Principles of information structure common to six levels of the human cognitive system (Q1072492)

This paper attempts to develop a theory of information structure in human cognitive systems, and to corroborate that theory independently in six levels of cognition. Ascending through the cognitive system, the levels are (1) simple cell function; (2) psychological scales; (3) perceptual organization; (4) categorization; (5) grammatical structure; and (6) planning. From the summary: It is argued that, in human cognition, information is organized as the algebraic structure G of a machine such that G is given a factorization sequence \(G=G_ 1\cdot G_ 2,...G_ n\) induced by a crucial dynamical systems criterion. This criterion states that, given any representation of a stimulus set S, the cognitive system attempts to remove the properties of S in order of their decreasing instability. The successive stabilization structure is represented as a dynamical system on G such that, under the dynamics, G is given a factorization sequence \(G_ 1\cdot G_ 2,...G_ n\) where the successive identity elements \(e_ n,e_{n-1},...,e_ 1\) \((e_ i\in G_ i)\) become successive sinks in the trajectory structure. It is proposed that any such decomposition \(G=G_ 1\cdot G_ 2,...G_ n\) is cognized as a structure of nested control in which each factor \(G_ j\) is cognized as a control group with respect to the dynamics of the cosets of its right subsequence \(G_{j+1}\cdot G_{j+2},...,G_ n.\) We argue that human cognitive reference frames are structured as such dynamically induced algebraic decompositions \(G_ 1\cdot G_ 2,...G_ n\). Three crucial properties of such frames are that the referential structure, which is induced among the underlying stimuli, is asymmetric, sequential, and explicit. Thus stimuli are referentially identified via the succession of sinks, \(e_ n,e_{n-1},...,e_ 1\), in the dynamical structure, rather than by invariants under G.