December 5, 2014

Cognitive Science, Learning, and ‘Theory of Mind’

As I have responded to Tanya Luhrmann’s book elsewhere and am on record in terms of thinking it makes a major contribution to the cognitive science of religion (CSR) both in terms of its argument and its use of mixed (ethnographic and experimental) methods, I decided that rather than recycling comments that I have already made, it would make much more interesting to give you something of a meta-review of the scientifically-oriented reviews of her book. My initial thought was to focus on the review symposium in Religion, Brain & Behavior from last winter, but with help from Luhrmann, who graciously emailed me some additional reviews, I decided — at the last minute — to expand my scope.

But let me begin by summarizing what I take to be the scientific heart of her study, that is, her effort to offer a scientific answer to the question of how people — in this case, evangelical, charismatically-oriented Christians in Vineyard congregations in Chicago and the Bay Area — come not only to believe that it is possible to talk to God and God will answer but to experience this as actually happening. In her introduction, she explicitly positions her study in relation to some of the research in CSR that stresses the prevalence of belief in gods and other invisible beings, making the case that, while this belief may be widespread, people have to learn to actually hear God and this learning process is (1) difficult, (2) more difficult for some than others, and (3) when it occurs actually changes how people process information.

Each of these claims required different methods. The first, documenting the learning process, was primarily an ethnographic task, which she accomplished through two years of intensive involvement not just with the congregations but with small groups in which people learned to pray that were attached to the churches in both Chicago and the Bay Area. She discovered that some found it easier to learn to hear god than others through participant observation as well. There were the “prayer warriors” at one end of the spectrum and the poor guy, who was raised in the Vineyard church, who desperately wanted to hear god, but never did, at the other extreme. But Luhrmann didn’t just observe these differences. When they started to become apparent, she had them fill out several psychological scales, some of which they liked better than others. Their responses on a scale that reports their tendency to lose themselves in something — to become absorbed — correlated with their ability to learn to hear god. That was an interesting discovery that suggested that individual differences made a difference in how easily people learned to hear god, but it didn’t say if the learning process itself — the religious practices and spiritual disciplines changed how people processed their perceptions and sensations. To test this, she designed an ingenious field experiment in which she gave people iPods with tapes that they had to listen to for 30 minutes six days a week for four weeks (202-210). The experimental group got a modified version of Ignatius’ spiritual exercises in which they were asked to place themselves in a biblical scene and interact with the characters. The control group got a series of Luke Timothy Johnson’s lectures on the New Testament. They were given a series of cognitive tasks before and after the experiment to see if the imaginative exercises made a difference in how they processed information, e.g., resulted in their seeing mental images more vividly. She described these as “training effects” and there were connections between the training effects and individual differences on the absorption scale.

Luhrmann took these measurable changes in how they were processing information as suggesting that they were learning a new theory of mind. Here’s the key passage from the introduction:

In effect, people train the mind in such a way that they experience part of their mind as the presence of God. They learn to interpret the familiar experience of their own minds and bodies as not being own at all–but God’s. They learn to identify some thoughts as God’s voice, some images as God’s suggestions, some sensations as God’s touch or the response to his nearness. … To do this, they need to develop a new theory of mind. That phrase theory of mind — has been used to describe the way a child learns to understand that other people have different beliefs and goals and intensions (Luhrmann, 2012, xxi-xxii).

This insight led to an expanded project, which she launched shortly after the book went to press and is still under way, in which she is comparing charismatic Christians in India (Chennai) and Ghana (Accra) to see if she finds local variations in theory of mind, that — she is hypothesizing — would suggest that culture influences the inferences we make about other people’s minds.

From a CSR perspective, this is actually her most controversial claim. Controversial, because theory of mind, which refers to our capacity to attribute mental states to ourselves and others is a central topic in developmental psychology and ethology, the study of other animals, where researchers are asking how this capacity develops in humans and whether it is present in other animals. What makes her claim controversial is that most developmental psychologists think this is a universal human capacity. There are other lines of research that question this and argue that “the standard psychological model for theory of mind is distinctively Euro-American.” Luhrmann is thus using her Vineyard research to raise the possibility that there are different theories of mind — local theories of mind — and asking whether these cultural differences make a fundamental difference in the way people draw inferences about another person’s intentions, feelings, and so on (Luhrmann, 2014, 83).

In the reviews I have looked at so far, reviewers did not address this controversial point head-on, but instead addressed it indirectly. Thus, some in the RBB symposium found her approach “too cognitive” – too dependent on an “information processing approach” — and advocated approaching the problem in light of research on embodied cognition (Strawn & Brown 2014), they and another reviewers thought she ought to pay more attention to explaining differences in terms of attachment processes (Strawn & Brown 2014, Sandage 2014). Others highlighted the role of emotion and still others the need to attend to motivation (Alcorta 2014, 54).

Whether her approach is “too cognitive” depends on how cognitive is defined. There is a narrow and a broader sense in which the term is used. Here I quote John Tooby & Leda Cosmides (1992, 65) on the definitional issues: “Some researchers use it in a narrow sense, to refer to so-called ‘higher mental’ processes, such as reasoning, as distinct from other psychological processes, such as ’emotion’ or ‘motivation’; that is, to refer to a concept that corresponds more or less to the folk notion of reasoning while in a calm frame of mind. In contrast, we [Tooby & Cosmides] are using the word cognitive in a different and more standard [cognitive science] sense. … We use … cognitive and information processing to refer to a language or level of analysis that can be used to precisely describe any psychological process: reasoning, emotion, motivation, and motor control can all be described in cognitive terms, whether the processes that give rise to them are conscious or unconscious, simple or complex. In cognitive science, the term mind refers to an information-processing description of the functioning of an organism’s brain.” So a cognitive science approach, as they and many others understand it, is a language for describing how we process all the incoming information available to us from our bodies and the entire environment in which we are situated – social, cultural, natural — in functional terms, that is in terms of what we do with it.

In terms of this understanding, what many of the reviewers distinguish as “not cognitive”, i.e., bodily sensations, emotions, and motivations, all can and should be considered within a cognitive, information processing approach as either input to be processed or output of the processing. Attachment processes are one of the ways of processing information. So in my mind, most of these reviewers missed the key question, which I think has to do with learning. If Luhrmann wants her work to speak to a CS audience, which I think she does, we need to be able to think about how Vineyard folks were learning — more or less well — in cognitive terms.

Pascal Boyer’s review for the HAU: Journal of Ethnographic Theory symposium basically does this, which to my mind makes it the most important — and illuminating — of the science-oriented reviews. Pascal Boyer is the author of Religion Explained and has been a leading figure in terms of CSR. His review firmly positions Luhrmann’s work within a cognitive, information processes perspective. Boyer picks up on Luhrmann’s central point: that people have to work hard to learn to hear God. He makes the obvious point that believing in god — however fervently — is not the same thing as feeling you are in the presence of and communicating with god. He points out that many anthropologists have glossed over this distinction, referring all the time to what the people they are studying believe and then offering this as an explanation for their behavior. Boyer generalized from Luhrmann’s research to suggest that most people hold their beliefs in a more tentative fashion, as a conjecture, that they have not necessarily, or perhaps only fleetingly, experienced first-hand as “real” in the sense that Luhrmann uses the term.

Boyer links this distinction between believing and experiencing to a basic cognitive science distinction between “intuitions,” which are normally processed rapidly and unconsciously, and “reflective beliefs” or “reflective processing” or “metacognition.” Most of what we find out about people’s beliefs are “reflective beliefs.” We process most bodily sensations, perceptions, and emotions intuitively without any reflection on the process at all. As he points out — and this is the key link — “religious representations are potentially much more compelling, attention-grabbing, and memorable when they are associated with intuitive content, preferably with perceptions” (Boyer 2013, 353). In other words, through slow learning processes we can learn to associate reflective beliefs with intuitively processed sensations, perceptions, and emotions. Doing so allows us to experience the reflective beliefs in a way that we would not otherwise be able to do. So the point here is that from an information processing perspective, the evangelical’s Luhrmann studied are learning to connect their conscious, reflective belief in god with intuitive, unconsciously processed emotions, perceptions, and sensations in a way that allows them to distinguish some of their inner thoughts as “theirs” and others as “god’s”.

So, let me tie this back into the theory of mind (ToM) issue. In her response to the RBB symposium, Luhrmann referred to a meeting of anthropologists and psychologists that she arranged at Stanford to see whether they had sufficient evidence to argue for “local theories of mind.” They came up with at least six such theories, one of them “the Euro-American modern supernaturalist theory of mind” (Luhrmann 2014, 82-83). In light of Boyer’s distinctions, what Luhrmann characterizes as local theories of mind are best understood as reflective or folk theories of mind. They are the beliefs that we and others have about how mind, body, and world are related. These beliefs may or may not have any relation to the evolved, intuitive, unconscious way that we intuit what is going on in the minds of others. In her RBB response, Luhrmann (2014, 83) acknowledges that there are a range of competing scientific views of ToM and cites two in her discussion – Scholl & Leslie (1999) and Astuti (2012) – that question her approach, stressing that theories about how something works may or may not tell us anything about how they actually work.

Thus, in Boyer’s terms, Luhrmann is asking whether reflective ToMs can influence intuitive ToM.   In asking that, we need to specify what we mean by influence. I think it is much more likely that reflective ToM can influence what inferences we draw about other people’s minds, but not so likely that it influences how we draw inferences about other people’s minds. By calling what people believe about the mind, body, world relationship “theory of mind,” Luhrmann risks over-associating these reflective beliefs with the intuitive process and overlooking the possibility that these beliefs about the relationship between mind, body, and world may come to seem real to people through association with other intuitive processes. So I think we need to consider the possibility that people can attach reflective/folk ToM to a range of sensations, perceptions, and emotions. Then the question we would need to ask would be how do people learn to experience THIS particular folk ToM, including our Euro-American secular one, as real?



Alcorta, Candace S. 2014. Modes of knowing: how kataphatic practice impacts our brains and behaviors. Book Symposium. Religion, Brain & Behavior 4 (1): 49-56.

Astuti, Rita. 2012. Some after dinner thoughts on theory of mind. Anthropology of this century 3 available in LSE Research Online at:

Boyer, Pascal. 2013. Why ‘belief’ is hard work: Implications of Tanya Luhrmann’s When God talks back. HAU: Journal of Ethnographic Theory 3 (3): 349-57.

Luhrmann, T. M. 2012. When God talks back. New York: Knopf.

Luhrmann, T. M. 2014. Response: Knowing God, attentional learning, and the local theory of mind. Symposium. Religion, Brain & Behavior 4 (1): 78-90.

Sandage, Steven J. 2014. Attachment theory, relational spirituality, and varieties of evangelicals. Book Symposium. Religion, Brain & Behavior 4 (1): 59-65.

Scholl, Brian J. and Alan J. Leslie. 1999. Modularity, development, and ‘theory of mind.’ Mind & Language 14 (1), 131-153.

Strawn, Brad D. and Warren S. Brown. 2014. Living with evangelical paradoxes. Book Symposium. Religion, Brain & Behavior 4 (1): 65-72.

Tooby, John and Leda Cosmides. 1992. The psychological foundations of culture. In J. H. Barkow, L. Cosmides, and J. Tooby, eds. The Adapted Mind, pp. 19-136. New York: Oxford University Press.


2 replies to “Cognitive Science, Learning, and ‘Theory of Mind’

  1. Richard Saville-Smith

    Hi Kevin, Nice piece.

    Here’s a link to Premack and Woodruff’s original 1978 paper ‘Does the Chimpanzee have a Theory of Mind ‘

    And there’s a handy 30 year follow up –

    There’s a lot more in these pieces about Chimpanzees than there is about Theory of Mind. My understanding is that child psychologists took up ToM, then autism researchers, then schizophrenia researchers (Sprong 2007). The tendency in this genealogy is to address the capacity to theorise about what ‘the other’ thinks, so there’s less about a ToM a person holds about their own mind – no doubt children, autism and schizophrenia made this a difficult research area. It seems to me that there are questions about a ToM in relation to others and a ToM in relation to self – they seem very different. But hey that’s not my concern.

    It seems to me that if ToM is addressed from the point of view of a believer they will not only imagine God in a culturally ‘appropriate’ form (inevitably) but they will naturally have a ToM which is self-fulfilling of their own expectations. From this very simple place it would seem obvious that the expectation of God speaking to the actuality of God speaking is a small leap which is well within the power of the mind to make. It’s possible to get all fancy about Dissociative Identity Disorder (DID) but the American Psychiatric Association are pretty clear that DID does not need to be a Disorder but as they limit their concern to the wrong side of distress, then there’s a whole territory of non-distressed dissociation in which the Conversation with God can operate, without anyone getting hurt.

    I think I’m saying what you’re saying but with different words.

    But you might also be intersted in a book by Ivan Leuder – Against Theory of Mind which is a bit of a corrective to all the fawning enthusiasm for the latest jargons!

    All the best



  2. Richard Saville-Smith

    Excuse me, I forgot and should add that there is a rather cute concept called “excess theory of mind”


Join the discussion

Your email address will not be published. Required fields are marked *