Towards an Information‐flow Model of Human Behaviour*

This paper is concerned with behaviour possible in an information‐flow system intended explicitly as a hypothetical model for comparison with human information‐handling system. A statistically self‐organizing system is described in which not only normal homeostatic behaviour but also such activities as invention of fruitful hypotheses, imagination of fictitious situations, and like would find a natural place. Discussion is confined mainly to manner of concept‐formation and concept‐handling in such a system. It has been suggested that correlate of perception (as distinct from reception) is activity which organizes an outwardly directed internal matching response to signals from receptors. This organizing activity amounts logically to an internal representation of feature in incoming signals to which it is adaptive, i.e. feature which is thus ‘perceived’. A hierarchic structure is postulated wherein much of organizing activity is concerned with modifying probabilities of other activity. Abstract concepts and hypotheses are represented by ‘sub‐routines’ of such organizing activity. These can in principle be evolved as a result of experience in a manner analogous to‐or at least fruitfully comparable with‐process of learning and discovery; and it is not necessary for designer to predetermine, nor possible for him to foresee, all conceptual categories in terms of which information received by system may be structured. Particular attention is directed to conditions under which such a process of concept‐formation could take place with sufficient rapidity. Some of symptoms of psychopathology find apparent correlates in possible modes of malfunction of such a system. But in this, as in other respects, it is not intended to press resemblances; intention is rather to stimulate critical comparisons, in order that differences between information‐flow model and real thing may continually lead to progressive refinement and enlarged understanding.