This commentary focuses on data suggesting that there are distinct aspects of spatial representation and thought and that not all spatial thought is functionally equivalent.
1.1 The basic thrust of Bryant's (1992) target article is that verbal descriptions of space are interpreted spatially, in terms of mental models, like spatial perception. This is not controversial (at least to most researchers who study spatial thought). He develops the idea that a dimensional Spatial Reference System (SRS) provides a general account of spatial mental models regardless of the form and channel by which the spatial information is presented. This further point has a great deal to recommend it (as indicated in the target article), but it is also more complex than Bryant presents. The focus of this commentary is on data suggesting that there are distinct aspects of spatial representation and thought and that not all spatial thought is functionally equivalent. This focus on certain complexities in Bryant's proposal is not intended to detract from our basic agreement with much of what he proposes.
1.2 One point already raised in Franklin's (1992) commentary, is that the specific way in which the SRS receives input from different perceptual (or knowledge) systems is not fully specified. Since Bryant proposes that the SRS can "use" 3 different frames of reference (egocentric, allocentric, external) to apply to the same information, we assume that Bryant is trying to account for what is often termed "spatial working memory" or spatial thought, as opposed to long-term memory (LTM) knowledge representations. Of course, the latter would also inform and be informed by SRS processes.
1.3 Bryant's proposal deals mainly with the paradigms of "mental models," in which all (or most) of the relevant spatial information is presented in the ongoing experiment and the focus is on combining immediate information from perceptual (and verbal) channels. Spatial models (or working memory) can also draw on LTM knowledge of environments and thus the form of such LTM knowledge bases needs to be considered in any complete account of spatial thought.
2.1 In our enthusiasm to point out that verbal descriptions of space are interpreted in spatial models, we should not gloss over the potentially important FUNCTIONAL distinctions in the way spatial information is used and represented. Not all spatial knowledge of the sort included by Bryant as "observed" is used in functionally equivalent ways in mental models. In work following up Evans & Pezdek's (1980) findings, Presson and colleagues (Presson & Hazelrigg, 1984; Presson, DeLange, & Hazelrigg, 1989) have shown that when people study maps or pictures (using secondary spatial learning), later recall is biased to the particular orientation in which it was learned (see also, Levine et al., 1982; Frederickson & Bartlett, 1987). In sharp contrast, when the information is learned by direct experience (primary spatial learning: e.g., looking at the environment in full scale or walking in it), later recall of the information is more flexible and not biased to the specific orientation (see also Sholl, 1987; Thorndyke & Hayes-Roth, 1982). Of course, the idea that there are important differences between perceiving environments and perceiving objects or figures is not a new one (e.g., Ittelson, 1973; Neisser, 1987; Schacter & Nadel, 1991).
2.2 The distinction between primary and secondary learning suggests that looking directly at a spatial array may lead to different functional properties of recall, compared to looking at the same information in a symbolic map or picture. Studies of verbal spatial descriptions are more like orientation-specific picture perception in that ease and accuracy of recall are biased by the specific orientation in which the information was initially described. Such orientation-specific recall was clear in Franklin & Tversky's (1990) data, even though it was not a focus of that work. In ongoing work, we have found that when simple routes are described, subjects always recall the information in a specific orientation, as if they had drawn a mental picture of the path. The orientation of the representation was typically with North at the top, regardless of the frame of reference used in the descriptions. If North was not described, then the direction of initial travel or straight ahead) was used as "up" (see also Palij et al., 1984).
2.3 In addition, the generality of the functional effects Bryant cites for mental models (e.g., front/back asymmetries) has not been explicitly demonstrated in some procedures; and it may be premature to assume that all procedural approaches will show parallel findings. For example, in a recent replication of Morrow, Greenspan, & Bower's (1987) work on mental models for text comprehension, Roepnack (1991) showed that differential priming for targets that were in front of vs. behind the protagonist in the goal room did NOT occur. She interpreted this as indicating that subjects were using a map-like representation and not imagining themselves inside the information (which the task did not require). This is part of a larger point that there may not be a spatial "competence" or single representational system separate from specific task contexts.
2.4 When Morrow et al.'s procedure was modified to require subjects to point to targets in various rooms in the model (Roepnack, 1991), there were different effects of orientation of protagonist for psychologically proximal targets (in the same room as the protagonist) and psychologically distal targets (in other rooms). For proximal targets, the protagonist's orientation did not affect accuracy. For distal targets, judgments were more accurate if the protagonist's orientation at the time of judgment was the same as the subjects' orientation to the map during the initial learning of the space. The latter effect is like the orientation-specific recall one often sees in judgments after map learning (Presson, 1987). These data indicate that there may be different representational bases for the pointing judgments in the goal room and in the other rooms (Roepnack, 1991).
3.1 On the one hand, findings such as those described above may limit the generality of some of Bryant's specific claims. On the other hand, they can be used to support Bryant's more general point that alternate frames of reference can be used in establishing mental models in spatial working memory (or SRS). We feel that the conditions under which different systems are used and the functional implications of such systems (the equivalences and nonequivalences) provide important challenges for an understanding spatial cognition.
Bryant, D.J. (1992). A spatial representation system in humans. PSYCOLOQUY 3(16) space.1.
Bryant, D.J., Tversky, B., & Franklin, N. (1992). Internal and external spatial frameworks for representing described scenes. Journal of Memory and Language, 31, 74-98.
Evans, G., & Pezdek, K. (1980). Cognitive mapping: Knowledge of real-world distance and location information. Journal of Experimental Psychology: Learning, Memory, and Cognition, 6, 13-24.
Franklin, N. (1992). Inquiring into the spatial representation system. PSYCOLOQUY 3(40) space.4
Franklin, N., & Tversky, B. (1990). Searching imagined environments. Journal of Experimental Psychology: General, 119, 63-67.
Frederickson, R.E., & Bartlett, J.C. (1987). Cognitive impenetrability of memory for orientation. Journal of Experimental Psychology: Learning, Memory, and Cognition, 13, 269-277.
Ittelson, W.H. (1973). Environment perception and contemporary perceptual theory. In W.H. Ittelson (Ed.), Environment and cognition (pp. 1-19). New York: Seminar Press.
Levine, M., Jankovic, I., & Palij, M. (1982). Principles of spatial problem solving. Journal of Experimental Psychology: General, 111, 157-175.
Morrow, D.G., Bower, G.H., & Greenspan, S.L. (1989). Updating situation models during narrative comprehension. Journal of Memory and Language, 28, 293- 312.
Morrow, D.G., Greenspan, S.L., & Bower, G.H. (1987). Accessibility and situation models in narrative comprehension. Journal of Memory and Language, 26, 165-187.
Neisser, U. (1987). A sense of where you are: Functions of the spatial module. In P. Ellen & C. Thinus-Blanc (Eds.), Cognitive processes and spatial orientation in animals and man, Vol. 2 (pp. 293-310). Boston: Martinus Nijhoff.
Palij, M., Levine, M., & Kahan, T. (1984). The orientation of cognitive maps. Bulletin of the Psychonomic Society, 22, 105-108.
Presson, C.C. (1987). The development of spatial cognition: Secondary uses of spatial information. In Eisenberg, N. (Ed.), Contemporary topics in developmental psychology (pp. 77-112), New York: John Wiley.
Presson, C.C., DeLange, N., & Hazelrigg, M.D. (1989). Orientation Specificity in spatial memory: What makes a path different from a map of the path? Journal of Experimental Psychology: Learning, Memory, and Cognition, 15, 887-897.
Presson, C.C., & Hazelrigg, M.D. (1984). Building spatial representations through primary and secondary learning. Journal of Experimental Psychology: Learning, Memory, and Cognition, 10, 716-722.
Roepnack, B.R. (1991). Effects of orientation in dynamic imagined scenes. Unpublished Master's Thesis, Arizona State University.
Schacter, D. L., & Nadel, L. (1991). Varieties of spatial memory: A problem for cognitive neuroscience. In H. Weingartner & R. Lister (Eds.), Cognitive neuroscience. New York: Oxford University Press.
Sholl, J. (1987). Cognitive maps as orienting schemata. Journal of Experimental Psychology: Learning, Memory, and Cognition, 13, 615-628.
Thorndyke, P., & Hayes-Roth, B. (1982). Differences in spatial knowledge acquired from maps and navigation. Cognitive Psychology, 14, 560-581.