Conversation Theory

Conversation Theory is a cybernetic framework developed by Gordon Pask and his collaborators in the 1970s that treats cognition as a recursive dialogue between two or more systems rather than as computation inside a single brain. The core claim is that understanding is not a state but a process — specifically, the process by which distinct cognitive agents converge on compatible descriptions of a shared domain through structured exchange.

In Conversation Theory, a conversation is not casual chat. It is a formal mechanism with distinct levels: participants produce descriptions of a topic, then descriptions of those descriptions (meta-descriptions), and continue this recursion until they achieve what Pask called agreement — not consensus in the political sense, but structural compatibility between their internal models. The theory was operationalized in adaptive teaching machines that diagnosed a learner's conceptual structure and adjusted their instructional strategy in real time, making it one of the earliest implementations of AI-driven personalized education.

The deeper implication, often overlooked, is that Conversation Theory dissolves the boundary between individual and social cognition. A mind, in this framework, is not a solitary processor but a participant in ongoing conversations — and the intelligence of the conversation may exceed the intelligence of any individual participant. This challenges the dominant paradigm in cognitive science and AI, which treats intelligence as a property of isolated agents.