Bryant (1992) proposes a unitary spatial representation system (SRS) constructs spatial knowledge representations based on input from either linguistic descriptions of environments or perceptual information acquired while directly experiencing environments, or both. In this commentary, I discuss some of the characteristics of the spatial-cognitive system that Bryant has touched upon, particularly those dealing with large-scale, or environmental, spaces. Attempts to build a common framework for understanding both spatial and linguistic cognition are interesting and laudable, especially if one believes that linguistic cognition evolved from nonlinguistic cognitive systems such as those responsible for spatial cognition. In addition to examining the spatial representations derived from linguistic descriptions, much of value could be gained by examining the linguistic descriptions derived from spatial representations, either external or internal.
2.1 Quite appropriately, Bryant discusses the reference-frames that are used to organize spatial knowledge about environments. He invokes the commonly made distinction between egocentric and allocentric frames (e.g., Hart & Moore, 1973). The former occurs when locational knowledge about objects or places is encoded relative to a person's own location in space ("the mountain is very far from me"); the latter occurs when locational knowledge is encoded independently of one's own location but relative to another externally located object or place ("the mountain is far from the ocean"), or to an external abstract coordinate system ("the mountain is at 32 degrees latitude north"). Bryant also mentions a third frame which he calls the "external frame." This is based on "the external spatial framework analysis, which is... based on the body but projected forward in the field of view" (Bryant, 1992, 2.3). As I understand Bryant, this frame might be described more parsimoniously as an egocentric frame based not on individuals' actual locations at a given time but on some other place at which they are imagining themselves to be located (see 2.3 below); it is an "imagined egocentric frame." As Bryant points out, an object-centered frame is probably not especially relevant to the SRS; its role is in object recognition, not object localization.
2.2 Bryant maintains that the SRS incorporates both egocentric frames (actual or imagined) and allocentric frames. I agree that both types of frames are important to spatial knowledge, but I believe some clarity on this issue could be gained by a consideration of the distinction between working memory (WM) and long-term memory (LTM) knowledge. Evidence suggests that human spatial information about environments is in fact stored allocentrically in LTM, and in a Cartesian rather than a polar coordinate system (Montello & Presson, 1992; Rieser, 1989). When the information is placed in WM in order to perform some spatial task (e.g., direction-giving, some cases of wayfinding), it is organized from the perspective of the actual or imagined location of "ego." This basically suggests that there is no such thing as a spatial image free of the thinker's real or imagined perspective. I would not maintain, however, that all spatial knowledge must be displayed as a spatial image in WM in order to be used (see 4.1 below).
2.3 Working-memory representations of environmental space are always organized with respect to a viewer's perspective, but this need not be the normal perspective of an earth-bound traveler. Literature on cognitive mapping (e.g., Siegel & White, 1975) has commonly made a distinction between "route maps" and "survey maps." This distinction refers to much more than just the perspective from which spatial knowledge is accessed, but it does suggest that people can imagine environments from a terrain- level perspective or a "bird's-eye view." I use the term "travel recreation" to refer to a temporally ordered sequence of environmental images that basically simulates a sequence of percepts experienced while moving through the environment. The term "survey-map scanning" refers to simultaneous, maplike spatial imagery that represents part of an environment more abstractly, from a bird's-eye view. Palij (1987) has made a similar distinction.
3.1 Bryant discusses the fact that not all locations in space relative to one's body position are equally accessible (i.e., different speed and accuracy in answering questions about objects in front vs. behind). However, he claims that "people generally have the same degree of access to inferred spatial relations as to explicitly described or observed ones" (Bryant, 1992, 3.7). With respect to experienced environments, both the theory (Siegel & White, 1975) and the evidence (Hanley & Levine, 1983; Klatzky et al., 1990; Montello & Pick, 1992; Sullivan et al., 1992) that I am familiar with clearly contradicts this notion. These studies have shown that people access knowledge more quickly or more accurately between places along routes they have traveled directly than along routes that are only inferred (though people can certainly infer spatial relationships with nonrandom accuracy).
4.1 Bryant also claims (and cites evidence) that spatial knowledge is automatically processed. For the most part, he cites research on pictorial space in order to buttress this argument. Although performance on some environmental spatial tasks is apparently relatively automatic (e.g., updating while walking about), not all spatial tasks are (e.g., navigation in unfamiliar places, inferring novel routes in familiar places, direction-giving). The latter may be expected to show interference effects, developmental differences, and the effects of instructional set (Book & Garling, 1981; Kozlowski & Bryant, 1977; Lindberg & Garling, 1983; Smyth & Kennedy, 1982; Sullivan et al., 1992; and other developmental references too numerous to list). So I don't think this part of Bryant's argument leads very far: Undoubtedly some components of both spatial and linguistic cognition are effortful and some are automatic. I believe that disentangling the automatic and effortful components of spatial processing is a critical task to be accomplished, but in any case I don't find this particularly damaging to the main point Bryant is trying to make.
5.1 Attempts to build a common framework for understanding both spatial and linguistic cognition are interesting and laudable, especially if one believes as I do that linguistic cognition evolved from nonlinguistic cognitive systems such as those responsible for spatial cognition (a long-debated issue). A final comment: In addition to examining the spatial representations (mental models) derived from linguistic descriptions, much of value could be gained, I would think, by examining the linguistic descriptions derived from spatial representations, either external or internal (e.g., direction-giving, descriptions of places, etc.; cf. Klein, 1982; Linde & Labov, 1975).
Book, A., & Garling, T. (1981). Maintenance of orientation during locomotion in unfamiliar environments. Journal of Experimental Psychology: Human Perception and Performance 7: 995-1006.
Bryant, D. J. (1992). A spatial representation system in humans. PSYCOLOQUY 3(16) space.1
Hanley, G. L., & Levine, M. (1983). Spatial problem solving: The integration of independently learned cognitive maps. Memory & Cognition 11: 415-422.
Hart, R. A., & Moore, G. T. (1973). The development of spatial cognition: A review. In R. M. Downs & D. Stea (Eds.), Image and environment (pp. 246-288). Chicago: Aldine.
Klatzky, R. L., Loomis, J. M., Golledge, R. G., Cicinelli, J. G., Doherty, S., & Pellegrino, J. W. (1990). Acquisition of route and survey knowledge in the absence of vision. Journal of Motor Behavior 22: 19-43.
Klein, W. (1982). Local deixis in route directions. In R. Jarvella & W. Klein (Eds.), Speech, place, & action (pp. 161- 182). Chichester: John Wiley & Sons.
Kozlowski, L. T., & Bryant, R. J. (1977). Sense of direction, spatial orientation and cognitive maps. Journal of Experimental Psychology: Human Perception and Performance 3: 590-598.
Lindberg, E., & Garling, T. (1983). Acquisition of different types of locational information in cognitive maps: Automatic or effortful processing? Psychological Research 45: 19-38.
Linde, C., & Labov, W. (1975). Spatial networks as a site for the study of language and thought. Language 51: 924-939.
Montello, D. R., & Pick, H. L. (1992, April 11). Integrating knowledge of large-scale spaces above and below ground. Paper presented at the Environmental Design Research Association Conference (EDRA 23) in Boulder, Colorado.
Montello, D. R., & Presson, C. C. (1992). Movement and orientation in surrounding and imaginal spaces. Manuscript submitted for publication.
Palij, M. (1987). On the varieties of spatial knowledge: Cognitive maps, imagined terrains, and other representational forms. Unpublished manuscript, State University of New York at Stony Brook, Stony Brook, NY.
Rieser, J. J. (1989). Access to knowledge of spatial structure at novel points of observation. Journal of Experimental Psychology: Learning, Memory and Cognition 15: 1157-1165.
Siegel, A. W., & White, S. H. (1975). The development of spatial representations of large-scale environments. In H. W. Reese (Ed.), Advances in child development and behavior, Vol. 10 (pp. 9-55). New York: Academic Press.
Smyth, M. M., & Kennedy, J. E. (1982). Orientation and spatial representation within multiple frames of reference. British Journal of Psychology 73: 527-535.
Sullivan, C., Montello, D. R., Pick, H. L., & Somerville, S. C. (1992). Integrating spatial knowledge. Manuscript in preparation.