Methods | Statistics | Clinical | Educational | Industrial | Professional items | World psychology |
Philosophy Index: Aesthetics · Epistemology · Ethics · Logic · Metaphysics · Consciousness · Philosophy of Language · Philosophy of Mind · Philosophy of Science · Social and Political philosophy · Philosophies · Philosophers · List of lists
Functionalism is the philosophical underpinning of much empirical research in psychology and cognitive science; however, as research goes on the functionalist approach is continually criticised for its shortcomings.
Chomsky (1975) commented that it is a peculiarity of intellectual history that physical structures, like legs, are taken to be genetically determined whilst mental structures, like those responsible for speech, are thought to be products of the social environment. Understanding that this distinction is incorrect is an important first step in understanding functionalism. This is because just as legs have been selected by evolution for their function so too have mental faculties. Turing (1950) noted this teleological functional property of brains and analogized it to the computer, which has been designed to function. In particular, whilst computers are physical devices with electronic substrate that perform computations on inputs to give outputs so brains are physical devices with neural substrate that perform computations on inputs which produce behaviours. While this comparison may be fictional rather than fundamental it helps show that functionalism is the theoretical level between the physical implementation and behavioural output (Marr, 1982). Therefore, it is different from its predecessors of Cartesian dualism (advocating discrete mental and physical substances) and Skinnerian behaviourism and physicalism (declaring only physical substances) because it is only concerned with the effective functions of the brain, through its organization or its ‘software programs’. More formally, functionalism says that “mental states are constituted by their causal relations to one another and to sensory inputs and behavioural outputs” (Block, 1996).
Early History of Functionalism
Setting their course against the Structuralist tenor of the times a group of psychologist lead by John Dewey, in  sought to ground psychology in terms of the function of psychological phemonena in adapting subjects to their environment. So conscious experience was viewed as being necessitated by the limitations of automatic reflexes and conferred a functional, evolutionary advantage.
Focused on Chicago where George Herbert Mead and James Rowland Angell were also influential other centres also supported developments. Thorndike and Woodsworth made progress at Columbia while William James led the way at Harvard.
Problems with Functionalism
Consider the character Kate being asked to establish if the replies on her computer were from a human or computer (see Turing, 1950), ticking “Human” on the supplied red paper and then exiting the laboratory passing a Chinese man on the way coming out of another door. Ignoring most of these details for the moment, consider if the red paper Kate made her response on were green to Kate because she had undergone surgery as an infant and had her ‘green’ and ‘red’ vision inverted although was still brought up to identify ‘green’ objects (really red) as red and vice versa. Indeed, if Kate were to have a twin who did not have the operation they would both report that the piece of paper was red, even thought Kate was experiencing green. This thought experiment using ‘inverted spectra’ (Block and Fodor, 1972) presents a prima facie argument against functionalism because Kate and her twin experience two states that are functionally commensurate but qualitatively dissimilar (Block, 1994). The notion that Kate and her twin can have two different qualia (the experience of experience) but remain functionally indifferent clearly demonstrates that functionalism is not robust enough to explain individual differences in qualia. Block (1980) also argues against the functionalist proposal of multiple realizabilty, where hardware implementation is irrelevant because only the functional level is important. Block considers if the billion or so neurons in a brain were given functionally equivalent electronic substitutes fitted with radios to communicate and distributed to the Chinese population if China would have qualia. He thinks it could not and argues that functionalism is inadequate accordingly.
The Chinese Room
If Kate were mistaken and a computer really were responding with such mastery then might we assume that the computer was conscious? Indeed, the problems of functionalism are most clear when its role in consciousness is examined. If functionalism is correct then all thinking, including the feeling of consciousness, can be explained by computations (Penrose, 1994). However, supposing Kate were talking to a computer except that the computer’s functional level were represented by the Chinese man she saw on the way out. The Chinese man would be in another room with a huge database describing the outputs to give from any inputs on his screen. At Kate’s end she is seeing conversationally appropriate English responses to her responses and yet the Chinese man (who speaks no English) understands nothing except that for x input he is to give y. This altered version of the thought experiment, The Chinese Room by Searle (1980) shows how functionalism could be false because in acting as the ‘program’ executing instructions the Chinese man lacks intentionality and is only concerned with syntactic procedures rather than semantic content. Searle pre-empted critics who claimed the ‘system’ as a whole understands by positing that if the Chinese man remembered all the possible instructions and went about his life he still would not understand English. If functionalism is to explain all mental processes it encounters problems in explaining how meaning can be derived from pure computation.
If Kate’s answer were correct because she was talking to human then we might suppose that the functioning of both her brain and her converser’s to be roughly the same during their conversation. For example, both would be reading words on their screen. In a psychological investigation some variable, such as word length, might be manipulated to measure the effect on another variable, say, reaction time from which some inference about reading might be made. This describes the inductive scientific method, where reasoning is made from observed facts. However, if the example is continued and the investigation finds that longer words take longer to respond to there are several interpretations that can be made. One is that the word recognition is serial, letter by letter. Another is that it is parallel (letters are processed all at once) but longer words require more lexical ‘post-recognition’ processing. The details here are not important; however, what is important is that inductive functionalism is very bad at accurately determining what functions are performed by the brain. The problem can be made clearer if Kate’s answer were incorrect and her converser were a computer. Assuming the outputs and inputs remained the same then it is not necessarily the case that processing is the same between human and computer, just as one clock’s mechanism may be different from another’s although they have the same inputs and show the same time. In any case, this is a serious problem for functionalist cognitive science because where multiple explanations exist it may be impossible to ascribe one correctly or, worse, possible to ascribe one incorrectly
As has been shown one of the central arguments against functionalism is that it fails to account for the qualitative aspects of minds or qualia. However, such arguments intuitively assume qualia exists but does not in anything else, which is tantamount to circularity. No evidence exists that qualia exists or does not exist and therefore it is not clear if humans do not have it or China does. Some functionalists believe China would have qualia but that due to the size it is impossible to imagine China being conscious (Lycan, 1987). Indeed, it may be the case that we are constrained by our theory of mind (e.g. Baron-Cohen, Leslie and Frith, 1985) and will never be able to understand what Chinese consciousness is like. Therefore, if functionalism is true either qualia will exist across all hardware or it will not exist at all but is illusory (Dennett, 1990). However, whichever of these is true the problem of obtaining semantic content from syntactic operations remains. The ‘systems reply’ is a rejoinder, which claims as a whole the system understands Chinese and that the ‘processor’ does not just as a neuron would not (Harnad, 2001). However, the issue is still moot. Finally, inductive functionalism is problematic because of the risk of false conclusions. These will be greatly reduced, but not avoided, by continuous empirical hypothesis-testing and peer criticism. Furthermore, a multidisciplinary approach integrating neuroscience, psychology and evolutionary biology will allow more data to converge on theories and models which will reduce the risk (Mundale and Bechtel, 1996). However, of all the arguments made against functionalism the risk of false conclusion is the most damaging and, unfortunately, is largely unavoidable.
Modern cognitive science is reliant on one position in its theoretical approach to research. This theory called functionalism stems from the notion that the brain has evolved to better the survival of its carrier by acting as an information processor. In processing information the brain is considered to execute functions, similar to those executed by a computer, in particular positing that mental states are defined by their causal relations between inputs, outputs and other functional states. However, this approach has been attacked on several fronts. The main argument is that whilst functionalism is adequate in explaining much of the brain it is not robust enough to explain subjective experience or qualia. Thought experiments have demonstrated that twins could be functionally commensurate but have different qualia and that multiple realizabilty could make China functionally identical to the human brain but without qualia. Equally, China could have qualia but which we are incapable of understanding. Similarly, qualia may not exist as a measurable phenomenon but as an illusion created by other processes of the mind. Another criticism levelled has been that computation lacks intentionality and that from syntax semantics cannot emerge; however no consensus has been reached on the validity of this claim. Finally, with functionalism there a serious risk of false conclusions being made, and this is perhaps where the framework is most vulnerable.
- Baron-Cohen, S., Leslie, A., & Frith, U. (1985). Does the autistic child have a "theory of mind"? Cognition 21, 37-46
- Block, N. & Fodor, J. , “What Psychological States Are Not,” Philosophical Review, 83, 159-181
- Block, N. (1980). Introduction: what is functionalism? Readings in philosophy of psychology. Cambridge, MA: Harvard University Press.
- Block, N. (1994). Qualia. In S. Guttenplan (ed), A Companion to Philosophy of Mind. Oxford: Blackwell
- Block, N. (1996). What is Functionalism? The Encyclopaedia of Philosophy Supplement. New York: MacMillan Reference Books
- Chomsky, N. (1975). Reflections on language. New York: Pantheon
- Dennett, D. (1990) Quining Qualia. In W. Lycan, (ed), Mind and Cognition. Oxford: Blackwells
- Harnad, S (2001) What's Wrong and Right About Searle's Chinese Room Argument? In Bishop, M. and Preston, J., (Eds.) Essays on Searle's Chinese Room Argument. Oxford University Press.
- Lycan, W. (1987) Consciousness. Cambridge, MA: MIT Press.
- Marr, D. (1982). Vision: A Computational Approach. San Francisco: Freeman & Co.
- Mundale, J. and Bechtel, W. (1996). Integrating Neuroscience, Psychology, and Evolutionary Biology Through a Teleological Conception of Function. Minds and Machines, 6, 481-505.
- Penrose, R. (1994). Shadows of the Mind. Oxford: Oxford University Press.
- Searle, J.R. (1980) Minds, brains, and programs. Behavioural and Brain Sciences, 3 (3), 417-457
- Thorndike, Edward L. & Woodworth, Robert S. (1901a). The influence of improvement in one mental function upon the efficiency of other functions (I). Psychological Review, 8, 247-261 Full text]
- Thorndike, Edward L. & Woodworth, Robert S. (1901b). The influence of improvement in one mental function upon the efficiency of other functions: II. The estimation of magnitudes. Psychological Review, 8, 384-395. Full text
- Thorndike, Edward L. & Woodworth, Robert S. (1901c). The influence of improvement in one mental function upon the efficiency of other functions: III. Functions involving attention, observation, and discrimination. Psychological Review, 8, 553-564.Full text
- Turing, A.M. (1950) Computing Machinery and Intelligence. Mind, 49, 433-460.
|This page uses Creative Commons Licensed content from Wikipedia (view authors).|