The Nature of Facial Expression Recognition Deficits Following
Orbitofrontal Cortex Damage
Megan L. Willis
Australian Catholic University, Sydney, Australia and
Macquarie University
Romina Palermo
ARC Centre of Excellence in Cognition and Its Disorders, Perth,
Australia and University of Western Australia
Ky McGrillen
Royal Prince Alfred Hospital, Sydney, Australia
Laurie Miller
ARC Centre of Excellence in Cognition and Its Disorders,
Sydney, Australia, and Royal Prince Alfred Hospital, Sydney,
Australia, and Central Medical School, University of Sydney
Objective: Orbitofrontal cortex (OFC) damage has been associated with facial expression recognition
deficits in some, but not all, previous studies. The pattern of performance of a group of patients with OFC
damage was assessed across a series of facial expression recognition tasks. We aimed to determine
whether some tasks were more sensitive at detecting deficits than others. Method: Seven patients with
damage to the OFC, 6 control patients with frontal lesions that spared the OFC, and a group of healthy
controls completed a series of facial expression recognition tasks including 2 labeling tasks and 2
matching tasks. Results: The OFC patient group demonstrated impaired labeling of negative facial
expressions (i.e., anger, disgust, fear, and sadness) shown for a short time (500 ms) relative to the
comparison groups. When facial expressions were shown for a longer time (5,000 ms), the OFC patient
group’s performance did not differ significantly from either comparison group. The OFC patient group
was impaired when matching subtle, low intensity negative facial expressions, but not when matching
high intensity, prototypical facial expressions. Conclusions: The pattern of performance across tasks
revealed that only certain facial expression recognition tasks appear to be sufficiently sensitive to detect
deficits in patients with OFC damage. These findings have important implications for the assessment of
facial expression recognition difficulties in patients with OFC damage and more broadly, for special
populations.
Keywords: facial expression recognition, emotion, orbitofrontal cortex, prefrontal cortex, frontal lobes
Patients who have sustained damage to the orbitofrontal
cortex (OFC), a brain region that occupies the ventral surface of
the frontal lobes (Kringelbach & Rolls, 2004), often perform
normally on standard neuropsychological batteries yet exhibit
abnormalities in emotional and social functioning, such as
frequently making social faux pas (Zald & Andreotti, 2010).
One factor that may be associated with their poor emotional and
social functioning is impaired facial expression recognition.
Impaired facial expression recognition is typical in other dis-
orders involving abnormal social functioning, such as schizo-
phrenia and autism spectrum disorders (e.g., Hall et al., 2004;
Humphreys, Minshew, Leonard, & Behrmann, 2007). The OFC
also is thought to be an important neural region in the facial
expression recognition network (Adolphs, 2002). Thus, the
extent to which damage to the OFC impairs facial expression
recognition has been of particular research interest. However,
the evidence is not consistent.
Some patient studies have shown that OFC damage is associated
with an impaired ability to categorize or rate the intensity of facial
expressions (e.g., Blair & Cipolotti, 2000; Dal Monte et al., 2013;
Heberlein, Padon, Gillihan, Farah, & Fellows, 2008; Hornak,
Rolls, & Wade, 1996; Marinkovic, Trebon, Chauvel, & Halgren,
2000). In these instances where deficits have been observed, they
typically reflect difficulty recognizing negative emotions (e.g.,
sadness, anger) as opposed to positive (e.g., happy) emotions (see
Heberlein et al., 2008; Zald & Andreotti, 2010, for reviews).
However, a number of other studies have failed to observe any
evidence of facial expression recognition deficits in patients who
have sustained damage to the OFC (e.g., Adolphs, Damasio,
Tranel, Cooper, & Damasio, 2000; Beer, Heerey, Keltner, Scabini,
& Knight, 2003; Hornak et al., 2003; Willis, Palermo, Burke,
McGrillen, & Miller, 2010). Explaining these conflicting findings
has been complicated by the divergent methodologies employed
This article was published Online First March 3, 2014.
Megan L. Willis, School of Psychology, Australian Catholic University,
Sydney, Australia and Department of Cognitive Science, Macquarie Uni-
versity, Sydney, Australia; Romina Palermo, ARC Centre of Excellence in
Cognition and Its Disorders, Perth, Australia and School of Psychology,
University of Western Australia, Perth, Australia; Ky McGrillen, Radiology
Department, Royal Prince Alfred Hospital, Sydney, Australia; Laurie
Miller, ARC Centre of Excellence in Cognition and Its Disorders, and
Neuropsychology Unit, Royal Prince Alfred Hospital, and Central Medical
School, University of Sydney, Sydney, Australia.
Correspondence concerning this article should be addressed to Megan
Willis, School of Psychology, Australian Catholic University, Mount St
Mary Campus, 25a Barker Road, Strathfield, NSW, 2135, Australia.
E-mail: dr.megan.willis@gmail.com
This document is copyrighted by the American Psychological Association or one of its allied publishers.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
Neuropsychology © 2014 American Psychological Association
2014, Vol. 28, No. 4, 613– 623 0894-4105/14/$12.00 DOI: 10.1037/neu0000059
613