Space
People
|
David Bryant |
Space
Peoples' conceptions of space depend on their perception of space and their actions in it, which in turn depend on the perceptual and motor apparatus evolution has provided. Different spaces are crucial for different kinds of actions and depend on different perceptions; these have consequences for conceptions of space. Thus, mental spaces are quite different from the space of physics or surveying or cartography, where the space is primary and universal, and locations of things defined with respect to the dimensions of space. For reviews, see Tversky, 2001, 2002, in press.
We have studied the space of the body (Morrison and Tversky, 2004), the space around the body (Franklin and Tversky, 1990; Bryant, Franklin, and Tversky, 1992; Tversky, Kim, and Cohen, 1999), the space of navigation (e.g., Tversky, 2000), the spaces created by people to augment their cognition (e.g., Tversky, 2001). Diagrams, language, gesture (and other behaviors) each reflect conceptions of space, and serve to instill conceptions of space. Despite overlap, they are not equivalent.
Space of the Body
Like other objects, we know about bodies from perception; unlike other objects, we know about bodies from inhabiting them, controlling them. We have studied mental conceptions of the space of the body in a body part verification task for the major body parts typically named across cultures (head, arm, hand, chest, back, leg, foot). Although large parts are verified faster in imagery, in perception, distinctive parts are verified faster than large ones. Distinctive parts are those with greater contour discontinuity. Insider knowledge of the body adds functional information from action, from kinesthetic and proprioceptive feedback. This knowledge is reflected in body part verification where naming is involved. Names appear to arouse function. (Morrison and Tversky, 2004).
Space around the Body
This is the space within reach of eye or hand, a space we conceive of in three dimensions. People can construct and revise mental models of the space around body from narrative, akin to the movie that runs through the mind while reading. Mental models established by narrative have the same spatial characteristics as those established by perception or by diagrams (e.g., Bryant and Tversky, 1999; Bryant, Tversky, and Lanca, 2001). They allow keeping track of the objects around the body under movements and turns. Data from numerous studies support the Spatial Framework Theory (Franklin and Tversky, 1990) according to which people construct a mental spatial framework out of the 3 axes of the body. The accessibility of the axes depends on their asymmetries and the only asymmetry of the world, the up-down axis created by gravity. For an upright observer, the head/foot axis is fastest due to its asymmetry and its correlation with gravity. Front/back is next because of asymmetry, and left/right is slowest (e.g., Franklin and Tversky, 1990). Variations in the spatial situation (e.g., Bryant, Tversky, and Franklin, 1992; Franklin, Tversky, and Coon, 1992; Tversky, Kim, and Cohen, 2000) and in the means of acquiring it (Bryant, Tversky, and Lanca, 2001) have predictable effects on the accessibility of objects on the axes.
Space of Navigation
This is the space too large to be seen in a glance, the space we mentally piece together from exploration, from maps, from descriptions. The mental representations of such spaces are sometimes referred to as cognitive maps. Yet they are systematically distorted by a number of mechanisms. People erroneously think that Rome is south of Philadelphia and Boston is east of Rio; these are a consequence of alignment, where people mentally organize two large elements as more aligned than they actually are. People erroneously think that Berkeley is east of Stanford; they mentally rotate the axes of a large element like the Bay Area to the axes of the larger environment (Tversky, 1981). People think ordinary buildings are closer to landmarks than vice versa. And so on (see Tversky, 2000 for a review). The melange of systematic biases and errors do not yield a coherent cognitive map; cognitive collage is a better metaphor for mental representations of environments (Tversky, 1993).
Describing the Space of Navigation: Route and Survey Perspectives
Environments are typically described from route or survey perspectives. A "route" perspective takes a view from within, and describes landmarks with respect to the changing position of "you" a traveler in the environment in terms of left, right, front, and back. A "survey" perspective takes a view from above and describes landmarks with respect to each other in terms of north, south, east, and west. When environments are well-learned from one perspective, people answer inference questions from both perspectives with equal speed and accuracy, suggesting that mental representations are perspective-free (Taylor and Tversky, 1992a, 1992b, 1996).
On-line establishment of mental representation is slowed by perspective switches (Lee and Tversky, 2004, in preparation). However, once established, mental models of simple environments seem perspective-free in that queries from the other perspective are answered quickly and accurately as queries from the same perspective.