[The following is an excerpt from Chapter 6 (The Embodied Mind) of The Myth of Mirror Neurons, which describes how early cognitive psychologists viewed their break from behaviorism as resting on the idea of information processing. The term "cognition" was later applied to this movement and ended up leading to some confusion because of the colloquial definition of cognition as applying to higher-level mental abilities...]
Computation and the information processing approach
The question then became, how does the brain process information? The digital computer was being developed around the same time and served as a convenient heuristic to think about how the brain might achieve such a feat. The basic idea is that there is information on the one hand and a set of processing routines (mental apps or computational algorithms) on the other. The information serves as input to the processing routines, which then transform it according to the set of computations defined in the program (e.g., if x, then y) and the processing routines output the results of the transformations. The output can then be stored as new information, serve as input to other programs, and control devices like a display or printer. Inputs to a computer -- key presses, mouse jiggles and pokes, image captures -- don’t directly control what’s displayed on the monitor or what the printer prints; rather, those inputs are processed by various apps to convert them into words and images, solve math problems, or play solitaire. It is the output of the apps that directly control what is displayed or printed.
The point of the computer analogy, or more accurately the computer program analogy, is that the mind/brain works the same way: the inputs to the brain – photons hitting the retina, air pressure fluctuations impinging on the ear drum, and so on – don’t control human behavior directly; rather, those inputs are processed by various neural apps to convert them to words and images, solve math problems, or play solitaire
Some early cognitive models looked very much like computer programs. In fact, some were computer programs. One of the most famous programs, developed in the 1950s by Allen Newell, J.C. Shaw, and Herbert Simon, was called the Logic Theorist. The Logic Theorist, or LT as it was nicknamed, was written to prove theorems using symbolic logic, similar in spirit to what a high schooler encounters in geometry class. The program was given a database of axioms (e.g., symbolic logic statements like “p or q implies q or p”) and a set of processing rules for using the axioms to generate proofs, rules such as substitution or replacement. LT was then presented with a series of new logic expressions and instructed to discover the proof for each using the “given” axioms and the rules. If it proved a theorem, it could store that proof along with the given axioms for use in subsequent proofs.
LT performed quite respectably, proving 73% of the theorems it was given. Writing a program that could pass a high school geometry class was an impressive accomplishment for the infant field of computer science, but it had far more significance for the information processing approach to understanding human behavior. LT showed that complex, human-like behavior could be approximated quite well with a purpose-built information processing system. And this is how Newell et al. presented the LT program, as a straight-up theory of human problem solving. To bolster their argument, the team presented evidence that LT’s problem solving “behavior” exhibited features characteristic of humans solving similar problems, such as its ability to learn, its demonstration of a kind of “insight” (trying at first to solve a problem with trial and error and then, once hitting upon the solution, using the same approach to solve similar problems), and its ability to break a problem down into sub-problems.
Now, Newell and company were careful to point out that their theory does not imply that humans are digital computers, only that humans appear to be running a program similar to LT.
We wish to emphasize that we are not using the computer as a crude analogy to human behavior—we are not comparing computer structures with brains, nor electrical relays with synapses. Our position is that the appropriate way to describe a piece of problem-solving behavior is in terms of a program: a specification of what the organism will do under varying environmental circumstances in terms of certain elementary information processes it is capable of performing. This assertion has nothing to do—directly—with computers. Such programs could be written (now that we have discovered how to do it) if computers had never existed. A program is no more, and no less, an analogy to the behavior of an organism than is a differential equation to the behavior of the electrical circuit it describes. Digital computers come into the picture only because they can, by appropriate programming, be induced to execute the same sequences of information processes that humans execute when they are solving problems. Hence, as we shall see, these programs describe both human and machine problem solving at the level of information processes.
It was a one-two punch for behaviorism. Chomsky and others had pointed out the inadequacy of simple associationist explanations of human behavior and symbol-manipulating computer programs provided a viable and directly observable account of how the brain processes information. Psychology whole-heartedly embraced the information processing approach.
The movement was later termed the cognitive revolution, an unfortunate term in my view because it implies that the revolution holds only for the everyday definition of “cognitive,” higher-order functions like language, memory, problem solving, and the like. It is true that the majority of the earliest work in the field focused on these complex human behaviors, but the real point of the revolution was that everything about human behavior -- perception, motor control, all of psychology -- is a result of information processing. Psychologist Ulric Neisser, who literally named the field and wrote the book on it with his 1967 text, Cognitive Psychology, defined the domain of cognition this way:
“Cognition” refers to all the processes by which the sensory input is transformed, reduced, elaborated, stored, recovered, and used. … Such terms as sensation, perception, imagery, retention, recall, problem-solving, and thinking, among many others, refer to hypothetical stages or aspects of cognition.
Neisser’s table of contents underlined his view that cognition was not limited to higher-order functions. His volume is organized into four parts. Part I is simply the introductory chapter. Part II is called “Visual Cognition” and contains five chapters. Part III is “Auditory Cognition” with four chapters. Finally, Part IV deals with “The Higher Mental Processes” and contains a single chapter, which Neisser refers to as “essentially an epilogue” with a discussion that is “quite tentative”. He continues,
Nevertheless, the reader of a book called Cognitive Psychology has a right to expect some discussion of thinking, concept-formation, remembering, problem-solving, and the like…. If they take up only a tenth of these pages, it is because I believe there is still relatively little to say about them….
Most scientists today working on perception or motor control, even at fairly low levels, would count their work as squarely within the information processing model of the mind/brain and therefore within Neisser’s definition of cognition. Consider this paper title, which appeared recently in a top-tier neuroscience journal: Eye Smarter than Scientists Believed: Neural Computations in Circuits of the Retina. If anything in the brain is a passive recording device (like a camera) or a simple filter (like polarized sunglasses) it’s the retina, or so we thought. Here’s how the authors put it:
Whereas the conventional wisdom treats the eye as a simple prefilter for visual images, it now appears that the retina solves a diverse set of specific tasks and provides the results explicitly to downstream brain areas.
Solves a diverse set of specific tasks and provides the results… sounds like a purpose-built bit of programing—in the retina! We observe similar complexity in the control of simple movements, such as tracking an object with the eyes, an ability that is thought to involve a cerebral cortex-cerebellar network including more than a half dozen computational nodes that generate predictions, detect errors, calculate correction signals, and learn. It is not much of an overstatement to say that there is universal agreement among perceptual and motor scientists in neuroscience and psychology that perception and action are complex systems that actively transform sensory information and dynamically control action. As Neisser wrote in 1967, “Information is what is transformed, and the structured pattern of its transformations is what we want to understand.” The information processing model of the mind -- cognitive psychology as defined eloquently by Ulric Neisser -- now dominates the study of the mind/brain from computation in the retina to motor control to complex problem solving.
[The rest of the chapter considers the contribution of embodied cognition in this context.]