Home Curriculum Vitae Publications Conference Reports Forthcoming Extramural Colloquia Expert Testimony Teaching Healthcare The Human Ecology of Memory Research Archive Publications and Reports Rants


What is Cognitive Science?

And What Does It Have to Do with Religion?


John F. Kihlstrom

University of California, Berkeley


Paper presented at a conference on "Religion and Cognitive Science: From conflict to connection", co-sponsored by the Graduate Theological Union and the University of California, Berkeley. Berkeley, California, January 17, 2008.


For those of you who are from out of town, or at least not denizens of either Berkeley or Holy Hill, let me welcome you again to this conference on religion and cognitive science. The undergraduate cognitive science program is pleased to be a co-sponsor of this conference. As director of that program, it falls to me to provide an introduction to the field of cognitive science, and to say some things about what it has to offer the study of religion -- and, for that matter, what the study of religion has to offer cognitive science. But I must point out that there are about 50 faculty members involved in cognitive science at Berkeley, and if you asked them all what cognitive science is all about, I suspect that you would get about 50 different answers (see also Bechtel & Graham, 1998; Boden, 2006; Gardner, 1985; Nadel, 2003; Osherson, 1995; Sobel, 2001; Stillings et al., 1995; Thagard, 2005; Wilson & Keil, 2001).

So here's mine.

Slide2.JPG (29218 bytes)Cognition is about knowledge and knowing, and cognitive science tries to understand the acquisition, representation, and use of knowledge by minds, brains, machines, and social entities. The topics of cognitive science are pretty much coterminous with the topics of cognitive psychology -- but with a difference in approach that I hope to make clear.


A Capsule History of Cognitive Science

Slide3.JPG (32781 bytes)The deep origins of cognitive science are in modern philosophy -- which, since Descartes, has been focused on problems of epistemology, as opposed to metaphysics, ethics, and other traditional areas. Think about the debate in the 18th century between the rationalists and the empiricists concerning the origins of knowledge. This was also a time of scientific revolution in both physics and biology, and one would have thought that psychology would have been part of that scene as well. But the legacy of Cartesian dualism, with its emphasis on the immaterial nature of mind, was to take psychology off the scientific table. As late as Kant, in the late 18th century, psychology was an impossible science: measurement was essential to science, but the mind, being an immaterial object, was not subject to measurement; therefore, so Kant reasoned, psychology could not be a science. In less than 50 years, though, Ernst Weber (in 1834) and Gustav Fechner (in 1860) had discovered the first psychophysical laws, quantifying the relationship between the intensity of physical stimulation and the intensity of the resulting sensory experience; Hermann von Helmholtz performed experiments to understand the mechanisms of distance and motion perception (1856-1866); and Franciscus Donders had introduced (in 1868) reaction time as a means of measuring the speed of mental processes (Boring, 1950; Wozniak, 1992).

Slide4.JPG (28557 bytes)Reflecting the British empiricists' emphasis on experience as the source of knowledge, early experimental psychology focused on phenomena of sensation and perception. Indeed, Wilhelm Wundt (1873-1874), generally regarded as the "father" of experimental psychology, argued that psychology as a true quantitative, experimental Naturwissenschaft was limited to the study of sensation and perception; all "higher" mental processes, including what Wundt called "cultural" psychology, were consigned to a nonexperimental Geisteswissenschaft (but see Greenwood, 2003). Still, as early as 1885 Ebbinghaus proved Wundt wrong by inventing methods for the quantitative, experimental study of memory; a little later, Mary Whiton Calkins (1896) developed paired-associate learning to study the formation of associations, and Pavlov and Thorndike (both 1898) developed similar methods for studying learning in animals; and finally, in 1920, Clark Hull put the icing on the cake by adapting Ebbinghaus' methods to the study of concept formation, a major aspect of thinking.

Slide5.JPG (32732 bytes)Unfortunately, just as psychology was ready to address problems in cognition at all levels, the dark days of behaviorism descended on the field. In a syllogism reminiscent of Kant's, John B. Watson (1913, 1920) and other behaviorists argued that science was based on objective, public observation, but mental life was inherently subjective and private; therefore, psychology -- if it were to be a true science -- had to banish the mental from its discourse. Along the same lines, Gilbert Ryle (1949) famously characterized the mind as "the ghost in the machine". William James's science of mental life quickly became B.F. Skinner's (1938, 1953) science of behavior. Psychology, as one wag put it, having lost its soul, now lost its mind as well (Woodworth, 1921, p. 2).

Slide6.JPG (36187 bytes)The hegemony of behaviorism within psychology lasted for more than half a century, but during and after World War II the development of high-speed computers, cybernetics, and information theory set the stage for what has since come to be known as the cognitive revolution in psychology (for details, see Baars, 1986; Boden, 2006; Gardner, 1985; Hirst & Miller, 1988).

Slide7.JPG (34309 bytes)Of particular importance was the year 1956, in which Jerome Bruner and his colleagues published A Study of Thinking -- the first experimental exploration of concept-formation since Hull. For two months that summer, psychologists, computer scientists, and others met at Dartmouth to consider the prospects for artificial intelligence. And at a symposium that fall at MIT, Newell and Simon described their computer simulation of problem-solving; Noam Chomsky presented a new way of viewing language, with an emphasis on syntactical rules; and George Miller discussed the limitations of human information processing. All three of these papers were delivered on the same day: September 11, 1956, which George Miller has since characterized as the birthdate of cognitive science.

Slide8.JPG (40308 bytes)This is the birthdate of cognitive science, not of cognitive psychology, because in 1956, and for many years thereafter, the hegemony of behaviorism was such that there was no institutional home for cognitivism in American academic psychology. Bruner, Miller, and their colleagues, for example, had to leave the confines of Harvard's psychology department, in Memorial Hall (actually, Bruner was in Emerson, in the parallel Department of Social Relations, but that is another story), to set up their Center for Cognitive Studies in rented quarters off Harvard Square, at the corner of Bow and Arrow Streets (Cohen-Cole, 2007).

Nevertheless, the 1960s were a period of rapid development on many fronts (Baars, 1986; Gardner, 1985; Hirst & Miller, 1988). Of special importance, Ulric Neisser published his seminal monograph on cognitive psychology, creating the textbook infrastructure for new courses on cognition, while the journal of the same name provided an outlet for publication of cognitive research.

Slide9.JPG (35177 bytes)Developments continued into the 1970s. Of particular importance was the support of the Sloan Foundation, which was keenly interested in promoting the leading edges of scientific progress. It established a program in cognitive science in 1976, and beginning in 1979 supported the development of training programs in cognitive science at a number of institutions, including UC Berkeley.

Slide10.JPG (37213 bytes)The result of all this activity was to institutionalize an interdisciplinary field of cognitive science, paralleling, and competing with, cognitive psychology. The Sloan Foundation was quite clear on this point: cognitive science was not to be a wholly owned subsidiary of psychology. Rather, it was conceived as "an autonomous science of cognition" -- an interdisciplinary effort bringing together a number of separate disciplines -- not just psychology and computer science, but philosophy, linguistics, neuroscience, and anthropology into the mix as well. It was to be, in Howard Gardner's phrase, "the mind's new science", with each of the fields comprising "the cognitive hexagon" making its own unique contribution to the emergent whole (Gardner, 1985).


Surveying the Cognitive Hexagon

Slide11.JPG (30570 bytes)Philosophy sits at the apex of the cognitive hexagon. This is, of course, where it all began, with epistemological concerns for the nature of knowledge and knowing, including the debate between nativists and empiricists. Philosophers also gave us the mind-body problem to contend with: how mind, or at least consciousness, could emerge from physical particles interacting in fields of force. Viewing the field in 1985, Howard Gardner asserted that the role of philosophers in cognitive science was merely to ask the questions and check the answers, and they have played that role well.

In the beginning, philosophers addressed these questions using the stock tools of their trade: introspection, reasoning, and, later, linguistic analysis. But more recently, the boundary between philosophical and scientific inquiry has been weakened, and one of the characteristics of contemporary philosophical inquiry into cognition is that philosophers are now interested in the actual results of empirical research -- so much so that some philosophers have argued that, in the new "neurophilosophy", the ordinary mental language of "folk psychology" -- percept, memory, belief, desire, and the like -- will be replaced by the constructs of neuroscience. We'll see.

Slide12.JPG (37625 bytes)For its part, psychology offers cognitive science a wide variety of experimental methods by which we can study cognitive processes empirically. This was true even before the cognitive revolution, where what used to be called "experimental psychology" focused its attention on problems of sensation and perception, learning, and memory (in the form of verbal learning). But in the wake of the cognitive revolution, psychology has added an increasingly sophisticated body of theory concerning how knowledge is acquired and represented in the mind, and how new knowledge can be generated through processes of reasoning, problem-solving, judgment, and decision-making.

But the cognitive perspective has permeated the approach of psychology to even its traditional topics. In the psychology of sensation, for example, the early (and, in my view, proto-behavioristic) emphasis on stimulus intensity and thresholds for sensation has been replaced by the theory of signal detection, which emphasizes the expectations and motives of the observer. In perception, the cognitive view stresses the inherent ambiguity of the stimulus, and the need for the observer to go "beyond the information given" by the stimulus, drawing on knowledge, expectations, and inferences to fill in the gaps, and construct a mental representation of the object or event in the environment. In memory, the injunction by Frederick Bartlett (1932) that "the psychologist, of all people, must not stand in awe of the stimulus" led to a view of the memory trace as similarly incomplete and ambiguous, and remembering as reconstructive activity involving problem-solving, reasoning, and inference. Even in animal learning, the earlier emphasis on the passive formation of associations between stimulus and response was replaced by a view of the behaving organism actively attending to unexpected stimuli, and seeking to predict and control events in its environment. The cognitive revolution began by offering an alternative to psychology, but at this point psychology is thoroughly imbued with the cognitive point of view.

Slide13.JPG (30016 bytes)Similarly, linguistics offers a set of specialized methods for the study of language. Traditionally, language was conceived as a particularly powerful means of communication, but the cognitive revolution in linguistics, initiated by Noam Chomsky, has moved the field far beyond that limited conception. In the first place, language is not just a means of communication: it is also a particularly powerful tool for thought, with words as symbolic representations, and the rules of syntax as a means of combining familiar concepts into entirely new thoughts. By affording us the ability to speak, and communicate, ideas that have never been thought -- much less spoken -- before, Chomsky persuasively argues that the capacity for language is the basis of human freedom.

While Chomsky stressed the strict separation of syntax from semantics (and focused his attention on the former), a movement in "cognitive linguistics" reasserted the communicative function of language, and argued for the priority of meaning, or semantics, over syntax. While Chomskian linguistics thinks of words as symbols whose meaning is constituted by a list of features, cognitive linguistics argues that words derive their meaning from the way they are used in communication: meaning is not given by the word, but rather is a matter of frame and metaphor. Moreover, while Chomskian linguistics thinks of language as following its own set of rules, cognitive linguistics thinks of language as a communication skill not unlike other cognitive skills. Finally, through the "speech act" theories promoted by John Searle and others, we have come to understand that language is not just a form of thinking, but also a form of action by which we can manipulate and transform the world around us, bringing the world in line with our ideas.

Slide14.JPG (36270 bytes)From the beginning, computer science, with its flowcharts of boxes and arrows, provided a means of conceptualizing the components of the mind and how they related to each other. But computer science has also offered the computer program itself as a medium for writing cognitive theory: Herbert Simon has long insisted that you don't really understand a cognitive process unless you have written an operating computer program to simulate it. Many of the contributions of computer revolve around the quest for artificial intelligence, and here I will just point out that there is an distinction between what John Searle has called "Weak AI", in which the computer serves merely as a medium for writing more complicated theories that would be possible verbally, or even mathematically; and "Strong AI", which seeks to duplicate human mental states and processes in a machine made of silicon chips. While Weak AI has been enormously successful, the jury is still out on Strong AI, with critics like Searle pointing out that the mere duplication of input-output relations does not mean that the intervening processes are identical in mind and machine. More recently, we have seen the emergence of what I have called "Pure AI", which tries to get machines to perform "intelligent" tasks without regard for how humans do it. A good example is "Deep Blue" (technically, a refined version known as "Deeper Blue") which in 1997 edged out Gary Kasparov, the reigning world chess champion. It was a stunning accomplishment, the Holy Grail of artificial intelligence; but IBM's software developers had absolutely no interest in simulating how a human grandmaster approaches the game.

Related to Pure AI is the a distinction between what John Haugeland (Haugeland, 1985) called "Good Old Fashioned Artificial Intelligence (GOFAI), based on traditional computer architectures, and the connectionist architectures which have become popular in machine vision and robotics. In GOFAI, knowledge is represented symbolically in strings of 0s and 1s that have a discrete address within the computer's memory; and then these symbols are manipulated by the computer's program. But in connectionist architectures, there is no distinction between symbols and rules: instead, knowledge is represented by patterns of activation distributed over a large number of processing elements. Connectionist architectures are sometimes described as more "neurally plausible" than symbol-and-rule architectures, and from the point of view of Pure AI, they certainly are very powerful systems for learning. But it remains to be determined whether they capture the essence of how the human mind performs its cognitive functions (McClelland & Patterson, 2002a, 2002b; S. Pinker & M. Ullman, 2002; S. Pinker & M. T. Ullman, 2002; Pinker & Ullman, 2003).

Slide15.JPG (29327 bytes)If, as Howard Gardner put it, the job of the philosophers is to ask the questions and check the answers, I suppose that the job of the neuroscientists is to figure out how the brain does it. Beginning with the assumption that the brain is the physical basis of mind -- that, as Steve Kosslyn has put it, "the mind is what the brain does" -- cognitive neuroscientists seek to understand how mental states and processes relate to states and processes in the brain.

Early in the evolution of cognitive science, the neuroscience component was largely focused on the neuron itself, and the analogy between the all-or-none property of neural discharge and the representation of information in strings of 0s and 1s. At higher levels of analysis, though, the search for neural correlates of particular cognitive functions was inhibited by Lashley's (Lashley, 1950) "Law of Mass Action", which asserted that higher cognitive functions were performed by an "association cortex" acting as a whole. But the discovery in the 19th century of Broca's and Wernicke's areas, seemingly specialized for speech and language, already threatened Lashley's law. And beginning in the 1950s, with the neurosurgical patient known as H.M., cognitive neuropsychologists discovered a whole host of discrete brain areas apparently specialized for various functions. The search for such centers has been greatly abetted by the developing of technologies such as PET and fMRI, which for the first time enable researchers to view the brain in action as subjects perform various tasks. This program is not without its controversies -- not least because the interpretation of the brain image is only as good as the experimental methodology used to generate it. But contemporary cognitive neuroscience seems close to fulfilling its promise of revealing the neural mechanisms underling complex cognitive functions. Cognitive neuroscience is now dominated by the doctrine of modularity, which asserts (among other things) that various cognitive functions are performed by dedicated neural centers and systems (Fodor, 1983). And, indeed, it seems that hardly a day goes by without some cognitive neuroscientist reporting that some cognitive task is performed by, or in, some discrete clump of brain tissue.

Slide16.JPG (36186 bytes)The sixth and final point in the Cognitive Hexagon is anthropology -- which, as Boden (Boden, 2006) has pointed out, is the "missing" or "unacknowledged" discipline of cognitive science. Cognitive science programs are full of philosophers, psychologists, linguists, computer scientists, and neuroscientists, but hardly any of them have any anthropologists, and those that do don't have very many. Which is too bad, because at first glance one would think that anthropology would be central to the field: after all, one way to think about culture is as a body of knowledge and belief shared by a group of people, passed through social learning from one generation to the next. Much early cognitive anthropology was stimulated by Western contact with non-western cultures, and an interest in exploring differences in thought processes between members of "developed" and "primitive" cultures. Later, Soviet historicism added the idea that economic and political development would change the way people thought. There was also the "Sapir-Whorf" hypothesis that language, clearly a part of culture, constrains (or at least influences) thought.

Ta great extent, Chomskian linguistics put a damper on cognitive anthropology, by virtue of its assertion that all languages are fundamentally the same. If this were so, then there is no point in seeking ways in which people who speak different languages might also think differently. And there were also empirical problems: it turned out that Eskimos don't really have 7 different words for snow; and if they do, do White, Anglo-Saxon, Protestants skiing at Alpine Meadows probably have at least as many (Martin, 1986; Pullum, 1989). More substantively, for example, despite the fact that different languages have different numbers of color terms, everyone makes the same discriminations among colors. Still, partly under the influence of post-Chomskian cognitive linguistics, we have come to understand that language does shape thought, by providing metaphors by which thought is expressed, and frames by which expressions are interpreted (Eve Sweetser, in her talk, will probably have more to say about this). There may be something to a weak version of the Sapir-Whorf hypothesis after all. Moreover, under the auspices of a revived "cultural psychology", we have begun to understand that there may be cultural diversity in thought processes after all -- for example, Euro-Americans may prefer linear, and East Asians "dialectical", forms of thinking (Nisbett, 2003). Again, much of this work is highly controversial, and in the same way that it might not be exactly true that Eskimos have seven words for snow, it might not be exactly true that "white men can't contextualize" (Shea, 2001).

Another positive development is that there has begun to be a cognitive revolution in social sciences other than anthropology (Turner, 2001). Economics, for example, has moved from the abstract principles that characterize the neoclassical view to a new "behavioral economics" that tries to understand how people actually think about money. And a new breed of "cognitive sociologists" have become interested in such matters as the social organization of knowledge -- how knowledge is distributed within groups, status differences in access to knowledge, and the like (Swidler & Arditi, 1994)); how knowledge is acquired, represented, and used by social organizations and institutions (Zerubavel, 1997); and, indeed, how certain aspects of reality are constructed through collective cognitive activity (Searle, 1995).


What Cognitive Science Is Not

So that, in a nutshell, is what cognitive science is: an interdisciplinary activity dedicated to understanding the acquisition, representation, and use of knowledge. But to further characterize the field, I should also say a few things about what cognitive science is not.

Slide17.JPG (37814 bytes)First, cognitive science is not neuroscience, because there are many varieties of neuroscience that don't have any interest in cognition per se. But cognitive science isn't even cognitive neuroscience. Neuroscience has its proper place in cognitive science, with its concern for the biological substrates of cognitive processes, how information is represented in the brain, what can be learned from brain-damaged patients, and specialized methodologies for brain-imaging. There are those who argue that neuroscience leads cognitive science, because knowledge of the structure of the nervous system will constrain theories of cognition at higher levels. Maybe, although it has to be said that there are no good examples of such constraint in the literature so far (Coltheart, 2006a, 2006b; Hatfield, 1988, 2000; Henson, 2005, 2006; Kihlstrom, 2007a, 2007b). But, in fact, the reverse seems to be true: the proper interpretation of brain function can come only when we have achieved a correct understanding of cognitive function at the psychological level of analysis. Or, as I like to put it: psychology without neuroscience is still psychology, while neuroscience without psychology is just neuroscience.

Slide18.JPG (34452 bytes)Second, cognitive science is not psychology, or even cognitive psychology. Recall that cognitive science was founded in the first place because academic psychology provided no institutional home for the study of cognition. That's all in the past now: contemporary psychology is thoroughly cognitive, almost to a fault. In some sense, psychology is broader than cognitive science, because it encompasses emotion, motivation, and behavior as well as cognition. But in another sense, cognitive science is broader than psychology, because it is a truly interdisciplinary activity in which psychologists work alongside others -- linguists, philosophers, neuroscientists, computer scientists, and anthropologists and other social scientists -- to solve problems of knowledge acquisition, representation, and use.

Slide19.JPG (32336 bytes)Third, at least from my point of view, the "cognitive" in cognitive science is not a euphemism for the mental. Cognitive science began with the insight that, in order to understand behavior, we had to understand the internal structures and processes that mediated between stimulus and response. Some of these structures and processes are cognitive in nature, but cognition isn't all there is to the mind. There's also emotion and motivation. In the wake of the cognitive revolution, some cognitive scientists have gone so far as to assert the hegemony of the cognitive -- that emotion and motivation are entirely derived from cognition, that feelings and desires are, essentially, beliefs about what we feel and desire. Maybe. On the other hand, Kant -- not to mention Plato -- argued that feelings and desires are irreducible -- that they have an ontological status that is independent of knowledge and cognition.

In the present state of the science, that seems like an equally viable hypothesis, and in fact we can now see the emergence of a separate "affective science", or "affective neuroscience", modeled on cognitive science (and cognitive neuroscience), but independent of it (Davidson, 2000; Lane & Nadel, 2000; Panksepp, 1998). A conative science, focused on motivation, cannot be far behind! Cognitive science is best construed as a specialized science of knowledge -- lest it become overbroad. Doubtless, cognitive, emotional, and motivational states interact with each other. Bruner's "New Look" in perception was based on the proposition that emotion and motivation influenced perception and other cognitive processes (J. Bruner, 1992, 1994; J. S. Bruner & Klein, 1960). Some emotional states to appear to flow from cognitive appraisals (Smith & Ellsworth, 1985), and people can regulate their emotions through cognitive transformations (Lazarus, 1991). Those relations deserve study too. But the people who study cognition, emotion, and motivation, all aspects of mental life together, are probably better called psychologists.


What Does Cognitive Science Offer Religion?

So what can cognitive science do for religion -- or, at least, for the study of religion?

Slide20.JPG (20143 bytes)First, the question might be rephrased as: What can cognitive science do to religion. It's pretty clear that cognitive science has provided new intellectual support for the village atheists among us (apologies to Masters, 19916). Previous arguments against religion were based on the problem of evil in the world, or dissections of proofs for the existence of God (I have to say, for myself, that I always thought that St. Anselm's proof was little more than a debater's trick). But cognitive science adds a new dimension to arguments from the outside. For Sigmund Freud religion was an illusion, a product of collective neurosis -- and thus a disorder of emotion (Freud, 1927/1968).  But for Richard Dawkins (Dawkins, 2006), the evolutionary biologist, belief in God is a delusion -- a disorder of cognition.  And Daniel Dennett (Dennett, 2006), the cognitive scientist, shows us how the delusion actually works: you don't believe in God; you just believe you do, as a result of adopting the intentional stance. For Jesse Bering, to take another example, religious belief may have been evolutionarily adaptive, but it is a mistake nonetheless -- an inappropriate generalization of the "theory of mind" by which we try to understand the thoughts and intentions of other people, to find intelligence and intentionality in the natural world as well (Bering, 2006a, 2006b).

Slide21.JPG (30782 bytes)On the positive side, cognitive science would seem to provide a useful theoretical and methodological apparatus to address questions that are central to religion. For example: What is the nature of religious belief? Philosophers and other cognitive scientists generally identify "belief" with any representational mental state, which combines with some proposition to generate what Russell called a propositional attitude. So, when someone says, "I believe in God", how does that differ from saying "I believe that it is raining outside"? Put bluntly: What is knowledge of God knowledge of?

How is knowledge of God represented in the mind? Cognitive science distinguishes among different types of representations, such as perception-based representations, which preserve knowledge about the physical structure of an object but not its meaning; and meaning-based representations, which go beyond physical description to include semantic and conceptual features. Some religions have very rich representations of God (or of gods), others do not. Does the nature of the representation have any consequences for the nature of the belief? Or vice-versa?

How is knowledge of God acquired? Cognitive science offers us two views of this matter, nativism and empiricism, plus a combination of the two. In some religions, God acts in history; in others, not so much: does this difference have consequences for the nature of religious belief? Moreover, some believers have a direct experience of God, whereas for others, religious belief is acquired vicariously, through precept or example. Was there a difference in religious belief between Moses, who actually received the Ten Commandments from God, and the rest of the Israelites, who were simply told about them (Exodus 19)? Or between Paul, who encountered the risen Christ on the road to Damascus (1 Corinthians 15:1-11), and the Corinthians, to whom he related the story?

What is the nature of religious experience itself? Cognitive science may help us to analyze the cognitive components of the experience, in terms of sensation, perception, memory, thought, and language. But cognitive science may not be enough to encompass such a topic, and it may have to be supplemented by affective science (or, at least, the rest of psychology), to understand the emotional dimensions as well. William James clearly thought so, which is why he gave so much space to the affective aspects of religious experience (James, 1902/1985).

Slide22.JPG (21470 bytes)It is important to understand that, in these respects, different religions may require different cognitive analyses. For example, H. Allen Orr (Orr, 2007), expanding on an idea of Philip Kitcher (Kitcher, 2007), has offered a tentative taxonomy of religions: providentialist, which holds that the universe was created by a benevolent God to whom the believer can pray; fundamentalist, asserting the literal truth of sacred texts ; supernaturalist, not so closely linked to literal interpretation; deist, based on the idea that there is a mind at the base of the material universe; spiritual religion, which focuses on ethical behavior and examples of lives rightly lived; and finally, secular humanism, which is uncomfortably close to spiritual religion but abjures any and all deities entirely. As Orr and Kitcher show, the various kinds of religion respond differently to the theory of evolution and other Enlightenment critiques. It may be that the cognitive science of religion will depend on what kind of religion it is a cognitive science of - -that is, what the adherents of that religion actually believe.


What Does Religion Offer Cognitive Science?

Slide23.JPG (25293 bytes)But I don't think that the relationship between cognitive science and religion is a one-way street. Religion can also make a positive contribution to cognitive science, by offering a unique perspective on certain topics. For example, let us return to the nature of religious belief. As I noted earlier, cognitive science usually thinks of belief as an umbrella term for all sorts of cognitive (and other mental) states; but religion reminds us that there is are real phenomenological and epistemological distinctions between believing something, on the one hand, and knowing, or perceiving, or imagining, or remembering, or thinking something, on the other. Beliefs are often defined as convictions in the truth of some statement, independent of, or in the absence of, sufficient evidence. Then there is the distinction between the belief that something is true, and the belief in something. Belief that, in whatever form it takes, always requires some sort of justification. But religious faith is described by St. Paul as the substance of things hoped for, the evidence of things unseen" (Hebrews 1:1). Believing that it's raining is not the same thing as knowing or perceiving that it's raining; but believing in God is not the same thing as believing that it's raining, and believing in God isn't the same thing as believing in Santa Claus, either. Cognitive scientists might get more clarity on these kinds of distinctions if they would undertake a serious, sympathetic inquiry into religious beliefs.

As another example, cognitive science is intensely concerned with the nature and function of consciousness. Altered states of consciousness offer one venue for consciousness, and there has lately been a resurgence of interest in the effects of meditation on consciousness. Much of this research has been focused on Hindu or Buddhist meditative practices, and much of it has been more concerned with physiology than with cognition (I say this as one who spent altogether too much of his junior year in college pestering Shibayama Roshi, abbot of Nanzen-ji, to allow me to slap electrodes on his head while be practiced zazen). Religious scholars remind us that there are meditative practices in Western religious traditions, too, and they may be equally deserving of our attention.  Shibayama's response to my entreaties was to ask whether I would make the same request of the Pope while he said Mass.  This was, I think, my own personal koan -- or maybe it was my own personal keisaku, the stick that Zen masters use to keep novices awake. Contemplative prayer is a part of Jewish, Christian, and Muslim traditions as well -- it is, as St. Teresa reminds us, the way we gain intimate knowledge of God; and I think that our ignorance of these traditions reflects a kind of scientific Orientalism on our part.

Moreover, just as research on cognition and culture would benefit from deeper understanding of the cultures from which the subjects are drawn, research on meditation would benefit from a deeper understanding of the religious context in which the act takes place. Meditation, ripped out of its religious context, may not be the same thing as the same practice in context. 

Slide24.JPG (32169 bytes)It is also important to understand that EEG tracings and brain images are not self-evident; only by knowing what purpose the meditation serves can we really understand what is going on in the brain: It makes a difference whether the goal of meditation is mindlessness or mindfulness, "unconditional loving-kindness and compassion" (Lutz, Greischar, Rawlings, Ricard, & Davidson, 2004, p. 16369) or the enlightened state of satori. If you're going to understand the cognitive science of religion, you've got to get the religion right: and this is no business for amateurs.  In this respect, Shibayama gave me another koan: even before he asked me about the Pope at Mass, he kept asking "What would it mean?".  I didn't have an answer then, but I think I have it now.  The meaning of meditation isn't going to be found in EEG tracings or illuminated pixels.  Rather, those neural measures are given meaning by the cognitive goal of the exercise.  If the cognitive goal of Zen meditation is the de-automatization of everyday thought patterns (Deikman, 1966) -- well, cognitive science knows how to measure that, and it's not with fMRI (see Postscript).  If the goal of Christian prayer is to gain intimate knowledge of God, as St. Teresa said -- well, perhaps not so much.  But the larger point is that the cognitive goal of meditation will differ from one religion to the other.  If you're going to understand the cognitive science of religion, you've got to get the religion right: and this is no business for amateurs.


Points of Departure

There are surely other ways in which cognitive science can contribute to religion, and in which religion can contribute to cognitive science. I may not be religious enough (or, for that matter, enough of a cognitive scientist!) to think of all of them. But I think these points of contact provide a useful starting-place for joint inquiry -- an inquiry that, I think, only makes sense if religious belief, and experience, is taken seriously by cognitive scientists -- and not just taken as something to be dismissed or explained away.



OK, so how do we measure de-automatization?  Back in 1966, when Deikman was writing about this, "de-automatization" was a pretty vague term.  That's all changed now, as "automaticity" has a fairly precise technical meaning in cognitive psychology.  In general, we can say that an automatic process has four features:

inevitable evocation by an environmental stimulus;

incorrigible completion once evoked; 

efficient execution, meaning that it consumes few or no attentional resources; and

parallel processing, meaning little or no interference with, or by, other processes.

066Stroop.jpg (100725 bytes)The classic demonstration of automaticity is the Stroop test, in which subjects are asked to name the colors in which words are printed, ignoring the meaning of the words themselves.  Subjects find this hard to do when the words are color names, and especially when the color names conflict with the colors.  The traditional explanation of this effect is that, for skilled readers, reading words has become automatized -- we just can't help it, and this automatic reading interferes with the task of naming colors.

So, if one of the consequences of meditation is the de-automatization of thought processes, we would expect that meditators would show reduced interference on the Stroop test.  

069AlexanderStroop.jpg (39569 bytes)And that's exactly what Alexander found in a study of Transcendental Meditation.



More systematic research by Wenk also showed that a secular (non-TM) meditation exercise also reduced Stroop interference.

072Wenk1Pre.jpg (53215 bytes) 073Wenk1Post.jpg (52205 bytes) 074Wenk1Change.jpg (54367 bytes)
076Wenk2aPre.jpg (51844 bytes) 077Wenk2aPost.jpg (51720 bytes) 078Wenk2aChange.jpg (50718 bytes)

Wenk also examined the effects of meditation on the generation of category instances.  Initially, she hoped that meditation would lead to a "freeing up" of thought, manifested in a tendency to generate less frequent, more atypical instances.  

In her first attempt, this didn't work out.

080Wenk2bPre.jpg (50981 bytes) 081Wenk2bPost.jpg (51250 bytes) 082Wenk2bChange.jpg (48054 bytes)

However, a later experiment found enhanced the production of atypical instances during a category generation task -- but only when subjects were specifically instructed to produce atypical as opposed to typical instances.

084Wenk3Typical.jpg (49707 bytes) 085Wenk3Atypical.jpg (46964 bytes)

Now here is a contribution of religion to cognitive science: Implicit in the standard concept of automaticity is the idea that automaticity, whether innate or achieved through extensive practice, is permanent.  By contrast, meditation research seems to indicate that automatization can be reversed.

The only fly in the ointment is that neither Alexander nor Wenk employed a meditation exercise that resembles zazen.  Alexander used Transcendental Meditation, and Wenk used another yoga-like exercise that focused on breathing.  Again, you've got to pay attention to the details of the religious discipline and its cognitive goals.  But the larger point is that the cognitive effects of meditation are a place where religion and cognitive science can meet on common ground.



Alexander, C. N., Langer, E. J., Newman, R. I., Changler, H. M., et al. (1989). Transcendental Meditation, mindfulness, and longevity: An experimental study with the elderly. Journal of Personality & Social Psychology, 57(6), 950-964.

Baars, B. J. (1986). The cognitive revolution in psychology. New York: Guilford Press.

Bechtel, W., & Graham, G. (Eds.). (1998). A companion to cognitive science. Malden, Ma.: Blackwell.

Bering, J. M. (2006a). The cognitive psychology of belief in the supernatural. American Scientist, 94(2), 142-149.

Bering, J. M. (2006b). The folk psychology of souls. Behavioral & Brain Sciences, 29, 453-498.

Boden, M. A. (2006). Mind as machine: A history of cognitive science. Oxford: Oxford University Press.

Boring, E. G. (1950). A history of experimental psychology (2nd ed.). New York: Appleton-Century-Crofts.

Bruner, J. (1992). Another look at New Look 1. American Psychologist, 47, 780-783.

Bruner, J. (1994). The view from the heart's eye: A commentary. In P. M. Niedenthal & S. Kitayama (Eds.), The heart's eye: Emotional influences in perception and attention (pp. 269-286). San Diego: Academic Press.

Bruner, J. S., & Klein, G. S. (1960). The function of perceiving: New Look retrospect. In W. Wapner & B. Kaplan (Eds.), Perspectives in psychological theory: Essays in honor of Heinz Werner (pp. 61-77). New York: International Universities Press.

Cohen-Cole, S. (2007). Instituting the science of mind: Intellectual economies and disciplinary exchange at Harvard's Center for Cognitive Studies. British Journal for the History of Science.

Coltheart, M. (2006a). Perhaps functional neuroimaging has not told us about the mind (so far)? Cortex, 42, 422-427.

Coltheart, M. (2006b). What has functional neuroimaging told us about the mind (so far)? Cortex, 42, 323-331.

Davidson, R. J. (2000). Cognitive neuroscience needs affective neuroscience (and vice versa). Brain & Cognition, 42(1), 89-92.

Dawkins, R. (2006). The god delusion. New York: Houghton Mifflin.

Deikman, A.J.  (1966).  De-automatization and the mystic experience.  Psychiatry, 29, 334-348.

Dennett, D. L. (2006). Breaking the spell: Religion as a natural phenomenon. New York: Viking.

Fodor, J. A. (1983). The modularity of the mind. Cambridge, Ma.: MIT Press.

Freud, S. (1927/1968). The Future of an Illusion. In J. Strachey (Ed.), The Standard Edition of the Complete Psychological Works of Sigmund Freud (Vol. 21). London: Hogarth Press.

Gardner, H. (1985). The mind's new science : A history of the cognitive revolution. New York: Basic Books.

Greenwood, J. D. (2003). Wundt, Volkerpsychologie, and Experimental Social Psychology. History of Psychology, 6(1), 70-88.

Hatfield, G. (1988). Neuro-philosophy meets psychology: Reduction, autonomy, and physiological constraints. Cognitive Neuropsychology, 5, 723-746.

Hatfield, G. (2000). The brain's "new" science: Psychology, neurophysiology, and constraint. Philosophy of Science, 67(Proceedings), S388-S403.

Haugeland, J. (1985). Artificial intelligence: The very idea. Cambridge, Ma.: MIT Press.

Henson, R. (2005). What can functional neuroimaging tell the experimental psychologist? Quarterly Journal of Experimental Psychology, 58A, 193-233.

Henson, R. (2006). What has neuropsychology told us about the mind (so far)? Cortex, 42, 387-392.

Hirst, W., & Miller, G. A. (1988). The Making of cognitive science : essays in honor of George A. Miller. Cambridge ; New York: Cambridge University Press.

James, W. (1902/1985). The varieties of religious experience. Cambridge, Ma.: Harvard University Press.

Kihlstrom, J. F. (2007a). Does neuroscience constrain social-psychological theory? [Expanded version]. Retrieved April 25, 2007, from http://socrates.berkeley.edu/~kihlstrm/SPSPDialogue06.htm

Kihlstrom, J. F. (2007b, May). Social neuroscience: The footprints of Phineas Gage. Paper presented at the Keynote address presented at a conference on "The Neural Bases of Social Behavior", Austin, Texas.

Kitcher, P. (2007). Living with Darwin: Evolution, design, and the future of faith. New York: Oxford University Press.

Lane, R. D., & Nadel, L. (Eds.). (2000). The cognitive neuroscience of emotion. New York: Oxford University Press.

Lashley, K. S. (1950). In search of the engram. In Symposia of the Society for Experimental Biology (Vol. 4, pp. 454-482). New York: Cambridge University Press.

Lazarus, R. S. (1991). Cognition and motivation in emotion. American Psychologist, 46, 352-367.

Lutz, A., Greischar, L. L., Rawlings, N. B., Ricard, M., & Davidson, R. J. (2004). Long-term meditators self-induce high-amplitude gamma synchrony during mental practice. Proceedings of the National Academy of Sciences, 101, 16369-16373.

Martin, L. (1986). "Eskimo words for snow": A case study in the genesis and decay of an anthropological example. American Anthropologist, 88, 418-423.

Masters, E. L. (19916). The village atheist. In Spoon River anthology. New York: Macmillan.

McClelland, J. L., & Patterson, K. (2002a). Rules or connections in past-tense inflections: what does the evidence rule out? Trends in Cognitive Sciences, 6(11), 465-472.

McClelland, J. L., & Patterson, K. (2002b). `Words or Rules` cannot exploit the regularity in exceptions. Trends in Cognitive Sciences, 6(11), 464-465.

Nadel, L. (Ed.). (2003). Encyclopedia of cognitive science. London: Macmillan Publishers.

Nisbett, R. (2003). The Geography of Thought : How Asians and Westerners Think Differently...and Why. New York: Free Press.

Orr, H. A. (2007). A religion for Darwinians? [review of Living with Darwin: Evolution, Design, and the Future of Faith by P. Kitcher]. New York Review of Books, 33-35.

Osherson, D. N. (Ed.). (1995). An invitation to cognitive science (2nd ed.). Cambridge, Ma.: MIT Press.

Panksepp, J. (1998). Affective neuroscience: The foundations of human and animal emotions. New York, NY, US: Oxford University Press.

Pinker, S., & Ullman, M. (2002). Combination and structure, not gradedness, is the issue. Trends in Cognitive Sciences, 6(11), 472-474.

Pinker, S., & Ullman, M. T. (2002). The past and future of the past tense. Trends in Cognitive Sciences, 6(11), 456-463.

Pinker, S., & Ullman, M. T. (2003). Beyond one model per phenomenon. Trends in Cognitive Sciences, 7(3), 108-109.

Pullum, G. K. (1989). The great Eskimo vocabulary hoax. Natural Language & Linguistic Theory, 7, 275-281.

Searle, J. R. (1995). The construction of social reality. N.Y.: Free Press.

Shea, C. (2001). White men can't contextualize. Lingua Franca, 11(6).

Smith, C. A., & Ellsworth, P. C. (1985). Patterns of cognitive appraisal in emotion. Journal of Personality & Social Psychology, 48(4), 813-838.

Sobel, C. P. (2001). The cognitive sciences: An interdisciplinary approach. London: Mayfield.

Stillings, N. A., Weisler, S. E., Chase, C. H., Feinstein, M. H., Garfield, J. L., & Rissland, E. L. (1995). Cognitive science: An introduction (2nd ed.). Cambridge, Ma.: MIT Press.

Swidler, A., & Arditi, J. (1994). The new sociology of knowledge. Annual Review of Sociology, 20, 305-329.

Thagard, P. (2005). Mind: An introduction to cognitive science (2nd ed.). Cambridge, Ma.: MIT Press.

Turner, M. (2001). Cognitive Dimensions of Social Science: The Way We Think About Politics, Economics, Law, and Society.  New York: Oxford University Press.

Wenk-Sormaz, H. (2005). Meditation can reduce habitual responding. Advances in Mind-Body Medicine, 21, 33-49.

Wilson, R. A., & Keil, F. C. (Eds.). (2001). MIT Encyclopedia of Cognitive Science. Cambridge, Ma.: MIT Press.

Woodworth, R. S. (1921). Psychology: A study of mental life. New York: Holt.

Wozniak, R. H. (1992). Mind and body: Rene Descartes to William James. Bethesda, Maryland and Washington, D.C.: National Library of Medicine and American Psychological Association.

Zerubavel, E. (1997). Social mindscapes : An invitation to cognitive sociology. Cambridge, Mass.: Harvard University Press.


This page last revised 05/28/2010 03:26:32 PM.