Individuation

In its most general sense, individuation is a process by which some entities or objects in the world are distinguished from others.  For example, when the visual system separates a figure from its background, it is individuated.  Pylyshyn (2003) has argued that most technical accounts of individuation involve composing descriptions of entities that can be used to pick them out – to identify them uniquely – from other possible entities that might be present.  Furthermore, a key element of creating such a description is identifying locations.  For instance, the traditional view of individuation would involve directing attention to a location, and retrieving the visual features that are present there.  However, Pylyshyn’s own theory of seeing and visualizing construes individuation in a more primitive fashion, one which does not involve composing or delivering descriptions, and on which does not make location explicit.  Indeed, he argues that one does not need to detect location to detect features, and supports a great deal of empirical evidence to support this surprising proposal.  He then proceeds to theorize that visual objects are individuated by being assigned visual indices, called FINSTs, that pick out the objects directly, and not indirectly via object properties or object locations.  Individuation is therefore analogous to assigning a demonstrative or a pointer to an entity (‘this thing’, ‘that thing’).  Once an entity is indexed, attention can be directed toward the index and the properties of the individuated entity can be explored.

References:

1. Pylyshyn, Z. W. (2003). Seeing and Visualizing: It's Not What You Think. Cambridge, Mass.: MIT Press.

(Added March 2011)