An fMRI study measurin
a
, G
Wiebke Tiede
b
, Irene Tracey
a,b
important for driving top-down pain inhibitory circuits. Using functional magnetic resonance imaging in practicing Catholics and
avowed atheists and agnostics during painful stimulation, here we show the existence of a context-dependent form of analgesia that
intensity of pain. These include diversion of attention
[6,29,35] as well as more high-level cognitive processes
such as placebo-induced analgesia [19,23,32], emotional
detachment [14] or perceived control over pain [26,33].
In contrast to the attentional modulation of pain, these
ain. Published by Elsevier B.V. All rights reserved.
*
Corresponding author. Address: Oxford Centre for Functional
Magnetic Resonance Imaging of the Brain (FMRIB Centre), Depart-
ment of Clinical Neurology, University of Oxford, John Radcliffe
Hospital, Headley Way, Oxford OX3 9DU, UK. Tel.: +44 1865 22488;
fax: +44 1865 222717.
E-mail address: kwiech@fmrib.ox.ac.uk (K. Wiech).
Pain 139 (2009) 40304-3959/$34.00  2008 International Association for the Study of Pwas triggered by the presentation of an image with a religious content but not by the presentation of a non-religious image. As con-
firmed by behavioral data, contemplation of the religious image enabled the religious group to detach themselves from the experi-
ence of pain. Critically, this context-dependent modulation of pain specifically engaged the right VLPFC, whereas group-specific
preferential liking of one of the pictures was associated with activation in the ventral midbrain. We suggest that religious belief might
provide a framework that allows individuals to engage known pain-regulatory brain processes.
 2008 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.
Keywords: Pain; fMRI; Cognitive; Modulation; Prefrontal cortex; Analgesia; Religion
1. Introduction
Religious lore is full of stories of physical pain with-
stood and vanquished through the power of religious
belief. Although such analgesic effects have not yet been
demonstrated in a controlled experimental setting, it is
not implausible that religious states and practices can
have an influence on pain [24,25,30,31]. Indeed, over
the past decade research has demonstrated a wide range
of ways in which top-down processes can modulate thea
Nuffield Department of Anaesthetics, University of Oxford, John Radcliffe Hospital, Headley Way, Oxford OX3 9DU, UK
b
Oxford Centre for Functional Magnetic Resonance Imaging of the Brain (FMRIB Centre), Department of Clinical Neurology,
University of Oxford, John Radcliffe Hospital, Headley Way, Oxford OX3 9DU, UK
c
Ian Ramsey Centre, Theology Faculty, University of Oxford, Bevington Road, Oxford OX2 6NB, UK
d
Psychology and Religion Research Group, Faculty of Divinity, University of Cambridge, West Road, Cambridge CB3 9BS, UK
e
Oxford Uehiro Centre for Practical Ethics, Faculty of Philosophy, University of Oxford, Oxford, UK
f
Oxford Future of Humanity Institute, Faculty of Philosophy, University of Oxford, Oxford, UK
Received 13 April 2008; received in revised form 23 June 2008; accepted 23 July 2008
Abstract
Although religious belief is often claimed to help with physical ailments including pain, it is unclear what psychological and neu-
ral mechanisms underlie the influence of religious belief on pain. By analogy to other top-down processes of pain modulation we
hypothesized that religious belief helps believers reinterpret the emotional significance of pain, leading to emotional detachment
from it. Recent findings on emotion regulation support a role for the right ventrolateral prefrontal cortex (VLPFC), a region alsoby religion as
Katja Wiech
a,b,
*
, Miguel Farias
c,ddoi:10.1016/j.pain.2008.07.030g analgesia enhanced
belief system
uy Kahane
e
, Nicholas Shackel
f
,
www.elsevier.com/locate/pain
67–476

high-level processes induce a reduction in pain by gener-
ating expectations, selecting alternative interpretations
or changing judgments about pain. Functional magnetic
resonance imaging (fMRI) studies indicate that the key
brain area implicated in such high-level pain-modula-
tory effect is the ventrolateral prefrontal cortex, predom-
inantly in the right hemisphere (VLPFC; [14,19,23,26,
32,33]. Although it is unclear what psychological and
neural mechanisms might underlie analgesic effects that
draw on religious belief, it is likely to involve such high-
level processes of pain modulation given that reported
strategies for religious coping include framing the pain-
ful event in terms that allow for its positive appraisal
[4,22]. As of yet, however, there has been little study
of how cultural and religious traditions and practices
might draw on high-level forms of pain modulation to
enable practitioners to deal with pain. We thus set out
to investigate (a) whether religious belief could be shown
to modulate pain in a controlled experimental setting,
and (b) whether such modulation of pain by religious
2. Methods
2.1. Subjects
Twelve religious (8 female, mean age 24.58, range 19–
33) and 12 atheist and agnostic (9 female, mean age
26.17, range 21–34) healthy subjects were included in
the study. Subjects were recruited by advertisements in
University Colleges and Roman Catholic churches in
Oxford. All religious participants were of Roman Cath-
olic denomination, attending mass at least weekly, pray-
ing everyday, and regularly performing other devotional
actions (e.g., taking part in retreats, going to confes-
sion). The control groups were constituted by partici-
pants who reported to be neither religious nor
spiritual. A questionnaire on religious beliefs was given
to all participants to verify the fulfillment of these crite-
ria [2]. All subjects had normal pain thresholds at the
site of stimulus application, no history of neurological
or psychiatric disease, no history of chronic pain, and
n a 2
-relig
duce
ed to
ainfu
ic stim
468 K. Wiech et al. / Pain 139 (2009) 467–476belief is mediated by the right VLPFC, reflecting high-
level cognitive modulation of pain.
Using fMRI we investigated the perception and neural
processing of pain in 12 practicing Catholics and 12
non-religious subjects. Both groups received repetitive
noxious electrical stimulation while they were either pre-
sented with an image of the Virgin Mary (‘‘religious con-
dition”) or a matched image without a religious
connotation (‘‘non-religious condition”; see Fig. 1 and
Section 2). Participants were instructed to look at the pre-
sented image for 30 s prior to the onset and during the
electrical stimulation. At the end of each stimulation per-
iod participants rated the perceived intensity of the
stimulation.
Fig. 1. Study design. We investigated the combination of two factors i
non-religious, factor II (within-subject): CONDITION religious vs. non
image of the Virgin Mary during the painful stimulation in order to in
without religious connotations was shown. Participants were instruct
application of the noxious stimulation. In each trial a series of 20 p
stimulation had stopped the participants rated the intensity of the electr(50 to +50) on a visual analogue scale. At the end of each block a circle apwere free to withdraw from the study at any time. The
study was conducted in accord with the declaration of
Helsinki and approved by the local research Ethics
Committee. All participants gave written informed con-
sent before participating in the study.
2.2. Experimental design and protocol
The study involved a 2  2 factorial design with the
factors GROUP (religious vs. non-religious) and CON-
DITION (religious vs. non-religious; see Fig. 1). The
experiment was divided into four sessions, each lasting
about eight minutes. In both groups the two conditions
(i.e., religious and non-religious context) were presented
 2 factorial design (factor I (between-subject): GROUP religious vs.
ious). In the ‘‘religious” condition participants were presented with an
a religious state. In the ‘‘non-religious” condition a comparable image
focus on the image that was displayed 30 s before and during the
l electric stimuli was applied to the back of the left hand. After the
uli (0–100) and to what an extent they had been affected by the imagepeared on the computer screen that signalled a baseline period of 12 s.