Electrocortical Processing of Food and Emotional Pictures in Anorexia
Nervosa and Bulimia Nervosa
JENS BLECHERT ,PHD, BERND FEIGE,PHD, ANDREAS JOOS, MD, ALMUT ZEECK, MD, AND BRUNNA TUSCHEN-CAFFIER,PHD
Objective: To compare the electrocortical processing of food pictures in participants with anorexia nervosa (n = 21), bulimia nervosa
(n = 22), and healthy controls (HCs) (n = 32) by measuring the early posterior negativity, an event-related potential that reflects stimulus
salience and selective attention. Methods: We exposed these three groups to a rapid stream of high- and low-calorie food pictures,
as well as standard emotional and neutral pictures. Results: Event-related potentials in the time range of 220 milliseconds to
310 milliseconds on posterior electrodes differed between groups: patients with eating disorders showed facilitated processing of both
high- and low-calorie food pictures relative to neutral pictures, whereas HC participants did so only for the high-calorie pictures.
Subjective palatability of the pictures was rated highest by patients with anorexia nervosa, followed by the HC and bulimia nervosa
groups. Conclusions: Patients with eating disorders show a generalized attentional bias for food images, regardless of caloric value.
This might explain the persistent preoccupation with food in these individuals. Key words: anorexia nervosa, bulimia nervosa, early
posterior negativity, event-related potentials, food pictures, attentional bias.
ANOVA = analysis of variance; AN = anorexia nervosa; BN = bulimia
nervosa; EDs = eating disorders; DSM-IV = Diagnostic and Statistical
Manual of Mental Disorders, Fourth Edition; EPN = early posterior
negativity; EEG = electroencephalography; ERPs = event-related
potentials; fMRI = functional magnetic resonance imaging; high-cal =
high-calorie; IAPS = international affective picture system; LPP = late
positive potential; low-cal = low-calorie; RSVP = rapid serial visual
presentation.
INTRODUCTION
I
n Western cultures, people usually have access to an abun-
dance of all types of food, most of which are affordable and
readily consumed. At the same time, the mainstream media has
idealized a slim body shape for women, which has in part
become the cultural norm. In anorexia nervosa (AN) and bu-
limia nervosa (BN), striving for a lean body shape by restricting
food intake has become a superordinate goal. Although most
nonpurging type AN patients are highly rigid with respect
to food intake, purging type AN patients and BN patients
chronically restrain eating but have intermittent binge eating
attacks, as described in the Diagnostic and Statistical Manual
of Mental Disorders, Fourth Edition (DSM-IV). As a result of
these abnormal eating patterns, patients with these eating dis-
orders (EDs) experience periods of food deprivation and star-
vation, interspersed by repletion.
Food is a highly salient biological stimulus category due to
its relevance for survival and its inherently rewarding nature. It
is, therefore, unsurprising that the brain is specifically geared to
process this stimulus class. For example, food deprivation leads
to increased visual attention toward food cues (1) and gives rise
to increased blood-oxygen dependent levels in inferotemporal
regions (amygdala, parahippocampal gyrus, anterior fusiform
gyrus), as well as in medial obitofrontal, medial prefrontal,
insular, and striatal regions (2,3). In patients with EDs, altered
blood-oxygen dependent levels-responses to food pictures were
reported in prefrontal and posterior cortical areas (4Y6) and,
thus, in regions implicated in cognitive control and sensory
processing. Patients with EDs also show specific attentional
biases in dot-probe tasks: They direct their attention toward
high-calorie (high-cal) food items and away from low-calorie
(low-cal) food items (7,8).
In sum, convergent evidence suggests that food cues are
processed differently in EDs. However, for a functional inter-
pretation of these findings, the time scale of stimulus is im-
portant: Attentional capture by motivationally salient stimuli
occurs during very brief and particularly early stages of stim-
ulus processing, which poses methodological problems for dot-
probe and functional magnetic resonance imaging (fMRI)
studies. In general, later phases of stimulus processing might
reflect cognitive regulation processes to a higher degree than
earlier processes control (9Y11). Due to their excellent temporal
resolution, event-related potentials (ERPs) can map precisely
the temporal dynamics of attentional processes from early
through late stages of stimulus processing. To our knowledge,
despite their clear relevance to attentional processing of food
cues, no study has yet investigated ERPs to food cues in EDs.
However, it would be important to know whether the results
from attentional bias and fMRI studies generalize to brain
electric responses.
Recent ERP evidence (12,13) suggested that caloric value
of food is reflected in the ERP already 120 milliseconds to
160 milliseconds after stimulus onset. Slightly later, in the
200-millisecond to 300-millisecond time range, the early pos-
terior negativity (EPN)Va negative, occipital-parietal deflec-
tion of the ERPsVemerges, which is higher for emotional than
for neutral pictures (14). An enhanced EPN also emerges for
food pictures, and this ‘‘food-EPN’’ has recently been shown to
be sensitive to the effect of food deprivation (15). Functionally,
the EPN is interpreted as a correlate of attention allocation (14).
The EPN is most frequently observed in rapid serial visual
presentation (RSVP) tasks: Pictures are presented with high
frequency (e.g., three pictures per second) and without per-
ceivable interstimulus interval. It is assumed that this rapid
visual stream places a high load on the visual system, which
leads to an efficient allocation of processing resources to the
Psychosomatic Medicine 73:415Y421 (2011) 415
0033-3174/11/7305Y0415
Copyright * 2011 by the American Psychosomatic Society
From the Department of Clinical Psychology and Psychotherapy (J.B., B.T.-C.),
Institute for Psychology; Department of Psychiatry and Psychotherapy (B.F.) and
Department of Psychosomatic Medicine and Psychotherapy (A.J., A.Z.), Uni-
versity of Freiburg, Freiburg, Germany.
Address correspondence and reprint requests to Jens Blechert, PhD,
Hellbrunnerstr. 34, 5020 Salzburg, Austria. E-mail: jens.blechert)gmail.com
Jens Blechert is currently at the Psychology Institute, Stanford University,
Stanford, California.
This research was supported, in part, by Grant TU78-6-1 from the German
Research Foundation (B.T.-C.).
Received for publication August 13, 2010; revision received November 19,
2010.
DOI: 10.1097/PSY.0b013e318211b871
Copyright © 2011 by the American Psychosomatic Society. Unauthorized reproduction of this article is prohibited.