Neuroscience & Mind Icon Neuroscience & Mind

Functionalism in Philosophy of the Mind

Michael Egnor


Yesterday I discussed John Searle’s Chinese Room argument, which denies that computers are capable of thought, and that computer functionalism is a valid description of the mind (“Is Your Brain a Computer?“). Readers may be interested to read a little more about the idea of functionalism.

With the eclipse of behaviorism and identity theory in philosophy of the mind, functionalism has become perhaps the dominant perspective in philosophy of the mind for philosophers and neuroscientists with a materialist perspective. Functionalism is the view that mental states are constituted entirely by their functional roles, rather than by their strict behavioral consequences, or by their physical correlates, or by their immaterial natures.

For example, a functionalist would define the mental state of experiencing pain as the functional connection between the inflammation of the nerve in your tooth and your moaning that your tooth hurts. A functionalist would define the mental state of believing that it’s raining as the connection between your perception that it’s raining outside and your act of unfolding your umbrella.

The functionalist perspective is often expressed by quip, “The mind is what the brain does.”

Because in functionalism mental states are functional relationships between physical events, inherent to most variants of functionalism is the theory of multiple realizability. Multiple realizability is the idea that mental states can arise in any system that carries out certain functions, whether or not the system is biological. Multiple realizability posits, for example, the computers can have mental states if functional relationships between electromechanical events in computers are sufficiently similar to functional relationships between electrochemical events in brains.

For example, if the electrical output of a neural network is similar in relevant ways (e.g. voltage, current, frequency, etc.) to the electrical output of the circuit in a computer, the computer will have the same mental state as the person with a neural network in his brain. Multiple realizability is the idea that any system, with the proper functional relationships, can have mental states regardless of the physical material out of which the system is made. Multiple realizability is the metaphysical basis for the theory of strong artificial intelligence.

Functionalism has gained popularity in the largely materialist philosophical community, as well as among neuroscientists who are mostly materialists, for several reasons.

First, it provides a means by which people who argue from a materialist perspective can evade the deep problems associated with behaviorism and identity theory that dominated materialist philosophy in the early and mid 20th century. Behaviorism has more or less disappeared as a credible approach to the understanding of the mind, and identity theory has been largely abandoned as well due to profound logical difficulties with the assertion that mental states are identical to brain states. Functionalism evades these notorious problems.

The second reason for functionalism’s modern popularity is its obvious analogy to computer science. From the functionalist perspective, the mind is to the brain as software is to hardware. Computer functionalism, which is the moniker by which functionalist theories based on computation are known, is quite appealing because it provides a basis for modeling the mind and brain activity as a kind of computation and it allows the substantial theoretical framework of computer science to be applied to cognitive neuroscience and cognitive psychology.

In computer functionalism, neural processes implement algorithms in the brain and this implementation is just what the mind is. Some neuro-philosophers, most notably Jerry Fodor, have proposed that human language is just a manifestation of the algorithms by which our brains function. This implies that there is a natural human “language of thought.” Fodor has called this putative language of thought “Mentalese.” This form of functionalism, which is quite popular today among philosophers, computer scientists, and neuroscientists, is sometimes called the computational/representational theory of thought.

Despite the attractiveness of functionalism and the attractiveness of the analogy between the mind and computation, there are serious problems with it as a theory of mind, as I noted yesterday and as I hope to discuss further.

Image: � denisismagilov / Dollar Photo Club.

Michael Egnor

Senior Fellow, Center for Natural & Artificial Intelligence
Michael R. Egnor, MD, is a Professor of Neurosurgery and Pediatrics at State University of New York, Stony Brook, has served as the Director of Pediatric Neurosurgery, and award-winning brain surgeon. He was named one of New York’s best doctors by the New York Magazine in 2005. He received his medical education at Columbia University College of Physicians and Surgeons and completed his residency at Jackson Memorial Hospital. His research on hydrocephalus has been published in journals including Journal of Neurosurgery, Pediatrics, and Cerebrospinal Fluid Research. He is on the Scientific Advisory Board of the Hydrocephalus Association in the United States and has lectured extensively throughout the United States and Europe.



Mind and TechnologyPhilosophyScienceViews