next up previous
Next: BINDING AND CONVENTION Up: IMPLEMENTING POLICIES AND BINDINGS Previous: IMPLEMENTING POLICIES AND BINDINGS

BINDING, DEIXIS, AND GAZE

To a first approximation, people can only visually recognize objects at which they are directly looking. People achieve the illusion of direct access to arbitrary objects by rapidly changing their gaze direction. Thus in addition to the normal state of the environment, our lived world contains an additional state component, our gaze direction. Since we can normally change our gaze direction without changing the world, and vice versa, our lived world E' can be separated into a parallel product of the objective environment and our gaze direction:

displaymath2362

Our access to this world is through our gaze, which allows us to focus in on one particular object at a time. Our gaze implements a binding, or more precisely, a binding map, since it depends on the direction of gaze. If we model gaze direction as a number indicating which object is presently foveated, we have that:

displaymath2363

A person could implement a single-object binding just by fixating the object they wish to bind. First they would set the D component to some egg, and then use tex2html_wrap_inline2370 as their binding. Since tex2html_wrap_inline2370 is really a binding map, however, rather than a true binding, the agent must pervasively structure its activity so as to ensure that its gaze need never be redirected.



Ian Horswill
Wed Apr 2 15:17:20 CST 1997