Berlin, pp 162–178 Hois J, Kutz O (2008) Natural language meets spatial calculi. In: Freksa C, Newcombe NS, Gärdenfors P, Wölfl S (eds) Spatial cognition VI—learning reasoning and talking about space. Springer, Berlin, pp 266–282 Klippel A, Worboys M, Duckham M (2008) Identifying factors of geographic event conceptualization. Int J Geograph Inform Sci 22:183–204 Kurata Y (2008) The 9?-intersection: a universal framework for modeling topological relations. Paper presented at the GIScience 2008, Park City 123 Cogn Process (2009) 10 (Suppl 2):S212–S214 Labov W (1973) The boundaries of words and their meanings. In: Bailey C-JN, Shuy RW (eds) New ways of analyzing variation in English. Georgetown University Press, Washington, DC, pp 340–373 Levinson SC (1996) Language and space. Annu Rev Anthropol 25:353–382 Levinson SC, Meira S, The Language and Cognition Group (2003) ‘Natural concepts’ in the spatial topological domain—adpositional meanings in crosslinguistic perspective: an exercise in semantic typology. Language 79:485–516 Robinette LE Re-viewing space: factors motivating use of the objectcentered frame of reference. University of Louisiana at Lafayette, Lafayette (in preparation) Robinette LE, Feist MI, Kalish M (2008) Motivating use of an objectcentered frame of reference. Poster presented at CogSci 2008, Washington DC