PDF hosted at the Radboud Repository of the Radboud University
Nijmegen
The following full text is a publisher's version.
For additional information about this publication click this link.
http://hdl.handle.net/2066/39569
Please be advised that this information was generated on 2017-07-31 and may be subject to
change.
Archive for the psychology of Religion 29 (2007) 67-85
www.brill.nl/arp
From Brain Imaging Religious Experience
to Explaining Religion: A Critique
Nina P. Azari, Ph.D. and 2Marc Slors, Ph.D.
1
1
Department of Psychology, University of Hawaii, Hilo, HI, USA
2
Department of Philosophy, University of Nijmegen, The Netherlands
Abstract
Recent functional neuroimaging data, acquired in studies of religious experience, have been used to
explain and justify religion and its origins. In this paper, we critique the move from describing brain
activity associated with self-reported religious states, to explaining why there is religion at all.
Toward that end, first we review recent neuroimaging findings on religious experience, and show
how those results do not necessarily support a popular notion that religion has a primitive evolutionary origin. Importantly, we call into question an assumption—key to that account of religion—
concerning a conceptual relation between ‘religion’ and ‘religious experience’. Then, we examine the
conditions that must be met in order to explain religion on the basis of brain imaging findings.
Moreover, we list principled reasons to be sceptical of explanations of religion in terms of the neural
underpinnings of experiences. We conclude that the data from neuroimaging studies are not suited
for an explanation of religion.
Keywords
Brain, Religion, Religious Experience, Neuroimaging, Explanation
I. Introduction
Recent developments in human brain functional imaging technologies—such as
positron emission tomography (PET), functional magnetic resonance imaging
(fMRI), and single photon emission tomography (SPECT)—afford the possibility of examining brain activity associated with a wide variety of mental phenomena (Raichle, 1998). Now there are functional neuroimaging reports on
religious experience (Azari et al., 2001b; Newberg et al., 2001). Such studies
have attracted popular attention, refuelling perceived controversy between science and religion (Begely, 2001a, 2001b; Woodward, 2001). Indeed, findings
from recent brain imaging studies of religious experience have been used to draw
broad conclusions about religion as such, in particular, to explain why there is
© Koninklijke Brill NV, Leiden, 2007
ARP 29_f6_65-85.indd 67
DOI: 10.1163/008467207X188630
5/19/07 1:46:36 PM
68
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
religion at all (Newberg & d’Aquili, 2001). Here we ask, how tenable are those
conclusions? Specifically, the aim of this paper is to critically examine the extent
to which functional brain imaging data of religious experience can be used to
explain religion.
II. Brain and Religious Experience
Early and influential studies
Inquiry into the physiological processes associated with religious phenomena is
not new. Indeed, there has been an ongoing debate for more than thirty years
about whether or not, in-between seizures (i.e., the inter-ictal period), patients
with temporal lobe epilepsy (TLE) (i.e., seizure foci in the temporal lobes) show
a distinctive personality disorder (Dewhurst & Beard, 1970; Ramachandran &
Blakeslee, 1998; Tucker et al., 1987), commonly referred to as deepened or hyper
‘religiosity’ (Bear, 1979; Bear & Fedio, 1977). The temporal lobes are part of the
limbic system, which plays an important role in emotional processes, in particular, emotional arousal. Correspondingly, those who have argued for a distinctive
TLE hyper religious personality, hypothesize that hyper-religiosity is rooted in
limbic over-excitation, expressed as a generalized hyper-emotionality to otherwise neutral events (Bear et al., 1982).
There is also a body of literature, according to which mystical and religious
experiences are a specific symptom of Schizophrenia (for a review see especially
Austin, 1998), and/or occur more frequently in persons who suffer from psychotic delusions and hallucinations than in normal, healthy individuals (Stifler
et al., 1993). A recent study, using human brain functional imaging technology,
reported an association between hyper-activation of the left temporal lobe with
occurrence of religious delusions in a single case study of Schizophrenia (Puri et
al., 2001).
A Popular Hypothesis
The findings of studies using particular patient populations (i.e., TLE, Schizophrenia), as well as those using healthy volunteers (Persinger, 1983, 1984), have
led to the notion that the neurophysiological basis of religious phenomena is
fundamentally a matter of an abnormal brain state, localized to the limbic system
(Persinger, 1983, 1987). More broadly, this view hypothesizes that limbic activity is necessary for (any and all) religious experience (cf. Newberg & d’Aquili,
ARP 29_f6_65-85.indd 68
5/19/07 1:46:37 PM
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
69
2001). Indeed, some have suggested that direct stimulation (electrically or magnetically) of the limbic system will artificially generate a religious experience,
even in a person who claims to be non-religious (e.g., an atheist) (Persinger,
1983). Conversely, others have speculated that removing limbic structures will
render someone incapable of having religious experience (Ramachandran &
Blakeslee, 1998).
The idea that religious experience is ‘marked’ by (dysfunctional) brain activity
involving limbic structures has come to be known as the ‘limbic marker hypothesis’ of religious experience ( Joseph, 2001). According to this hypothesis, only if
the limbic system is active during a reported experience, can one conclude that
religious experience has occurred. The absence of limbic activity would mean
that religious experience was not present. In sum, the ‘limbic marker hypothesis’
forwards a conceptualisation of religious experience as rooted in a primitive
brain-reflex response, for which thoughts and beliefs of the experient are of little
(if any) importance.
There has been considerable evidence cited against the TLE-religiosity
hypothesis (Tucker, 1987). In addition, most serious scholars of religion do not
endorse this view of religious experience (Austin, 1998; Glassman, 2002; Hood
et al., 1996; McNamara, 2002; Peterson, 2001, 2002; Teske, 2001). However,
the ‘limbic marker’ hypothesis of religious experience persists, and, in fact has
only become all the more popular as new neuroscientific techniques to study the
live human brain have emerged (e.g., Alper, 2001).
From religious experience to religion: An assumption
The ‘limbic marker hypothesis’ has encouraged efforts to explain the natural origin of religion on the basis of brain activity data acquired in studies of religious
experience. The general idea has been that, if one can make the case that religious
experience is rooted in evolutionarily ‘primitive’ neural circuitry (i.e., involving
‘lower-order’ brain systems that humans share with less-evolved species), then
one can argue that religion also has primitive evolutionary roots, that religion
emerged as part of the natural order. On this basis, then, one could conclude that
there is a scientifically justified naturalistic explanation for religion, and particularly why there is religion at all.1
1
The quest for an evolutionary basis or root to religion and religious experience is not new. See,
Walter H. Capps, Religious Studies: The Making of a Discipline (Minneapolis: Fortress Press, 1995),
p 74ff.
ARP 29_f6_65-85.indd 69
5/19/07 1:46:37 PM
70
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
Key to this line of argument is an assumption concerning a conceptual relation between ‘religious experience’ and ‘religion’. That is, religious experience is
conceptualized as the essence of religion. On this approach, in order to explain
(the naturalistic roots of ) religion, all one needs to do is to explain (the naturalistic roots of ) religious experience.2 Of course, this agenda additionally turns
on there being a primitive (physiological) basis for religious experience in
the first place. Cognitive neuroscience—in particular, human brain functional
imaging—is the singular domain of science, which is expected to deliver best on
this challenge.
Functional neuroimaging findings on religious experience
Recent developments in human brain functional imaging technologies—such as
PET, fMRI, and SPECT—afford the study of the brain function in normal
healthy individuals (Raichle, 1998). Predictably, there is a rapidly growing literature on the neurophysiological processes associated with a wide variety of normal human mental phenomena, such as imagery, arithmetic, abstract problem
solving, and memory recall (Binkofski et al., 2000; Cabeza et al., 1997; Dehaene
et al., 1999; Duncan et al., 2000; Kosslyn et al., 1993). Now there are two recent
neuroimaging reports on religious experience (Azari et al., 2001a,b; Newberg et
al., 2001).3
Azari et al. (2001b) used PET-CBF to study a group of self-identified ‘religious’ subjects, who were Protestant Christians, and a group of control subjects
who were self-identified as ‘non-religious’. Specifically, the subjects in that study
were 12 normal, healthy adults aged 28+/– 5 years (mean +/– standard deviation). All were right-handed, native German speakers. Six subjects (2 women, 4
men) were self-identified as religious, and were members of a Christian ‘Free
Evangelical Fundamentalist Community’. They were teachers at a private Christian secondary school, and had been selected for their teaching posts on the basis
of rigorous faith-based criteria. Each of them reported having had a documented
conversion experience, and interpreted biblical text literally as the word of God.
According to those religious subjects, the first verse of Psalm 23 (i.e., ‘The Lord
is my shepherd . . .’) was essential for each to get into and sustain a religious state,
2
An essentialist-experientialist approach to the study of religion has been especially prominent
in the scientific study of religion, most notably with William James. See, Walter H. Capps, Religious
Studies: The Making of a Discipline (Minneapolis: Fortress Press, 1995).
3
Three earlier studies were designed to examine cerebral blood flow changes during meditative
relaxation, conceptualized in a broader sense than in terms of religious experience (Herzog et al.
1990; Lazar et al. 2000; Lou et al. 1999).
ARP 29_f6_65-85.indd 70
5/19/07 1:46:37 PM
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
71
defined as ‘being in a personal relationship with God as Jesus Christ’. Six subjects
(2 women, 4 men) were self-identified as non-religious, and were students at the
University of Düsseldorf, Germany, studying various subjects in the natural sciences. For those subjects, religion did not play any significant role in their daily
life, and, in fact, they reported feeling somewhat indifferent to religious matters
(i.e., none was a committed atheist, but also none felt strongly about any particular set of religious beliefs or practices). The two groups were matched on age,
gender, and level of education.
Azari et al. employed a self-induction functional neuroimaging paradigm,
which involved asking the subjects to induce in themselves a religious state (as
they understood that to mean) with the help of biblical text that the religious
subjects themselves suggested would be most effective, namely, Psalm 23. Subjects were PET-scanned in six conditions (occurring in a different order for each
subject). For each scan, the subjects were asked to induce in themselves the requisite target state, with the help of particular textual material (i.e., ‘cues’). Directly
before and after each scan, the felt-quality of each target state was assessed objectively, using a standardized questionnaire (called the Positive Affect Negative
Affect Scale, or PANAS; Watson et al., 1988). In addition, after each PET scanning session, the subjects were asked to assess the extent to which each felt he/
she had attained the requisite target state (i.e., religious, happy, neutral; these
self-assessment ratings ranged on a scale of 0-10). The scanning conditions were
as follows: (1) religious-read, (2) religious-recite, (3) happy-read, (4) happyrecite, (5) neutral-read, (6) rest. The textual cues used for the different task conditions were: religious—the first verse of biblical Psalm 23; happy—a well-known
German children’s nursery rhyme; and neutral—instructions on using a phone
card, taken from the Düsseldorf telephone book. Texts were matched on length,
and the rhyme was not associated with music. All subjects were able to recite
from memory both the religious and happy texts at the time of the PET scanning. In the read conditions, the texts were presented on a screen that was visible
to the subjects as they lay on the scanner bed. In the recite conditions, the subjects had their eyes covered, and recited the textual material silently to themselves. In the rest condition, they lay quietly with their eyes covered.
According to their self-assessment ratings, only the religious subjects achieved
the religious state (while reciting the religious text). The functional brain imaging results during their religious state showed a brain activation pattern that corresponded to their individual self-perspectives. Specifically, PET images acquired
in the religious subjects in the religious state showed peak blood flow activation
in the right dorsolateral prefrontal cortex as compared to the non-religious subjects. This activation pattern was observed also in contrast to the happy state and
ARP 29_f6_65-85.indd 71
5/19/07 1:46:37 PM
72
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
the neutral read condition. Other activations in the religious state were the dorsomedial frontal cortex (or, pre-SMA) and a posterior parietal area identified as
the right precuneus. A more recent network analysis of the PET data further
distinguished the non-religious happy emotion state from the religious state,
again highlighting the particular importance of the prefrontal cortex in the latter (Azari et al., 2005).
The prefrontal cortex—a structure whose connectivity patterns were associated with the religious experience—is most recently understood to play an especially important role in particular kinds of social-relational cognitive processes
(e.g., ‘mentalising’ or ‘theory of mind’) (Adolphs, 2003a,b). Specifically, findings
from recent functional imaging studies suggest that widespread connections
involving especially the prefrontal and medial frontal cortex are critical for reasoning processes that are necessary for social interactions, namely, those that
involve ‘mentalising’—i.e., attributing independent mental states to oneself and
others; reflecting on one’s own inner states and reading another’s mind (Calarge
et al., 2003; Fletcher et al., 1995; Goel et al., 1995; Happe 2003; Kampe et al.,
2003; Siegal & Varley, 2002; Vogeley et al., 2001); agency detection (Blakemore
et al., 2003); intentional (vs. incidental) self-processing (Kirchner et al. 2002);
decision-making in social context (Sanfey et al., 2003), and complex emotions
such as empathy and sympathy (Decety & Chaminade 2003; Wicker et al.,
2003).
Importantly, the PET data indicated increased blood flow in the left amygdala
(i.e., a limbic structure) only for the happy emotion; no significant changes in
blood flow involving limbic structures were evident for the religious experience.
This particular finding not only challenges the ‘limbic marker hypothesis’ of religious experience, but, as well, supports the conviction that religious experience
and emotion, while related, are not exactly the same thing (Azari & Birnbacher,
2004).
In sum, these recent PET findings are consistent with the idea that complex
cognition—belief-thought, subjective awareness, learning, memory—are central
to (at least one kind of ) religious experience, and, that limbic activity may not be
necessary for (all kinds of ) religious experience.
In another recent neuroimaging investigation, Newberg et al. (2001) used
SPECT to study Buddhist meditation. They, like Azari et al., employed a selfinduction neuroimaging paradigm, in which they asked their subjects—four
men and four women, self-described as experienced, practicing Tibetan Buddhist meditators—to meditate while their brains were being scanned. The participants were allowed to use incense during the scanning sessions (i.e., ‘cues’).
Brain scan data were collected when the subject indicated that he/she was in the
ARP 29_f6_65-85.indd 72
5/19/07 1:46:38 PM
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
73
most intense part of the meditation. Newberg et al. reported that changed blood
flow in brain areas involved in cognitive processes (i.e. complex visual perception, attention, orientation, and verbal conceptualisation) correlated with the
Buddhist meditation experience (e.g. areas such as the prefrontal cortex and the
posterior parietal lobe, thought to be involved in creating mental representations
of the self and orientation in space). These investigators also did not observe
changes in brain activity in limbic structures during the religious experience of
their Buddhist subjects.
Taken together, the neuroimaging results suggest that complex cognitive processes are central to at least some kinds of religious experience, and thus support
a picture of religious experience as cognitive (Dupre, 1998; Gelphi, 1994; Henry,
1991; McIntosh, 1995; Proudfoot, 1985; Schlitt, 2001; Spilka & McIntosh,
1995; Spilka et al., 1985, 1992, 1996). On this understanding, religious experience is itself structured in terms of past and present beliefs, perceived meaning,
and interpretation of the given experiential context. In this way, recent human
brain imaging findings challenge the alternative, popular view that religious
experience—and religion—is a matter of a pre-cognitive, brain reflex-response
localized to the limbic system—i.e., the ‘limbic marker hypothesis’. Indeed, brain
activity in limbic structures—which is central to a conceptualisation of the biological roots of religious experience (and religion)—simply was not observed in
the neuroimaging results.
Despite this fact, the functional imaging data on religious experience have
been used to argue for the naturalistic basis, and, in fact, evolutionary necessity
of religion. That is, it has been argued that brain imaging data of reported religious experience show that the human brain is hardwired for religious experience, as it is for religion (Alper, 2001; Newberg & d’Aquili, 2001). Religion
serves an evolutionary function (has a naturalistic basis), and recent brain imaging data can directly demonstrate this. How tenable is such a conclusion?
III. Conditions for Explaining Religion using Functional Imaging Data
Emphasising the evolutionary necessity of religion is, of course, not merely an
attempt at explaining why God, or rather belief in God, won’t go away. It is at
least also a way of explaining why belief in God is there in the first place. It would
be odd indeed to explain the sustenance of religion in naturalistic, evolutionary
terms, but not its emergence. If we’re looking for a naturalistic explanation of
religion, it is not enough to explain its sustenance and leave its emergence a mystery. So, like the ‘limbic marker hypothesis’, cognitivist attempts at explaining
ARP 29_f6_65-85.indd 73
5/19/07 1:46:38 PM
74
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
religion in neuroscientific terms must involve attempts to explain the emergence
of religion.
In this section we shall concentrate merely on the question of emergence as
regards neuroscientific explanations of religion. First we determine two conditions, which have to be met by neuroscientific data in order to be suitable for
explaining religion. Next we argue that the limbic marker hypothesis can meet
only one of these demands. Hence it cannot explain the emergence of religion.
After that, the cognitivist alternatives of the hypothesis are put to the test.
Although it may seem at first that at least one of these alternatives, the view of
Newberg and d’Aquili, can avoid the problems with the limbic marker hypothesis
and meet both conditions, we shall argue that this can only be the case on a much
too narrow, and in fact incorrect, understanding of the myths sustaining religion.
What does it take to explain the emergence of religion primarily on the basis of
brain data? Two constraints
If the emergence of religion is to be explained primarily in terms of neuroscientific
data, then these data must meet at least two constraints. First of all the existence
of these data as well as our characterization of them must be independent of an
existing religious tradition. If we have to invoke a religious tradition in order to
claim the existence of these data, or in order to describe or characterize their
peculiarities relevant to the explanation of religion, then obviously these data
cannot be used to explain the emergence of that very religious traditions, on pain
of vicious circularity. Let us call the required independence of the neuroscientific
data from the religious tradition, whose emergence are alleged to explain the
noncircularity constraint.
Apart from the noncircularity constraint there is another (equally obvious)
constraint on neuroscientific data that purport to explain the emergence of a
religious tradition. These data must—ceteris paribus—indicate sufficient conditions for the emergence of such a tradition. Of course the ceteris paribus clause is
important here. Brain properties alone may well be strictly insufficient for the
emergence of religion, since all kinds of further conditions may have to be met
too in order for religion to emerge as a consequence of these properties. Think,
for example, of the existence of a language or of the availability of conditions for
physical survival. As long as these extra conditions can trivially or unproblematically be assumed to be obtained, they can be considered background conditions
for a neuroscientific explanation of the emergence of religion. Thus the
neuroscientific data can be considered ceteris paribus sufficient for the emergence
of religion. We may call this the sufficiency constraint.
ARP 29_f6_65-85.indd 74
5/19/07 1:46:38 PM
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
75
The Limbic marker hypothesis
The ‘limbic marker hypothesis’ and associated theories are about neuroscientific
evidence that appears to meet the noncircularity constraint. The experiences for
which these theories cite neuroscientific evidence are entirely non-cognitive and
can hence be characterized in complete abstraction from any religious tradition.
Moreover, there is no problem whatsoever in imagining these experiences occurring prior to the emergence of institutionalized religion. Add the overly plausible
claim that our current brains are relevantly similar to those of people living
before institutionalized religion emerged, and it is clear that the ‘limbic marker
hypothesis’ can claim to pass our first test. Unfortunately, though, the limbic
marker hypothesis is not confirmed by the neuroimaging studies mentioned,
which does at least diminish its credibility.
However, the real problem with the limbic marker hypothesis, interpreted as
an attempt at explaining the emergence of religion, is that it cannot meet the
sufficiency constraint. The very reason why it can meet the noncircularity constraint rather easily—the non-cognitive character of the claimed neural/mental
precondition for the emergence of religion—is just the problem here. For noncognitive states to give rise to a religion of sorts, they have to be interpreted as
being experiences of some supernatural reality. Now of course this would be easy
in the presence of a (dominant) religious tradition. Given a certain religious tradition, it is no mystery why and how certain non-cognitive emotional states can be
interpreted as being experiences of some supernatural reality described in that tradition. But of course this route would lead to immediate circularity: the religious
tradition would be a precondition for the explanation of its own emergence.
So, the question is how these non-cognitive states can be interpreted as giving
rise to beliefs about some religious reality in the absence of a pre-existing religious tradition.4 This is not a rhetorical question; it is possible to interpret noncognitive (emotional) states as experiences of something or other. It is certainly
4
For the sake of brevity, we shall ignore an extra problem that arises as a result of the fact that
various forms of ‘mental externalism’ are by now the orthodoxy in the philosophy of mind and the
philosophy of the cognitive sciences. Originally, externalism was the thesis that the contents of
mental states can be individuated only by referring also to the world outside of the mind (Putnam
1975), or to the language community in which a thinker is brought up (Burge 1979). Recently,
more radical forms of externalism have been developed, such as Clark and Chamers’ ‘extended
mind’ (Clark and Chalmers 1998), according to which the mind itself is coconstituted by items
outside of the head. On the Burge-form of externalism, it is impossible altogether to assign content
to an experience without drawing on a pre-existing language community. If radical externalism is
true, information about the neural basis of an experience may well be insufficient to determine
content altogether. We shall ignore the issue of externalism here, because all forms of externalism
ARP 29_f6_65-85.indd 75
5/19/07 1:46:38 PM
76
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
imaginable, for instance, that specific emotional states are interpreted as experiences of the presence of some sentient being. The point, however, is that this is
not enough to view such non-cognitive states or emotional experiences as the
sources for the emergence of some religious tradition. For such a tradition is
founded on beliefs. And though interpreting a non-cognitive state as experience
of X is an important step towards the formation of a belief concerning the existence of X, it is not sufficient for the formation of such a belief in general and
certainly not for the kind of belief relevant to the present context: a publicly
shared belief.
Experiences of X do not spontaneously give rise to beliefs that X is the case.
My experience of the book in front of me does not spontaneously lead to a fullblown belief that there is indeed a book in front of me. It may lead to my assuming
there to be a book in front of me, but I only really believe that when, for example,
I reach for it and indeed feel a book-shaped object to be there. In such a case my
visual experience is corroborated by an experience of the same object in a different
sense-modality. Or it may lead to a full blown belief, because someone else that I
observe looking at the same book, reports seeing the same book from a different
angle; i.e. because I can—in the words of Donald Davidson—triangulate my
experience with someone else’s experience of the same object. Corroboration
and triangulation may perhaps not be necessary for the formation of subjective
beliefs. But when it comes to beliefs that are publicly shared in the way that religious beliefs are shared—shaping the identities and lives of millions as well as
shaping society to a significant degree—we move past the realm of subjective
idiosyncrasies. Some extra epistemic (or quasi-epistemic) warrant is required.
Now the question is whether these extra epistemic procedures are available
options in the case of religious experiences of the kind the ‘limbic marker hypothesis’ claims are responsible for the emergence of religion. It does not require a lot
of argument to claim that neither of the procedures mentioned are available
options. There is no other sense-modality by means of which we can experience
the same religious reality (and the ‘limbic marker hypothesis’ does not claim there
is). As for triangulation, this is not possible either. Triangulation requires us to be
able to observe another subject observe or experience the numerically same reality that we are observing or experiencing. And that is clearly impossible as well.
are still contested, and the debate over the viability of externalism is far from over. We would like to
stress, though, that from the viewpoint of the philosophy of mind and the philosophy of the cognitive sciences, explanations of religion based on neuro-scientific evidence presuppose a theory about
the content of mental states—internalism, the negation of externalism—that is by now very much
a minority position.
ARP 29_f6_65-85.indd 76
5/19/07 1:46:39 PM
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
77
Here it can be claimed that corroboration and triangulation are indeed
required for obtaining reliable beliefs only. They are logical demands. But here we
are talking about psychological reality, and more often than not people are somewhat irrational.5 Couldn’t religious beliefs be the result of a rationally indefensible epistemic procedure, to which people are nevertheless prone given their
psychological make-up?
Maybe. However, what is of interest for the alleged neuroscientific explanation of religion is whether or not this specific irrationality can be given a purely
neuroscientific explanation. It seems to us that this cannot be the case. The reason for this is that there are many experiences that are epistemologically on a par
with religious experiences and for which we can find neurological sufficient conditions that nevertheless do not give rise to beliefs of the kind involved in religion. For instance, hallucinations, dreams and experiences of voices heard by
psychotics do not usually give rise to beliefs and practices similar to religion. And
then there are experiences that are epistemologically similar to religious experiences and that do give rise to beliefs, but of a slightly different kind than religious
ones. Experiences of ghosts may be cases in point. These did give rise to beliefs
and belief-systems we know from spiritualism. But spiritualism is a belief-system
that is subtly but significantly different from institutionalized religion. Things
like creation myths, notions of salvation, and purely symbolic (i.e. non-manipulative) rituals, are sometimes present in spiritualist belief-systems. If so, however,
they derive from religion which is very often mixed with spiritualist beliefs. Spiritualism itself does not give us these notions, which can be seen by considering
the fact that there are secular versions of spiritualism that do not contain them.
So, our point is this: experiences that resemble religious experiences in that
the neural substrates are (ideally) known and in their epistemological status, are
nevertheless different from religious experiences in that they do not lead to systems of belief and practice that resemble religion. Our point is not that this
implies that it is impossible to explain how specific experiences that are allegedly
hardwired in our brains can give rise to religion. Our point is that though these
hard-wired experiences may or may not be necessary for the explanation of religion, they can never be sufficient. For such an explanation, neuroscientific data
presumably need to be supplemented with various sociological, psychological
and epistemological considerations that are obviously much too substantial to
count as mere background conditions.
5
This is why e.g. Nick Zangwill’s rational critique of the possibility of religious experience is
largely irrelevant for the present discussion. See Zangwill (2004).
ARP 29_f6_65-85.indd 77
5/19/07 1:46:39 PM
78
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
The upshot of this is that there cannot be a naturalistic explanation of religion
primarily in terms of neuroscientific data. Hence the sufficiency condition is
violated.
The Cognitivist Alternatives
So the limbic marker hypothesis does well with regard to the noncircularity constraint but fails at the sufficiency constraint. Things appear to be significantly
different with the more cognitivist alternatives for the hypothesis that are supported by the findings of the neuroimaging studies discussed above. Prima facie,
these alternatives appear to have no problems with the sufficiency constraint and
serious problems with the noncircularity constraint. But we shall argue that
things are exactly the other way round.
What Newberg et al. and Azari et al. found was that among the areas of the
brain involved in the religious experiences they studied are those regularly associated with cognitive tasks. In other words, these religious experiences are found
to have a cognitive component; roughly: they involve beliefs and concepts.
Hence, these cognitivist studies of religious experiences do much better with
respect to the sufficiency constraint than the limbic marker hypothesis. For this
time there are no particular problems with interpreting a nonconceptual experience in terms such that the experience gives rise to beliefs; the beliefs are already
part of the experience. So let us put the sufficiency constraint to one side for a
moment.
As soon as we ask where these beliefs come from, we can see that there are
serious prima facie difficulties with the noncircularity constraint. It is more than
likely that these concepts and beliefs reflect the religious traditions in which the
subjects were involved. Not only do the descriptions the subjects give of their
own religious experiences betray the traditions in which they stand, but the ways
in which the experiences in these studies are induced also involve clear reference
to pre-existing religious traditions. In the studies by Azari et al., religious experiences are induced through recitation of Psalm 23. In Newberg et al., the subjects
used the meditation methods of Tibetan Buddhism, e.g. using incense that is
associated with religion or religious experiences through tradition. The religious
experiences studied by the experiments by Newberg and Azari, then, clearly
presuppose a pre-existing religious tradition. Therefore, these experiences cannot
be used to explain the emergence of religious traditions on pain of vicious
circularity.
But here’s the catch: neither Azari nor Newberg claim that religious experiences alone can explain religion. Newberg, together with d’Aquili (and unlike
ARP 29_f6_65-85.indd 78
5/19/07 1:46:39 PM
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
79
Azari), does claim that religion can be given a primarily neuroscientific explanation, but this explanation is not based purely on experiences. In fact, in the fourth
chapter of their book, Newberg and d’Aquili claim to give a neuroscientific explanation for the emergence of myths, which they plausibly regard as the foundations of religion and religious traditions. For the sake of the argument we may
presume that such an explanation of myth, if cogent, is sufficient to account—in
neuroscientific terms—for exactly that what is presupposed by the religious
experiences in the studies by Newberg et al and Azari et al. These experiences
may serve to explain the sustenance of religion while their underlying myths may
serve to explain the emergence of religion. Thus, the noncircularity constraint is
not violated.
The sufficiency constraint is not violated, however, if and only if the neuroscientific explanation of myth offered by Newberg and d’Aquili is cogent. Our
point is that it isn’t.
According to Newberg and d’Aquili, the emergence of myths can be explained
in terms of (i) what they term the ‘cognitive imperative’, the irresistible, biologydriven need to comprehend things by means of a cognitive analysis of reality,
(ii) the ability to anticipate and to fear what is anticipated (most importantly
one’s own death), (iii) what they call ‘the causal operator’, i.e. the tendency to
postulate causes for unexplained phenomena, and (iv) what they term ‘the binary
operator’, i.e. the tendency to understand the reality in which we live in terms of
opposing concepts such as ‘light’ and ‘darkness’, ‘heaven’ and ‘hell’, etc. All four
elements, we may assume, have neuroscientific explanations. Elements (i) and
(ii) explain the fact that we humans tend to be concerned with crucial questions
about life and the human condition (e.g. what happens when we die?). Element
(iv) explains the fact that we tend to frame those questions in terms of binary
oppositions (e.g. life and death). And finally, (iii) explains how we find answers
in terms of some ‘hidden’ part of reality—an afterlife, a spiritual world—that
accounts for the manifest features of reality that cause fear and/or anxiety. The
main purpose of these myths, then, according to Newberg and d’Aquili, is to take
away this fear or anxiety.
One problem with this explanation is that many crucially important features
of myths are not accounted for. Most or all religious myths, for instance,
contain stories about the creation of the universe that are such that the structure
of the tribe or society in which the myth functions is justified. A river, for
instance, can be identified with, say, a mythical snake while the various sections
of the river represent the various clans in the tribe who holds the myth and their
interrelations. Another example: peoples are depicted as ‘the chosen’, and are attributed rights and duties accordingly. Yet another: Chiefs of tribes are sometimes
ARP 29_f6_65-85.indd 79
5/19/07 1:46:40 PM
80
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
identified with celestial constellations, which invokes the need to devise cyclical
conceptions of time and rituals in which the chief ‘becomes’ his forefather. Etc. etc.
This sociological aspect of myth—the import of which cannot be underestimated—
is simply ignored by Newberg and d’Aquili.
Another problem is the usually non-logical character of myth. If the main
purpose of myth is to take away fear and anxiety by means of stories portraying a
hidden reality that ‘causes’ manifest reality, why not stick to simpler and more
logical stories (in fact the fictional examples Newberg and d’Aquili give are fairly
logical, but in that respect they do not resemble actual myths).
But the main problem is this: despite the fact that all elements of the explanation for the emergence of myths apply to all normally functioning people, we
must conclude that not all people develop myths or live by already developed
myths. Despite the fact that (i)-(iv) apply to all normally functioning people, we
cannot but observe that for a significant part of the western European and American population at least, God has gone away. There are simply too many atheists
and agnostics at present to consider them not normally functioning. And since
(i)-(iv) will apply to them as well, we must conclude that (i)-(iv) are not sufficient
to explain the emergence of real myths and consequently religion.
So, even if we grant the claim that (i)-(iv) can be given purely neuroscientific
explanations, this is not sufficient to claim that myths can be given a purely
neuroscientific explanation. Hence, even if we grant that the step from myth to
religion is small enough to ignore, religious experiences of the kind involved in
the studies by Newberg et al. and Azari et al. are not explained in purely
neuroscientific terms since they presuppose myths and traditions that are not
themselves explained in purely neuroscientific terms.
In other words, the sufficiency constraint is not met. This result is a consequence of the way in which violating the noncircularity constraint was avoided
(by invoking the idea of a neuroscientific explanation of myth). So, the cognitivist alternatives for the limbic marker hypothesis fail in a similar way to explain
religion in primarily neuroscientific terms: by avoiding the noncircularity constraint, the sufficiency constraint must be violated.
III. Summary and Conclusion
In this paper, we started with a discussion of explanations of religion on the basis
of abnormal brain states, localized to the limbic system—the limbic marker
hypothesis. Next we discussed neuroimaging studies on religious experiences
and concluded that they did not corroborate the limbic marker hypothesis and
ARP 29_f6_65-85.indd 80
5/19/07 1:46:40 PM
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
81
concurring explanations of religion. What these studies suggest is that religious
experiences have a strong cognitive component, something that does not fit with
the hypothesis. But we concluded that at least some of the neuroimaging
researchers still claim that religion can be explained in terms of the neuroscientific
study of religious experience (i.e., functional brain imaging data).
To investigate the viability of that claim, we next articulated the conditions to
be met by such an explanation: (1) the noncircularity constraint and (2) the
sufficiency constraint. We argued that neither the limbic marker hypothesis nor
its cognitivist alternatives can meet the sufficiency constraint due to the different
ways in which they manage to avoid the noncircularity constraint. We conclude,
therefore, the data from functional neuroimaging studies of religious experience
are not suited for an explanation of religion.
Acknowledgments
We would like to thank the referees for their helpful and constructive suggestions.
References
Adolphs, R. 2003a. Cognitive neuroscience of human social behavior. Nature Reviews Neuroscience
4: 165-178.
Adolphs, R. 2003b. Investigating the cognitive neuroscience of social behavior. Neuropsychologia
41: 119-126.
Alper, M. 2001. The ‘God’ Part of the Brain: A Scientific Interpretation of Human Spirituality and
God. New York, NY: Rogue Press.
Austin, J.H. (1998) Zen and the Brain. Boston, MA: MIT Press.
Azari NP, Birnbacher, D. (2004) The role of cognition and feeling in religious experience: An interdisciplinary inquiry. Zygon 39(4): 901-918.
Azari, N.P., Missimer, J., Seitz, R.J. (2005) Religious experience and emotion: Evidence for distinctive cognitive neural patterns. International Journal for the Psychology of Religion 15(4): 263-281.
Azari, N.P., Nickel, J.P., Wunderlich, G., Niedeggen, M., Hefter, H., Tellmann, L., Herzog,
H., Stoerig, P., Birnbacher, D., Seitz, R.J. (2001a) Neural Circuitry of Religious Experience.
In 31st Annual Meeting of the Society for Neuroscience, 1, 382. San Diego, CA: Society for
Neuroscience.
Azari, N.P., Nickel, J.P., Wunderlich, G., Niedeggen, M., Hefter, H., Tellmann, L., Herzog, H., Stoerig,
P., Birnbacher, D., Seitz, R.J. (2001b) Neural Correlates of Religious Experience. European J
Neurosci 13: 1649-52.
ARP 29_f6_65-85.indd 81
5/19/07 1:46:40 PM
82
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
Bear, D., Levin, K., D. Blumer, D., et al. (1982) Interictal Behavior in Hospitalized Temporal Lobe
Epileptics: Relationship to Idiopathic Psychiatric Syndromes. The Journal of Neurology, Neurosurgery, and Psychiatry 45: 481-488.
Bear, D.M. (1979) Temporal Lobe Epilepsy—a Syndrome of Sensory-Limbic Hyperconnection.
Cortex 15: 357-384.
Bear, D.M., Fedio, P. (1977) Quantitative Anaylsis of Interictal Behavior in Temporal Lobe Epilepsy. Archive of Neurology 34: 454-467.
Begley, S. (2001a) Religion and the Brain. Newsweek 137: 50-7.
Begley, S. (2001b) Searching for the God Within. Newsweek 137, no. 5: 59.
Binkofski, F., Amunts, K., Stephan, K.M., Posse, S., Schormann, T., Freund, H.J., Zilles, K., Seitz,
R.J. (2000) Broca’s region subserves imagery of motion: a combined cytoarchitectonic and fMRI
study. Hum Brain Mapp 11: 273-285.
Blakemore, S.J., Boyer, O., Pachot-Clouard, M., Meltzoff, A., Segebarth, C., Decety, J. (2003) The
detection of contingency and animacy from simple animations in the human brain. Cereb. Cortex
13: 837-844.
Burge, T. (1979), “Individualism and the Mental”, Midwest Studies in Philosophy 4: 73-121.
Cabeza, R., Grady, C.L., Nyberg, L., McIntosh, A.R., Tulving, E., Kapur, S., Jennings, J.M.,
Houle, S., Craik, F.I. (1997) Age-related differences in neural activity during memory encoding
and retrieval: a positron emission tomography study. J Neurosci. 17:391-400.
Calarge, C., Andreasen, N.C., O’Leary, D.S. (2003) Visualizing how one brain undersands another:
a PET study of theory of mind. Am. J. Psychiatry 160: 1954-1964.
Capps, Walter H. Religious Studies: The Making of a Discipline. Minneapolis: Fortress Press, 1995.
Clark A. andChalmers D. (1998), “The Extended Mind”, Analysis 58 (1): 7-19.
Decety, J., Chaminade, T. (2003) Neural correlates of feeling sympathy. Neuropsychologia 41:
127-138.
Dehaene, S., Spelke, E., Pinel, P., Stanescu, R., Tsivkin, S. (1999) Sources of mathematical thinking:
behavioral and brain-imaging evidence. Science 284: 970-974.
Dewhurst, K., Beard, A.W. (1970) Sudden Religious Conversions in Temporal Lobe Epilepsy. The
British Journal of Psychiatry 117: 497-507.
Duncan, J., Seitz, R.J., Kolodny, J., Bor, D., Herzog, H., Ahmed, A., Newell, F.N., Emslie, H. (2000)
A Neural Basis for General Intelligence. Science 289: 457-460.
Dupre, L. (1998) Religious Mystery and Rational Reflection: Excursions in the Phenomenology and
Philosophy of Religion. Grand Rapids, MI: Eerdmans Publishing Company.
Fletcher, P.C., Happe, F., Frith, U., Baker, S.C., Dolan, R.J., Frackowiak, R.S.J., Frith, C.D. (1995)
Other minds in the brain: a functional imaging study of “theory of mind” in story comprehension. Cognition 57: 109-128.
Gelphi, D.L. (1994) The Turn to Experience in Contemporary Theology. Mahaw, NJ: Paulist Press.
ARP 29_f6_65-85.indd 82
5/19/07 1:46:40 PM
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
83
Glassman, R.B. (2002) “Miles within Millimeters” and Other Awe-Inspiring Facts About Our
“Mortarboard” Human Cortex. Zygon 37: 255-277.
Goel, V., Grafman, J., Sadato, N., Hallett, M. (1995) Modeling other minds. Neuroreport 6: 17411746.
Grafman, J. & Hendler, J. (1991) Planning and the brain. Brain and Behavioral Science 14: 563564.
Happe, F. (2003) Theory of mind and the self. Ann NY Acad. Sci. 1001: 134-144.
Henry, C.F.H. (1991) Theology, 20th-Century Trends In: From 1900. In New 20th-Century Encyclopedia of Religious Knowledge, ed. James Dixon Douglas, 820-824. Grand Rapids, MI: Baker
Bookhouse Company.
Herzog, H., Lele, V.R., Kuwert, T., Langen, K.J., Kops, E.R., Feinendegen, L.E. (1990) Changed
Pattern of Regional Glucose Metabolism During Yoga Meditative Relaxation. Neuropsychobiology 23: 182-7.
Hood, R.W. Jr., Spilka, B., Hunsberger, B., Gorsuch, R. (eds.) (1996) The Psychology of Religion: An
Empirical Approach. New York, NY: The Guillford Press.
Joseph, R. (2001) The Limbic System and the Soul: Evolution and Neuroanatomy of Religious
Experience. Zygon 36: 105-136.
Kampe, K.K.W., Frith, C.D., Frith, U. (2003) “Hey John”: Signals conveying communicative intention toward the self activate bain regions associated with mentalizing, regardless of modality.
J. Neuroscience 23: 5258-5263.
Kirchner, T.T.J., Brammer, M., Bullmore, E., Simmons, A., Bartels, M., David, A.S. (2002) The
neural correlates of intentional and incidental self processing. Neuropsychologia 40: 683-692.
Kosslyn, S.M., Behrmann, M., Jeannerod , M., (1995) The cognitive neuroscience of mental imagery. Neuropsychologia 33:1335-44.
Lazar, S.W., Bush, G., Gollub, R.L., Fricchione, G.L., Khalsa, G., Benson, H. (2000) Functional
Brain Mapping of the Relaxation Response and Meditation. Neuroreport 11: 1581-5.
Lou, H.C., Kjaer, T.W., Friberg, L., Wildschiodtz, G., Holm, S., Nowak, M. (1999) A 15O-H2O
PET Study of Meditation and the Resting State of Normal Consciousness. Human Brain Mapping 7: 98-105.
McIntosh, D.N. (1995) Religion as Schema, with Implications for the Relation between Religion
and Coping. The International Journal for the Psychology of Religion 5: 1-16.
McNamara, P. (2002) The Motivational Origins of Religious Practices. Zygon 37: 143-160.
Newberg, A., Alavi, A., Baime, M., Pourdehnad, M., J. Santanna, J., d’Aquili, E. (2001) The Measurement of Regional Cerebral Blood Flow During the Complex Cognitive Task of Meditation:
A Preliminary Spect Study. Psychiatry Res 106: 113-22.
Newberg, A., Alavi, A., Baime, M., Pourdehnad, M., Santanna, J., d’Aquili, E. (2001) The Measurement of Regional Cerebral Blood Flow During the Complex Cognitive Task of Meditation: A
Preliminary Spect Study. Psychiatry Research 106: 113-22.
ARP 29_f6_65-85.indd 83
5/19/07 1:46:41 PM
84
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
Newberg, A.B., d’Aquili, E. (2001) Why God Won’t Go Away: Brain Science and the Biology of Belief.
New York. NY: Ballantine Books.
Persinger, M.A. (1983) Religious and Mystical Experiences as Artifacts of Temporal Lobe Function: A General Hypothesis. Percept Mot Skills 57: 1255-1262.
Persinger, M.A. (1984) People Who Report Religious Experiences May Also Display Enhanced
Temporal-Lobe Signs. Percept Mot Skills 58: 963-975.
Persinger, M.A. (1987) Neurophysiological Bases of God Beliefs. New York, NY: Praeger.
Peterson, G.R. (2001) Think Pieces: Religion as Orienting Worldview. Zygon 36: 5-19.
Peterson, G.R. (2002) Thinkpieces: Mysterium Tremendum. Zygon 37: 237-253.
Proudfoot, W. (1985) Religious Experience. Berkeley: University of California Press.
Puri, B.K., Lekh, S.K., Nijran, K.S., Bagary, M.S., Richardson, A.J. (2001) Spect Neuroimaging in
Schizophrenia with Religious Delusions. International Journal of Psychophysiology 40: 143-148.
Putnam, H. (1975), “The Meaning of ‘Meaning’,” in H. Putnam (ed.), Mind, Language and Reality,
Cambridge, Cambridge University Press: 215-71.
Raichle, M.E. (1998) Behind the Scenes of Functional Brain Imaging: A Historical and Physiological Perspective. Proceedings of the National Academy of Sciences 95: 765-772.
Ramachandran, V.S., Blakeslee, S. (1998) Phantoms in the Brain. New York, NY: William Morrow
and Company, Inc.
Sanfey, A.G., Rilling, J.K., Aronson, J.A., Nystrom, L.E., Cohen, J.D. (2003) The neural basis of
economic decision-making in the ultimatum game. Science 300: 1755-1758.
Schlitt, D.M. (2001) Theology and the Experience of God. New York, NY: Peter Lang.
Siegal, M., Varley, R. (2002) Neural systems involved in “theory of mind.” Nat. Rev. Neurosci. 3:
463-471.
Spilka, B., G.A. Brown, G.A., Cassidy, S.A. (1992) The Structure of Religious Mystical Experience
in Relation to Pre-and Postexperience Lifestyles. The International Journal for the Psychology of
Religion 2: 241-257.
Spilka, B., Ladd, K.L., McIntosh, D.N., Milmoe, S. (1996). The Contents of Religious Experience:
The Roles of Expectancy and Desirability. International Journal for the Psychology of Religion 6:
95-105.
Spilka, B., McIntosh, D.N. (1995) Attribution Theory and Religious Experience. In Handbook of
Religious Experience, edited by R.W. Hood, 421-445. Birmingham, AL: Religious Education
Press.
Spilka, B., Shaver, P., Kirkpatrick, L.A. (1985) A General Attribution Theory for the Psychology of
Religion. Journal for the Scientific Study of Religion 24: 1-20.
Stifler, K., Greer, J., Sneck, W., Dovenmuehle, R. (1993) An Empirical Investigation of the Discriminability of Reported Mystical Experiences among Religious Contemplatives, Psychotic inPatients, and Normal Adults. Journal for the Scientific Study of Religion 32: 366-372.
Teske, J.A. (2001) Neuroscience and Spirit: The Genesis of Mind and Spirit. Zygon 36: 93-103.
ARP 29_f6_65-85.indd 84
5/19/07 1:46:41 PM
N. P. Azari, M. Slors / Archive for the psychology of Religion 29 (2007) 67-85
85
Tucker, D.M., Novelly, R.A., Walker, P.J. (1987) Hyperreligiosity in Temporal Lobe Epilepsy:
Redefining the Relationship. Journal of Nervous and Mental Disorders 175: 181-184.
Vogeley, K., P. Bussfeld, A. Newen, S. Herrmann, F. Happe, P. Falkai, W. Maier, N.J. Shah, G.R. Fink,
K. Zilles, K. (2001) Mind reading: Neural mechanisms of theory of mind and self-perspective.
NeuroImage 14: 170-181.
Watson, D., Clark, L.A., Tellegren, A. (1988) Development and Validation of Brief Measures of
Positive and Negative Affect: The Panas Scales. Journal of Personality and Social Psychology 54:
1063-1070.
Wicker, B., Perrett, D.I., Baron-Cohen, S., Decety, J. (2003) Being the target of another’s emotion:
a PET study. Neuropsychologia 41: 139.
Woodward, K.L. (2001) Faith Is More Than a Feeling. Newsweek 137: 58.
Zangwill, N. (2004) The Myth of Religious Experience. Religious Studies 40: 1-22.
ARP 29_f6_65-85.indd 85
5/19/07 1:46:41 PM