Pain 107 (2004) 16–21
www.elsevier.com/locate/pain
Development of sensitivity to facial expression of pain
Kathleen S. Deyoa, Kenneth M. Prkachina,*, Susan R. Mercerb
a
Department of Psychology, University of Northern British Columbia, 3333 University Way, Prince George, British Columbia, Canada V2N 4Z9
b
University of Otago, Dunedin, Otago, New Zealand
Received 25 October 2002; received in revised form 13 March 2003; accepted 10 June 2003
Abstract
The ability to perceive pain in others is an important human capacity. Its development has not been studied. The present study examined
the development of sensitivity to evidence of pain from childhood to early adulthood. One hundred and thirty-four males and females from
four age groups (5 – 6, 8 – 9, 11 – 12 years and young adult) took part. They judged the amount of pain displayed on videotaped excerpts of the
facial expressions of pain patients. Excerpts were selected to display no pain, some pain and strong pain, based on facial measurements, and
were displayed to participants in a signal-detection paradigm. All participant groups were more sensitive to evidence of strong than some
pain. The ability to detect pain expressions increased across the young, middle and older groups of children, but older children did not differ
from adults. Increasing age was generally associated with increasing sensitivity to more subtle facial signs of pain. The results indicate that
the ability to perceive pain in others is already significantly developed by the ages of five to six, but refinements in the ability continue
through to early adulthood. These findings represent the first description of the development of the ability to perceive pain in others.
Important areas for future research into the neurobiological, personal and social determinants of this ability are highlighted.
q 2003 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.
Keywords: Pain; Expression; Facial expression; Facial action coding system; Development
Suffering produced by pain is often evident on the face of its
victim. The sufferer’s behavior commands attention from
others and can provoke behavior changes, such as offering
assistance, and emotional responses, such as concern,
sympathy, empathy and distress (Prkachin and Craig,
1994). There is evidence that rudiments of the capacity to
recognize distress in and empathize with others are present
from a very young age (Eisenberg and Fabes, 1998). For
example, Martin and Clark (1982) reported that neonates
will cry when exposed to the cries of other human infants
but not to those of other primates or themselves.
Zahn-Waxler et al. (1992) found that behavioral evidence
of concern for others increased between the ages of 14 and
20 months among children exposed to experimental
simulations of distress. There is considerable evidence
that empathic and sympathetic responding are present by the
age of 2 –3 years (Eisenberg, 2000).
Behavioral and affective responses indicative of
empathic or sympathetic responding to another’s pain
* Corresponding author. Tel.: þ 250-960-6633; fax: þ 250-960-5536.
E-mail addresses: [email protected] (K.M. Prkachin); susan.mercer@
stonebow.otago.ac.nz (S.R. Mercer).
presuppose the ability to detect and identify pain in others.
Although it is evident that both adults and children are able
to identify and respond to painful distress in others, little is
known about the development of this capacity. Knowledge
of how this capacity develops may play an important role in
helping us understand not only pain, but phenomena as
varied as empathy and clinical decision making on the one
hand and the perceptual integration of experience and
expression on the other.
A substantial body of research has described facial
changes that occur during pain and their important role in
social communication (Craig et al., 2001; Prkachin and
Craig, 1994). A limited set of changes—principally
contraction of the muscles surrounding the eye and drawing
together of the eyebrows—occurs on the face when a person
is in pain (Prkachin, 1992a). These actions are graded in
intensity (Prkachin and Mercer, 1989) and convey evidence
of the subjective intensity of pain experience (Prkachin
et al., 1994). During painful experiences, homologous
changes occur in the faces of children and newborns, even
those born prematurely (Craig et al., 1993, 1994; Grunau
and Craig, 1987; Grunau et al., 1990; Johnston et al., 1993,
0304-3959/$20.00 q 2003 International Association for the Study of Pain. Published by Elsevier B.V. All rights reserved.
doi:10.1016/S0304-3959(03)00263-X
K.S. Deyo et al. / Pain 107 (2004) 16–21
1995). This suggests that the behavioral capacity to register
pain and its neurological substrates are present from the
earliest moments of sentient existence.
Observers are sensitive to variations in pain expression in
adults (Poole and Craig, 1997; Prkachin et al., 1983;
Prkachin and Craig, 1985; Prkachin, 1992b) and children
(Craig et al., 1988; Hadjistavropoulos et al., 1994, 1997)
although there are documented shortcomings in their
performance (Prkachin et al., 1994, 2001). The capacity to
recognize and take action upon the suffering of others is of
obvious adaptive benefit to the species (Prkachin, 1986,
1997; Williams, in press). Individuals who can perceive
evidence of pain in others are in a position to render them
assistance, thus enhancing their fitness, or to protect
themselves from a threat shared with the afflicted. Thus,
sensitivity to evidence of pain in others should be a nearly
universal human capacity.
Despite this, virtually nothing is known about the
influences upon this capacity. Is it present from an early
age? When does it achieve its zenith? Does it change across
the life-span? There have been, to our knowledge, no
descriptive studies of the development of the capacity to
perceive pain in others. Accordingly, the purpose of the
present study was to collect basic descriptive information
concerning the development of the ability to perceive pain
in others.
Study of the development of this ability presents certain
methodological obstacles common to the study of perceptual processes in children. It is necessary first to construct
procedures sensitive to the ability to perceive pain in others
and to individual and group differences in those abilities. In
addition, the procedures must be sufficiently interesting to
sustain the attention and motivation of the youngest
participants and sufficiently undemanding to avoid fatigue.
Since any study of perception is dependent on the ability to
observe a differential response associated with the stimuli of
interest, the capacity to respond appropriately is a critical
issue. Most often, perceptual sensitivity is indexed by a
verbal response. It follows that the response demands for a
developmental study must be consistent with the verbal and
cognitive capacities of the youngest participants. Moreover,
the response demands must be consistent across the age
ranges under investigation in order for age comparisons to
be valid. In the present study we describe a method for
investigating age differences in the perception of pain in
others. Observers from four age categories: young, middle
and older children and adults judged the pain of others in a
signal-detection paradigm (Swets, 1996).
1. Method
1.1. Participants
One hundred and thirty-four children and adults from
four age groups took part. Young children were 5- and
17
6-year-olds (16 boys, 17 girls), middle children were 8- and
9-year-olds (17 boys, 18 girls) and older children were 11and 12-year-olds (16 boys, 17 girls). A group of young
adults, 17 men, 16 women (age, M ¼ 22, SD ¼ 6), also took
part. The children were recruited from a number of sources:
a participating school, summer day camps, by referral and
from personal contacts. The adults were university students.
1.2. Materials
Participants viewed videotaped excerpts of the facial
expressions of patients undergoing assessment of their
shoulder injuries by active and passive range of motion
tests. The excerpts were sampled from records taken in a
previous study (Prkachin and Mercer, 1989), which had been
coded for the amount of pain expression displayed by the
patients. Pain expression scores were based on measurements
performed using Ekman and Friesen’s (1978) Facial Action
Coding System. All excerpts had been assigned a score that
was the sum of the products of the intensity and duration of
four facial actions—brow lowering, orbit tightening, lipraising/nose-wrinkling and eye closure—that have been
associated with pain (Prkachin, 1992a; Prkachin and Craig,
1994). Excerpts were selected from the Prkachin and Mercer
(1989) archive if they fell into one of three categories defined
by their pain expression scores: no pain, some pain and a lot
of pain. Pain expression scores ranged from 0 to 150
(M ¼ 7:97; SD ¼ 15.23). Excerpts were categorized as no
pain if their pain expression score was 0, some pain if their
score ranged from 1 to 10 and a lot of pain if their scores
exceeded 10. Thirty excerpts were selected from each
category and assembled on a videotape. The excerpts were
randomized in blocks such that each three excerpts contained
one example of every category. The order of presentation of
patients and patient gender was also random.
The excerpts were 2 s in length. They were preceded by a
number, identifying the trial, and followed by 5 s of black
screen. This test series was itself preceded by 12 practice
trials. The total time of the tape was just under 13 min.
1.3. Procedure
Participants viewed the videotape on a portable, 13-inch
television/video cassette player. They sat at a self-selected
distance from the screen. The majority of children (80) were
tested in their own homes. The rest were tested in another
place familiar to them—a friend’s or relative’s home, a
daycare or a day camp. The adults were tested in a room at
the university.
Pilot testing had established that even the youngest
children were capable of making use of a five category
rating scale if the experimenter recorded the responses.
Participants were told that they were to watch each face on
the video and decide if the person they saw hurt. The five
category rating scale was explained. A laminated sheet of
paper showing the five categories (0 ¼ no pain, 1 ¼ maybe
18
K.S. Deyo et al. / Pain 107 (2004) 16–21
pain or cannot tell, 2 ¼ a little amount, 3 ¼ a middle
amount or moderate pain and 4 ¼ a lot of pain) was placed
in front of the participant. Participants viewed each stimulus
and gave their ratings aloud. The ratings were recorded by
the experimenter.
2. Results
Participants’ ratings of the three levels of pain were
analyzed using the methods of signal-detection theory. For
ease of description, the three levels of pain expression
manipulated will be referred to as none, mild and strong. Two
indices of sensitivity, one to mild pain expressions and one to
strong pain expressions were calculated, using the ratings of
no pain as a reference category for each. The conditional
probabilities of using each rating-scale category to refer to
no, mild or strong pain were first calculated. These
probabilities were then accumulated from the highest to the
lowest category, in this manner treating the rating-scale
categories as a series of successively less stringent response
criteria. The measure of sensitivity to pain expression was
P(A) (McNicol, 1972), an estimate of the area under the
receiver – operating – characteristic curve formed when the
cumulative probability of using each category to describe a
‘signal’ (in this case, mild or strong pain expressions) is
represented on the ordinate, while the cumulative probability
of using each category to describe ‘noise’ (in this case, no
pain expression) is represented on the abscissa. Thus, two
measures of the ability to discriminate pain expressions were
calculated, one representing the ability to distinguish mild
pain from noise and one representing the ability to
distinguish strong pain from noise. These values, referred
to as P(A)M and P(A)S could vary between 0 and 1.0. A value
of 0.5 represents chance performance.
An initial analysis was conducted to determine whether
participants were, in fact, able to detect pain expressions at
above-chance levels. Single-sample t-tests were used to
evaluate whether the mean P(A)M and P(A)S value for each
age group differed from the chance value of 0.5. Because
this entailed eight separate comparisons an alpha level of
0.007 was set as the criterion for statistical significance in
order to adjust for Type 1 error. All the statistical tests were
significant (lowest t (32) ¼ 7.04, p , 0:001). Thus, both
mild and strong pain expressions were clearly discriminable
from no pain across all participant age groups.
P(A)M and P(A)S values were then entered as dependent
variables in a 2 (sex) £ 4 (age) £ 2 (intensity) analysis of
variance (ANOVA) with repeated measures on the last
factor. The analysis revealed a significant effect for
intensity, Fð1; 126Þ ¼ 291:91, p , 0:001, h ¼ 0.70. There
was also a significant main effect for age, Fð3; 126Þ ¼ 85:22,
p , 0:001, h ¼ 0.67.
Both effects can be observed in Fig. 1, which shows that
facial expressions of strong pain were clearly more
discriminable than facial expressions of mild pain. Older
participants were also clearly more sensitive to variations in
pain expression. Post-hoc tests revealed that middle children
were more sensitive than young children, while older
children were more sensitive than middle children.
However, the average performance of older children and
adults did not differ significantly.
A further analysis was conducted in order to examine
how participants in the different age groups made use of the
facial cues. Each participant’s pain rating was correlated
with each patient’s scores on the component facial actions
contributing to the pain expression index. Because mouth
opening was related to pain in the original study from
which the pain expression excerpts were sampled (Prkachin
and Mercer, 1989), and because, from observation, it was
Fig. 1. Mean (^ se) discriminability of the mild and strong pain expressions of shoulder pain patients, judged by four age groups: young children (ages 5–6
years, n ¼ 33), middle children (ages 8 –9 years, n ¼ 35), old children (ages 11 –12, n ¼ 33) and adults (Mean age ¼ 22 years, n ¼ 33).
K.S. Deyo et al. / Pain 107 (2004) 16–21
Table 1
Average correlations between component facial actions and participants
pain ratings across four age groups (standard errors in parentheses)
Facial action
Young
children
Middle
children
Older
children
Brow lowering 0.31* (0.02) 0.46* (0.02) 0.52* (0.02)
Orbit tightening 0.17 (0.02) 0.35* (0.02) 0.39* (0.02)
Lip-raising/
0.18 (0.02) 0.28* (0.02) 0.31* (0.02)
nose-wrinkling
Eye closing
0.18 (0.01) 0.33* (0.01) 0.33* (0.01)
Mouth opening 0.29* (0.02) 0.41* (0.02) 0.45* (0.02)
Adults
0.49* (0.02)
0.40* (0.02)
0.31* (0.02)
0.35* (0.01)
0.43* (0.02)
*Average p , 0:01. Note: data were subject to r-to-z transformations
prior to averaging. Values represent z-to-r back-transformations subsequent
to averaging.
a salient feature of the excerpts, the measure of degree of
mouth opening was also included. Individual participants’
action-rating correlations were subjected to r-to-z transformations. A 2 (sex) £ 4 (groups) £ 5 (actions) ANOVA,
with repeated measures on the last factor was performed on
z-transformed r values. This analysis resulted in a
significant main effect for group, Fð3; 126Þ ¼ 44:09,
p , 0:001, h ¼ 0.51, and a significant group – action
interaction, Fð9; 377Þ ¼ 3:10, p , 0:001, h ¼ 0.07. Posthoc analysis of the interaction by Tukey’s B-test indicated
that, for each component action, the youngest participants’
average correlations were significantly lower than those of
the other three age groups, which did not differ. The backtransformed average correlations between each component
facial action and pain ratings for the four separate age
groups are shown in Table 1. It can be seen that, except for
the correlations between orbit tightening, lip-raising/nosewrinkling, eye closing and pain ratings among the youngest
participants, the average correlations were all significant.
For the youngest participants, only two facial actions—
brow lowering and mouth opening—were related to pain
judgements.
3. Discussion
Children as young as 5 –6 years of age are clearly
sensitive both to the occurrence of and quantitative variation
in pain expression. Despite this, the ability to detect facial
evidence of pain in others increases in a linear fashion to the
pre-adolescent years, at which age performance resembles
that of young adults. By the age of 8– 9 years, the relation
between children’s pain judgements and the facial movements that contribute to a pain expression suggests that
children are sensitive to, and make use of, the suite of
changes that occur on the face during pain.
The findings provide evidence of a process of maturation
of the ability to detect pain in others that unfolds over time
and is largely in place by late childhood. There were
significant improvements in the ability to detect both levels
19
of pain expression between the young and the middle
children and the middle and the older children. The
improvement between the young and middle groups was
accompanied by a change in the relationship between their
pain judgements and the facial signals that differentiated the
pain levels. The judgements of the youngest children
reflected primarily two actions—the degree of brow lowering and mouth opening on the patients’ faces. Children
who were, on average, 3 years older, appeared to make
independent use of all the facial changes involved in pain
expression. Thus, the improvement in ability to detect pain
expression over this time appears to reflect the emergence of
differential sensitivity to a more comprehensive array of
information. The fact that children in the older age group
were more sensitive to the different levels of pain expression
than the middle children, suggests that the older children
may employ a more effective algorithm for weighing the
array of information available in the facial displays. Adults,
who were not significantly better at detecting pain than the
older children, also made use of the full array of information
in the processing of their judgements. The findings suggest
that the maturation of sensitivity to pain expression unfolds
sequentially across the age spectrum examined in this study.
It is interesting to note the relative similarity across ages
of the correlations with pain ratings associated with each
facial action. For all participant groups the correlations
associated with brow lowering and mouth opening were of
greatest magnitude and differed only slightly. These appear,
therefore, to be the most salient cues to observers of all ages.
The correlations associated with orbit tightening, eye
closing and nose-wrinkling/lip-raising were likewise comparable across age groups, though lower in magnitude. This
pattern of findings is somewhat curious, since orbit
tightening is the action that has shown the strongest and
most consistent relationship to sufferers’ reports of pain in
empirical studies of the subject (Craig et al., 2001). This
kind of sub-optimal cue utilization has also been shown in
judgement studies employing untrained adult observers
(Prkachin et al., 1994) and adults with varying types of
experience with pain sufferers (Prkachin et al., 2001). This
may not be surprising, given that the facial signs associated
with orbit tightening (and nose-wrinkling/lip-raising) are
more subtle than the other changes (Ekman and Friesen,
1978). Nevertheless, from an adaptational perspective, it is
difficult to explain why observers would be relatively
insensitive to a more reliable indicator.
Although it may not be surprising that sensitivity to
evidence of pain in others improves with age, the findings
nevertheless highlight the importance and the possibility of
understanding the processes that contribute to this development. In addition, the data from the very youngest
participants indicated that they were surprisingly adept at
making discriminations between apparent pain states. The
ability to detect suffering in others is clearly already
functional to a significant degree at the ages of five and six.
Indeed, despite the fact that pilot work had established that
20
K.S. Deyo et al. / Pain 107 (2004) 16–21
children in the youngest age group studied could make use
of the rating scale employed in this study, it is also true that
the use of quantitative ratings is quite a sophisticated
procedure for children of this age. This could have limited
the magnitude of the sensitivity indices obtained in the
present study. With improved methodology, it may be
possible to document sensitivity to evidence of suffering in
others at an even earlier age.
What are the sources of such sensitivity and its
development over time? It is likely that some elements of
the capacity to perceive suffering in others, as a component
of our evolutionary heritage, are innate. A general tendency
to know that others are hurt would clearly confer an
adaptive advantage to the group, insofar as the perceptual
ability is linked to lending assistance or feeling threatened in
situations of peril. It is known that children as young as 4– 7
months of age are sensitive to the occurrence of some facial
expressions of emotion (Labarbera et al., 1976; Nelson and
de Haan, 1996). This suggests that, at a very early age,
children are differentially responsive to social signals that
have implications for their well-being and provides reason
to believe that the ability to perceive some social signals is
either inherent, or requires relatively little in the way of
experience to develop. As an experience that has clear
implications for well-being, whether it is occurring to
oneself or someone else, pain would seem to be in the same
category.
Recent findings from the study of emotion may have
implications here. Pain has several properties in common
with emotions. It is an unpleasant state, it is accompanied by
an expressive signal and it organizes a variety of behaviors
whose goal is to manage it and to facilitate survival.
Consequently, it is not implausible to expect that aspects of
pain expression and the ability to perceive its expression
would be regulated in ways that resemble the regulation of
comparable emotional processes. There is increasing
evidence from neuropsychological and neuroimaging
studies that specific emotional states and the ability to
perceive the expression of comparable emotional states in
others are regulated in comparable brain sites. For example
Adolphs and colleagues have reported that bilateral damage
to the amygdala, a region that has been implicated in the
regulation of fear (LeDoux, 1995), was associated with
impaired ability to detect facial expressions of fear (Adolphs
et al., 1994, 1995). Likewise PET and fMRI studies have
shown enhanced activity in the amygdala among normal
participants processing facial expressions of fear (Morris
et al., 1998; Phillips et al., 1997). Given the parallels in
properties of pain and emotional expression, it may well be
that the ability to perceive pain expression in others is
regulated by specific brain sites and that development of the
capacity would be dependent on development of the
associated neural structures.
No doubt experiential influences play a role in development of the capacity to perceive pain in others. In fact,
recent findings suggest that variations between people in
the nature of their exposure to the pain of others can affect
adults’ abilities to perceive pain expression (Prkachin et al.,
2001). Precisely what kinds of experience may affect
development of the ability to perceive pain expression in the
age range examined in the present study is unclear. Von
Baeyer et al. (1998) have documented the frequent
experience of pain in the natural environments of children.
Such exposure would appear to provide the critical
environmental conditions for learning the signals associated
with pain in others. In a related context, it should be
emphasized that in the present study sensitivity to others’
pain expressions was examined by exposing participants to
the displays of adult patients. It is not unreasonable to think
that aspects of the perception of pain expressions of children
may differ from perception of the expressions of adults.
It seems likely that further study of the development of
the ability to perceive pain in others and of individual
differences in this ability would provide valuable information not just about the natural ecology of pain, but also
about other important human processes. Sensitivity to
suffering is a hallmark of sympathy and empathy. Consequently, learning about the factors that may increase or
decrease such sensitivity would undoubtedly advance
understanding of this important human capacity. Pain
theorists have suggested that responses of members of the
social environment to evidence of suffering may contribute
to the development or maintenance of chronic pain
problems (Fordyce, 1976). Pain-evoked social responses,
such as sympathy or encouraging forbearance presuppose
the ability to perceive variations in pain expression
(Prkachin and Craig, 1994). Thus, studies of the determinants of sensitivity to pain expression are potentially of
great importance to advance understanding of how social
experience may affect pain. This study demonstrates a
method for addressing these and other important issues.
Acknowledgements
This paper is based on a thesis submitted in partial
fulfilment of requirements for the M.Sc. in Psychology at
the University of Northern British Columbia. Some of the
findings were presented at the Annual Meeting of the
Canadian Psychological Association, Edmonton, Alberta,
1998. We gratefully acknowledge the assistance provided
by Glenda Prkachin and Reiko Graham in the preparation of
videotape materials and helpful commentary by Sherry
Beaumont and Judith Lapadat.
References
Adolphs R, Tranel D, Damasio H, Damasio AR. Impaired recognition of
emotion in facial expressions following bilateral damage to the human
amygdala. Nature 1994;372:669 –72.
K.S. Deyo et al. / Pain 107 (2004) 16–21
Adolphs R, Tranel D, Damasio H, Damasio AR. Fear and the human
amygdala. J Neurosci 1995;15:5879 –91.
Craig KD, Grunau RVE, Aquan-Assee J. Judgment of pain in newborns:
facial activity and cry as determinants. Can J Behav Sci 1988;20:442–51.
Craig KD, Hadjistavropoulos HD, Grunau RVE, Whitfield MF. A
comparison of two measures of facial activity during pain in the
newborn child. J Pediatr Psychol 1994;19:305 –18.
Craig KD, Prkachin KM, Grunau RVE. The facial expression of pain. In:
Turk DC, Melzack R, editors. Handbook of pain assessment, 2nd ed.
New York: Guilford; 2001.
Craig KD, Whitfield MF, Grunau RVE, Linton J, Hadjistavropoulos HD.
Pain in the preterm neonate: behavioral and physiological indices. Pain
1993;52:287 –99.
Eisenberg N. Emotion, regulation and moral development. Annu Rev
Psychol 2000;51:665 –97.
Eisenberg N, Fabes RA. Prosocial development. In: Damon W, Eisenberg
N, editors. Handbook of child psychology: social, emotional and
personality development. New York: Wiley; 1998. p. 701 –8.
Ekman P, Frieson WV. Manual for the facial action coding system.
Consulting Psychologists Press, Palo Alto, CA, 1978.
Fordyce WE. Behavioral methods for chronic pain and illness. St. Louis:
Mosby; 1976.
Grunau RVE, Craig KD. Pain expression in neonates: facial action and cry.
Pain 1987;28:395– 410.
Grunau RVE, Johnston CC, Craig KD. Neonatal facial and cry responses to
invasive and non-invasive procedures. Pain 1990;42:295–305.
Hadjistavropoulos HD, Craig KD, Grunau RVE, Whitfield MF. Judging
pain in newborns: facial and cry determinants. J Pediatr Psychol 1994;
19:485–91.
Hadjistavropoulos HD, Craig KD, Grunau RE, Whitfield MF. Judging pain
in infants: behavioral, contextual, and developmental determinants.
Pain 1997;73:319– 24.
Johnston CC, Stevens BJ, Craig KD, Grunau RVE. Developmental changes
in pain expression in premature, fullterm, two, and four month old
infants. Pain 1993;52:201–8.
Johnston CC, Stevens BJ, Yang F, Horton L. Differential response to pain
by very premature neonates. Pain 1995;61:471–9.
LaBarbera JD, Izard CE, Vietze P, Parisi SA. Four- and six-month-old
infants’ visual responses to joy, anger and neutral expressions. Child
Dev 1976;47:535– 8.
LeDoux JE. Emotion: clues from the brain. Annu Rev Psychol 1995;46:
209–35.
Martin GB, Clark RD. Distress crying in neonates: species and peer
specificity. Dev Psychol 1982;18:3–9.
McNicol D. A primer of signal detection theory. London: George Allen &
Unwin 1972.
21
Morris JS, Friston KJ, Buechel C, Frith CD, Young AW, Calder AJ, Dolan
RJ. A neuromodulatory role for the human amygdala in processing
emotional facial expressions. Brain 1998;121:47–57.
Nelson CA, de Haan M. Neural correlates of infants’ visual responsiveness
to facial expression of emotion. Dev Psychobiol 1996;29:577–95.
Phillips ML, Young AW, Senior C, Brammer M, Andrews C, Calder AJ,
Bullmore ET, Perrett DI, Rowland D, Williams SCR, Gray JA, David
AS. A specific neural substrate for perceiving facial expressions of
disgust. Nature 1997;389:495–8.
Poole GD, Craig KD. Judgments of genuine, expressed, and faked facial
expressions of pain. Journal of Personality and Social Psychology 1992;
63:797–805.
Prkachin KM. Pain behavior is not unitary. Behav Brain Sci 1986;9:754 –5.
Prkachin KM. Afterword. In: Rosenberg E, Ekman P, editors. What the face
reveals. Basic and applied studies of spontaneous expression using the
facial action coding system (FACS). New York: Oxford University
Press; 1997. p. 196–7.
Prkachin KM. The consistency of facial expressions of pain: a comparison
across modalities. Pain 1992a;51:297– 306.
Prkachin KM. Dissociating spontaneous and deliberate expressions of pain:
signal detection analyses. Pain 1992b;51:57–65.
Prkachin KM, Craig KD. Influencing nonverbal expressions of pain: signal
detection analyses. Pain 1985;21:399–409.
Prkachin KM, Craig KD. Expressing pain: the communication and
interpretation of facial pain signals. J Nonverbal Behav 1994;19:
191– 205.
Prkachin KM, Berzins S, Mercer SR. Encoding and decoding of pain
expressions: a judgement study. Pain 1994;58:253–9.
Prkachin KM, Currie AN, Craig KD. Judging nonverbal expression of pain.
Can J Behav Sci 1983;15:408–20.
Prkachin KM, Mercer S. Pain expression in patients with shoulder
pathology: validity, coding properties and relation to sickness impact.
Pain 1989;39:257 –65.
Prkachin KM, Solomon PE, Hwang T, Mercer SR. Does experience
influence judgements of pain behaviour? Evidence from relatives of
pain patients and health-care providers. Pain Res Manage 2001;6:
105– 12.
Swets JA. Signal detection theory and ROC analysis in psychology and
diagnosis: collected papers. Mahwah, NJ: Erlbaum; 1996.
Von Baeyer C, Baskerville S, McGrath PJ. Everyday pain in three- to fiveyear-old children in day care. Pain Res Manage 1998;3:111 –6.
Williams AC de C. Facial expression of pain: an evolutionary account.
Behav Brain Sci 2003;25:439 –88.
Zahn-Waxler C, Robinson JL, Emde RN. The development of empathy in
twins. Dev Psychol 1992;28:1038 –47.