Book Description

This dissertation describes a computer simulation model of paranoia. The model mimics the behavior of a patient participating in a psychiatric interview by answering questions, introducing its own topics, and responding to negatively-valued (e. g., threatening or shame-producing) situations. The work focuses on the motivational mechanisms required to instigate and direct the modelled behavior. Major components of the model are: (1) A production system (PS) formalism accounting for the instigation and guidance of behavior as a function of internal (affective) and external (real-word) environmental factors; (2) A model of affects (emotions) as an anticipation mechanism based on a small number of basic pain-pleasure factors; and (3) a formalism for intensional behavior (directed by internal models) requiring a dual representation of symbol and concept. An intensional object (belief) can be accessed either by sensing it in the environment (concept) or by its name (token). Similarly, an intensional action (intention) can be specified either by its conditions in the immediate environment (concept) or by its name (token). Issues of intelligence, psychopathological modelling, and artificial intelligence programming are discussed. The paranoid phenomenon is found to be explainable as an extremely skewed use of normal processes. Applications of these constructs are found to be useful in AI programs dealing with error recovery, incompletely specified input data, and natural language specification of tasks to perform.