Wednesday March 18, 10:20-11:20

Rethinking the Unconscious Influences of Past Experiences

Pierre Perruchet

(Faculte des Sciences, Universite de Bourgogne, France)

All of us have intuition that earlier events may influence subsequent behavior either through the conscious and strategic exploitation of the memory of these events, or through unconscious ways. This intuition has been confirmed by a vast amount of experimental data. On these bases, the human mind has usually been inferred to be composed of two separate and potentially independent cognitive centers, whether the dichotomy is construed in terms of systems or processes. In the prevalent view, the whole set of cognitive operations linking incoming information with final behavior may operate in conscious or unconscious modes, with the unconscious mode being thought of as mirroring the conscious mode on many aspects. This dualist conception is reflected in the pervasive opposition of so-called implicit and explicit knowledge, and in the exclusive focus on the concept of dissociation.

This view is questionable, however. Both intuitive and experimental data can be accounted for as well within a radically different conception. Instead of thinking about conscious and unconscious adaptive modes as independent, both aspects can be integrated within a unitary, functional perspective. An integrative framework is outlined, in which intrinsically unconscious processes serve the function of generating, transforming, and manipulating intrinsically conscious representations. A computer-implemented model,PARSER, is presented to illustrate how ubiquitous unconscious associative processes may shape perceptual and representational units that become increasingly congruent with the world structure. PARSER accounts for performances that seemingly require complex and unconscious computation,while it has no need of concepts such as those of implicit representations or implicit knowledge. The implications of this framework for some current approaches in the implicit memory and implicit learning fields are discussed.