Academic journal article Cosmos and History: The Journal of Natural and Social Philosophy

Information, Physics and the Representing Mind

Academic journal article Cosmos and History: The Journal of Natural and Social Philosophy

Information, Physics and the Representing Mind

Article excerpt


Representing the world is a primary function of the mind. The ability to form, manipulate and use representations is not unique to humans: honeybees use dance to represent and communicate the direction and distance to food sources [1]; crows, dogs and other animals are capable of making and using tools [2], [3], [4]; several studies indicate that some species of birds and mammals are capable of planning [5], [6], [7]. Human use of representation has fundamentally altered the world around us. The agricultural revolution dramatically increased food production, generating an exploding human population and the rise of cities [8], [9], [10]. The industrial revolution reorganized production of material goods, transforming human society and radically altering the global environment [11]. Now, as the information revolution unfolds around us [12], humanity is applying our representation capability to representation itself. Although the full impact is not yet manifest, the information revolution is clearly giving rise to disruptive change on a worldwide scale. Each of these major worldwide revolutions has grown out of our ability to construct representations, manipulate those representations to form plans and understand their effects, execute the plans, and build improved representations using feedback on successes, failures and unintended side effects.

Underlying the information revolution is scientific study of the phenomenon of representation itself. Shannon's influential theory of information has found wide application [13]. Database technology, concerned with representing, storing and accessing information in computers [14], [15], [16], is a critical element of the infrastructure of today's business enterprise. Artificial intelligence employs computational knowledge representations to allow computers to exhibit intelligent behavior on tasks once thought to require humans [17]. Cross-fertilization between artificial intelligence and cognitive science has resulted in computational theories of human cognition and, conversely, artificial intelligence formalisms inspired by empirical research on human problem solving [18], [19], [20].

Knowledge representation and reasoning methods have grown more sophisticated as the problems have grown more challenging. Some of the most successful information processing methods originated in computational physics. The key insight of these physics-based methods is to represent the space of possible solutions as a fictitious multi-dimensional physical space in which good solutions have low energy. The problem of finding a good solution to a problem is thus transformed into the problem of finding a low-energy state in a physical state space. This enables the application of techniques from computational physics to solve information problems.

The successful analogy with computational physics may reflect something fundamental about how organisms form and manipulate representations and perform goal-directed action. This paper suggests a rethinking of the traditional metaphor of cognition as execution of algorithms on a digital computer. It may be both more fruitful and more accurate to conceive of representation as mapping problem features to an energy surface, learning as identifying representations that map good solutions to low free energy, and problem solving as efficient search for low free energy states. This conception of cognition is in natural accord with Stapp's [21] theory of efficacious conscious choice.


Newell and Simon [18] pioneered the now-common practice of developing computational theories of intelligent behavior, implementing the theories as computer programs, and evaluating the theories by comparing with human problem solving behavior. They stressed that cognition is performed by the physical brain and nervous system of an embodied agent situated in a physical environment. They offered the physical symbol system hypothesis as a scientific, empirically testable hypothesis about the nature of intelligence. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.