Feeling Hangry? When Hunger Is Conceptualized as Emotion

Abstract

Many people feel emotional when hungry-or "hangry"-yet little research explores the psychological mechanisms underlying such states. Guided by psychological constructionist and affect misattribution theories, we propose that hunger alone is insufficient for feeling hangry. Rather, we hypothesize that people experience hunger as emotional when they conceptualize their affective state as negative, high arousal emotions specifically in a negative context. Studies 1 and 2 use a cognitive measure (the affect misattribution procedure; Payne, Hall, Cameron, & Bishara, 2010) to demonstrate that hunger shifts affective perceptions in negative but not neutral or positive contexts. Study 3 uses a laboratory-based experiment to demonstrate that hunger causes individuals to experience negative emotions and to negatively judge a researcher, but only when participants are not aware that they are conceptualizing their affective state as emotions. Implications for emotion theory, health, and embodied contributions to perception are discussed.

Full text

Feeling Hangry? When Hunger Is Conceptualized as Emotion
Jennifer K. MacCormack and Kristen A. Lindquist
University of North Carolina at Chapel Hill
Many people feel emotional when hungry—or “hangry”—yet little research explores the psychological
mechanisms underlying such states. Guided by psychological constructionist and affect misattribution
theories, we propose that hunger alone is insufficient for feeling hangry. Rather, we hypothesize that
people experience hunger as emotional when they conceptualize their affective state as negative, high
arousal emotions specifically in a negative context. Studies 1 and 2 use a cognitive measure (the affect
misattribution procedure; Payne, Hall, Cameron, & Bishara, 2010) to demonstrate that hunger shifts
affective perceptions in negative but not neutral or positive contexts. Study 3 uses a laboratory-based
experiment to demonstrate that hunger causes individuals to experience negative emotions and to
negatively judge a researcher, but only when participants are not aware that they are conceptualizing their
affective state as emotions. Implications for emotion theory, health, and embodied contributions to
perception are discussed.
Keywords: hunger, emotion, embodiment, psychological construction, affect misattribution
Supplemental materials: http://dx.doi.org/10.1037/emo0000422.supp
Empirical evidence demonstrates that emotions impact every
aspect of our waking lives, from visual perception to decision-
making and interpersonal processes (e.g., Balcetis & Dunning,
2006; Barrett & Bar, 2009; Keltner, Ellsworth, & Edwards, 1993;
Lerner, Small, & Loewenstein, 2004; Loewenstein, 2000; Van
Kleef, 2009). Less research, however, examines the impact of
states such as hunger on perceptions, decisions, and interpersonal
processes. Yet people appear to be at least implicitly aware of the
fact that hunger impacts their emotions—the idea that hunger can
impact emotional experiences and behaviors is captured in the
colloquial expression hangry, defined by the Oxford Dictionary as
feeling “bad tempered or irritable as a result of hunger” (Hangry,
2015).
A small body of scientific research affirms that hunger-induced
emotionality or feeling “hangry” is more than mere colloquialism.
For example, individuals who have not eaten (i.e., in a glucose-
depleted state) tend to be more impulsive, punitive, and aggressive
(e.g., Anderberg et al., 2016; Benton, 2002; Bushman, Dewall,
Pond, & Hanus, 2014; Denson, Pedersen, Friese, Hahm, & Rob-
erts, 2011; Symmonds, Emmanuel, Drew, Batterham, & Dolan,
2010; Williams, Pizarro, Ariely, & Weinberg, 2016). Other liter-
ature links hunger to negative mood (e.g., Hepburn, Deary, &
Frier, 1992, 1994; Hermanns, Kubiak, Kulzer, & Haak, 2003,
Hermanns et al., 2007; Taylor & Rachman, 1988). Yet the psy-
chological mechanisms by which people become emotional when
hungry, or “hangry,” remain little understood. The purpose of the
present studies is to begin to identify the psychological mecha-
nisms of hunger-induced emotional states.
Potential Mechanisms of Hunger’s Impact on
Emotions, Judgments, and Behavior
One common assumption, both in folk and experimental psy-
chology, is that hunger impacts emotions, judgments, and behav-
iors because it impairs self-regulation. In this view, hunger re-
leases the constraints that typically keep people from feeling
unbridled emotions, making impulsive judgments, or aggressing
against others (e.g., Bushman et al., 2014; DeWall, Deckman,
Gailliot, & Bushman, 2011; DeWall, Pond, & Bushman, 2010).
Until recently, much research on self-regulation was guided by the
“regulation as muscle” analogy, which hypothesizes that self-
control fails when biological resources such as glucose are de-
pleted (Baumeister, 2003, 2014; Gailliot & Baumeister, 2007;
Gailliot et al., 2007; Muraven & Baumeister, 2000; Vohs et al.,
2014).
This regulatory depletion hypothesis was first inspired by work
demonstrating that mental effort can deplete blood glucose (Fair-
clough & Houston, 2004; Hall & Brown, 1979). Thus, it is as-
sumed that negative, high arousal emotions or outbursts of aggres-
sion when hungry occur because individuals cannot regulate their
feelings without sufficient blood glucose (e.g., DeWall et al.,
2011). However, the regulatory depletion hypothesis has been
critiqued in recent years following failed replications and mixed
findings (e.g., Carter, Kofler, Forster, & McCullough, 2015; Job,
Walton, Bernecker, & Dweck, 2013; Kurzban, 2010; Miles et al.,
2016; Vadillo, Gold, & Osman, 2016; see review in Inzlicht,
Schmeichel, & Macrae, 2014). Moreover, the underlying biologi-
cal premise may be unfounded, as it is unlikely that short-term
shifts in cognitive exertion alter blood glucose levels substantially
Jennifer K. MacCormack and Kristen A. Lindquist, Department of
Psychology and Neuroscience, University of North Carolina at Chapel Hill.
Correspondence concerning this article should be addressed to Jennifer
K. MacCormack, Department of Psychology and Neuroscience, University
of North Carolina at Chapel Hill, 325 Davie Hall, Chapel Hill, NC 27599.
E-mail: jkmaccor@unc.edu
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.
Emotion
© 2018 American Psychological Association 2018, Vol. 1, No. 999, 000
1528-3542/18/$12.00 http://dx.doi.org/10.1037/emo0000422
1

in the central nervous system (e.g., Coker & Kjaer, 2005; Peters et
al., 2004).
An alternate hypothesis is that feeling “hangry” is psychologi-
cally constructed (Lindquist & Barrett, 2008). Psychological con-
structionist theories propose that all mental events, including emo-
tions, arise from the dynamic co-action of domain-general
psychological processes. Specifically, instances of emotion and
other mental events occur when the brain uses prior experiences
and category knowledge to conceptualize or predict the meaning of
affective representations from the body (i.e., “core affect” ranging
from pleasantness-unpleasantness and activated-deactivated sensa-
tions, e.g., Barrett, 2017; Clore & Ortony, 2013; Cunningham,
Dunfeld, & Stillman, 2013; Lindquist, 2013; Russell, 2003;
Shaked & Clore, 2017). Research testing a psychological construc-
tionist hypothesis tends to focus on shifts in affect caused by
external factors (e.g., a situational event), yet ongoing changes in
the body related to maintaining allostasis (i.e., homeostatic pro-
cesses such as blood sugar, inflammation, etc.) also shift affect and
have consequences for downstream experiences, judgments, and
behaviors (MacCormack & Lindquist, 2017).
Indeed, the physiology underlying hunger is consistent with the
idea that hunger impacts core affect. For instance, when blood
sugar drops, ghrelin, the metabolic hormone that signals hunger,
triggers a cascade of hormones, such as cortisol, that act on the
sympathetic nervous system, in turn inducing unpleasant, highly
arousing affective bodily changes (Christensen, Alberti, & Brands-
borg, 1975; Corrall, Frier, Davidson, Hopkins, & French, 1983;
Cryer, 1999; Heller, Macdonald, Herbert, & Tattersall, 1987;
Marks & Rose, 1981). Moreover, brain regions such as the anterior
cingulate cortex, insula, and amygdala show increased activation
during hunger (Chen, Papies, & Barsalou, 2016; Dagher, 2009;
Porubská et al., 2006; Tataranni et al., 1999) but also are more
generally activated during affect and emotion (e.g., Lindquist,
Satpute, Wager, Weber, & Barrett, 2016; Lindquist, Wager, Kober,
Bliss-Moreau, & Barrett, 2012). These findings suggest that there
may be similar neural processes involved in these two seemingly
distinct states, at least at the level of gross anatomical brain
activity.
Building on psychological constructionist theory and the afore-
mentioned physiological findings, we hypothesize that people ex-
perience instances of greater emotionality when hunger-induced
affect is conceptualized as emotions in a given context (e.g., a
negative situation). In this case, hungry people would be more
likely to experience negative, high arousal emotions and engage in
more antisocial interpersonal behaviors than they otherwise might
when satiated in the same context. Notably, despite the colloquial
implication that “hanger” is about anger specifically, we do not
expect these effects to be unique to anger; rather we expect that
hungry individuals can construct any negative, high arousal emo-
tional state, such as feeling irritable, stressed, and so forth, de-
pending on how the context guides such conceptualizations for that
individual.
The Construction of Hunger Into Emotion
A rich theoretical history in psychology supports the psycho-
logical constructionist hypothesis that people experience emotions
when they make bodily changes meaningful in the present context.
Schachter and Singer (1962) classically demonstrated that individ-
uals are more likely to engage in affiliative behavior when given a
shot of epinephrine in the presence of a jovial as opposed to angry
stranger. Zillmann (1971) demonstrated that experimentally induc-
ing higher levels of physiological arousal through exposure to
sexual stimuli or exercise increased aggressive behavior. Dutton
and Aron (1974) found that male participants were more likely to
experience physical attraction when they encountered an attractive
female as opposed to male research assistant after crossing a
rickety, arousal-inducing suspension bridge versus a low, stable
bridge. Schwarz and Clore (1983) found that transient external
conditions (such as bad weather) shifted individuals’ affective
states and influenced subsequent judgments of life satisfaction—
but only when individuals were unaware of the source of their
affective feelings.
The above literature highlights the power of context, but also
awareness of one’s state (or lack thereof) in the construction of
emotions from bodily states such as physiological arousal. Psy-
chological constructionist approaches do not presume that individ-
uals have conscious access to the brain’s construction of mental
experience (e.g., Barrett, 2017; Barrett, Ochsner, & Gross, 2007).
However, what an individual attends to in a given context could
play a powerful role in shifting and updating the brain’s current
predictions about the meaning of incoming stimuli. As such, we
hypothesized that instances in which people “misattribute” the
meaning of their affective state to a different source is a subset of
psychologically constructed phenomena and that in the case of
feeling “hangry,” people draw on the external situation to make
meaning of their hunger-induced affect to construct a variety of
negative emotion instances. We reasoned that as in misattribution
theory, hunger-induced negative, high arousal emotions might be
all the more likely to occur when people’s attention is directed
away from emotions and focused instead on the external circum-
stance (e.g., the person who just insulted me; the traffic jam, etc.).
Across three studies, we test the hypothesis that “hanger,” in the
form of negative, highly aroused emotions, is constructed when
people make meaning of their hunger-induced affect as the expe-
rience of negative emotions.
The Present Studies
1
Studies 1 and 2 used a cognitive tool called the affect misattri-
bution procedure (AMP; Payne, Cheng, Govorun, & Stewart,
2005; Payne et al., 2010) to establish whether hunger increases
individuals’ tendency to judge an ambiguous stimulus as negative
when that stimulus was experienced in a negative context. Study 1
tested this hypothesis by manipulating negative versus neutral
contexts before participants judged the meaning of ambiguous
stimuli, whereas Study 2 further manipulated positive versus neg-
ative versus neutral contexts to demonstrate that hunger’s impact
on judgments of ambiguous stimuli is specific to negative contexts,
but not positive or neutral contexts.
Study 3 extended Studies 1–2 in several ways. First, it was a
laboratory-based experiment that directly assessed how hunger
influences self-reported emotions and interpersonal judgments.
1
All studies were approved by the University of North Carolina at
Chapel Hill Institutional Review Board and were conducted in accordance
with APA ethical conduct of research with human subjects (IRB#s 13–
3050 and 15–3169).
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.
2MACCORMACK AND LINDQUIST

Study 3 also manipulated hunger versus satiation to better assess
the causal effect of hunger. Drawing from the affect misattribution
literature, Study 3 further manipulated emotional awareness. We
hypothesized that hungry individuals who were unaware of mak-
ing meaning of their affective state as emotional would be more
likely to experience “hanger” than individuals who were hungry
and aware of making meaning of their affective state as emotional
or any individuals who were satiated. In particular, we predicted
that individuals who were hungry and unaware of making meaning
of their state as emotional would be the most likely to report
unpleasant, highly aroused emotions and harsher interpersonal
judgments in a negative context.
Importantly, we did not expect participants to exclusively en-
dorse feeling “anger” because negative situations can cause all
manner of unpleasant emotions depending on how individuals
make meaning of the context (Ellsworth & Scherer, 2003; Kirk-
land & Cunningham, 2012; Lindquist & Barrett, 2008). Further-
more, from a psychometric standpoint, people rarely report expe-
riencing just one negative emotion when given multiple options.
Instead many individuals endorse multiple negative emotion ad-
jectives, thus communicating what those adjectives share in com-
mon (i.e., that they feel generally negative; Barrett & Fossum,
2001; Feldman, 1995). Thus, although the colloquial word “hanger”
implies that people feel specifically angry when hungry, we ex-
pected people to report multiple negative, high arousal emotions as
a result of hunger. No previous literature has examined hunger’s
relation to discrete emotions above and beyond general unpleas-
antness, so this provided a first study to identify which emotions
people are more likely to report when experiencing hunger in a
negative social context.
Finally, Study 3 supported a psychological constructionist the-
ory of hunger’s impact on emotions and interpersonal behaviors by
ruling out the alternate hypothesis that hunger leads to emotion
merely via depleted self-control (e.g., Bushman et al., 2014; De-
Wall et al., 2011).
Study 1
In Study 1, we first sought to establish that hungrier people
would perceive ambiguous stimuli as more negative, but only
when those stimuli were perceived in a negative context. Hunger
induces negative, high arousal affect (e.g., Cryer, 1999), but we
hypothesized that people’s degree of hunger would only influence
their perceptions of ambiguous stimuli when their negative, high
arousal affect could be made meaningful in the presence of an
affect-congruent (i.e., negative) context.
To test these hypotheses, we used the AMP (Payne et al., 2005,
2010). The AMP is a cognitive tool that is commonly used to
measure implicit attitudes (e.g., pleasant and unpleasant reactions
to stimuli; see Payne & Lundberg, 2014). The AMP achieves this
goal by assessing the extent to which a person misattributes the
meaning of an initial stimulus (e.g., an unpleasant picture; a picture
of an opposite-race face) to an ambiguous Chinese pictograph that
does not itself have meaning to the participant. In this context, the
AMP is useful for examining the psychological processes under-
lying “hanger” because it measures the implicit process whereby
an individual perceives the affect induced by one source (hunger
interacting with the context of the affective picture) as caused by
another (an ambiguous Chinese pictograph). The extent to which
participants rate the ambiguous Chinese pictographs as more neg-
ative or positive following a negative or positive image is thus an
index of their degree of affect misattribution. In Study 1, we
hypothesized that hunger would interact with negative images,
such that participants who were hungrier would experience the
ambiguous pictographs as even more negative when preceded by a
negative, but not neutral context.
Method
Participants. Two-hundred and 50 Mechanical Turk workers
from the United States participated in the study for monetary
compensation. As the AMP uses Chinese pictographs for ambig-
uous stimuli, four participants were excluded from analyses be-
cause they reported either a familiarity or fluency with Mandarin
Chinese. Twenty-five participants failed attention checks and were
excluded from subsequent analyses; another three participants had
computer issues (e.g., the task froze) and were unable to complete
the study, leading to a final sample of 218 participants (46%
female; M
age
⫽35, SD
age
⫽10.41, 18- to 71-years-old). The
sample size was determined ahead of time based on an a priori
power analysis and data were not analyzed until data collection
was complete.
Although no prior work has examined the interaction of hunger
with context using implicit cognitive tasks such as the AMP, we
estimated that there would be moderate-to-small effect sizes for
our main effects (e.g., ␤⫽.3) and interactions (e.g., ␤⫽.15), as
observed in Payne et al. (2010) and Lee, Lindquist, and Payne
(2017). Given the hierarchical, partially within-subjects nature of
the study design, we planned a priori to use multilevel modeling.
Power analyses for multilevel modeling are based on power sim-
ulations, which suggest that with a Level 1 sample (trials) greater
than 30 nested within a Level 2 sample (participants) greater than
40, we would have 90% power to observe an effect (p⬍.05–.01;
see simulations in Scherbaum & Ferreter, 2009). As we were not
manipulating hunger, and we expected data loss due to the online
sample we used (Thomas & Clifford, 2017), we aimed for well
above this sample size including Level 1 n⫽40 trials, Level 2 n⫽
250 participants.
Procedure. After consent, participants were oriented to the
AMP in Qualtrics. Study 1 used a mixed model design, with the
negative or neutral affective context preceding a Chinese picto-
graph as the within-subjects factor and self-reported hunger as the
continuous between-subjects factor. On all trials, participants first
saw a “context” image followed by an ambiguous Chinese picto-
graph. As is typical of the AMP, participants were instructed that
they would see affective images before each Chinese pictograph,
and that they should try their best to ignore the affective image,
instead focusing on their judgment of the pictograph. The manip-
ulated context in Study 1 was a negative, highly arousing versus
neutral context.
The AMP was implemented in Qualtrics using the QRTEngine
(Barnhoorn, Haasnoot, Bocanegra, & van Steenbergen, 2015). All
participants first completed a practice block of eight trials before
proceeding to the 40 experimental trials. Trials were presented in
counterbalanced blocks based on image type. For each AMP trial
(see Figure 1), participants saw a fixation cross on the center of the
screen for 125 ms, then the negative versus neutral image for 75
ms, followed by a gray visual noise mask for 125 ms and the
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.
3
CONCEPTUALIZING HUNGER AS EMOTION

randomly assigned pictograph for 100 ms. Participants then saw
another fixation cross prior to rating how pleasant versus un-
pleasant they found the pictograph to be. Immediately after the
AMP procedure, participants rated their engagement and atten-
tion during the AMP. Participants then reported how hungry
they felt during the task. The study ended with demographics
and debriefing.
Materials.
Affective images. Participants saw a total of 24 neutral and 24
negative images selected from the International Affective Picture
System to serve as the affective context (IAPS; Lang, Bradley, &
Cuthbert, 2008). In the IAPS valence scale, negative images range
from 1–4, neutral images from 4–6, and positive images from
6–9. Based on IAPS norms, we selected images that fell in the
middle of each range for the negative and neutral images. These
excluded highly graphic negative images, which tend to be ex-
treme in valence and likely to elicit ceiling effects. The negative
images we chose ranged from 2.5–3.5, M
valence
⫽3.17 (SD ⫽.50);
M
arousal
⫽5.37 (SD ⫽.50), and neutral images ranged from
4.5–5.5, M
valence
⫽4.78 (SD ⫽.17); M
arousal
⫽2.99 (SD ⫽.70).
Chinese pictographs. Pictographs were randomly chosen
from the standard, previously validated AMP pictograph set
(Payne et al., 2010) and randomly assigned for pairing with the
affective contexts. Participants rated each pictograph on a bipolar
Likert scale from 1 ⫽extremely pleasant to4⫽neither pleasant
nor unpleasant to 7 ⫽extremely unpleasant (M⫽4.05, SD ⫽
1.37). Participants’ trial-by-trial ratings of the ambiguous picto-
graph targets were our dependent measures.
Self-reported engagement. All participants responded to six
questions assessing how engaged they were in the task, how easy
it was to focus and stay on task, and how successful they thought
they were at ignoring the affective primes using a Likert scale from
1to5(1⫽strongly disagree,5⫽strongly agree). A mean score
was created from these six items (M⫽4.42, SD ⫽.53, range
2.83–5.00, ␣⫽.75).
Self-reported hunger. Participants reported their degree of
hunger by responding to “How hungry were you during the rating
task?” on a Likert scale from 1 to 6 (1 ⫽not at all hungry,4⫽
somewhat hungry,6⫽extremely hungry; range 1–5; M⫽2.15;
SD ⫽1.32).
Analyses. Inspection of all variable histograms and scatter-
plots revealed no statistical outliers (greater than two standard
deviations from the mean). As per Bulmer (1979), where skewness
between ⫺1to⫺.50 and .50 to 1 indicates moderate skew, ratings
for self-reported hunger exhibited moderate skew (.71, SE ⫽.17)
such that a greater proportion of participants in our sample re-
ported that they were not hungry to moderately hungry as com-
pared with highly hungry. However, reports did not demonstrate
extreme skew and so we did not transform the distribution.
Due to the nested nature of the data, multilevel modeling with a
random intercept (Raudenbush & Bryk, 2001) was used to analyze
the data in SPSS. Context type (negative vs. neutral) was dummy-
coded and served as the predictor of participants’ pictograph
ratings at Level 1. Participants’ self-reported hunger ratings were
the predictor at Level 2, and we examined the cross-level interac-
tion between context and hunger. For each analysis in the study,
we ran two models as part of standard model-building practices
where Model 1 is a random effects ANOVA with no predictors to
demonstrate the degree of dependence in the data. Model 1 for
Study 1 is presented in Table 1, but is not discussed further. After
examining the degree of dependency between and within partici-
pants’ pictograph ratings, we used a random intercepts model to
examine the cross-level interaction between context at Level 1 and
hunger at Level 2. Standardized betas (␤) are presented throughout
(calculated as per Cohen, Cohen, West, & Aiken, 2003) as these
allow for effect size comparison, but see Table 1 for unstandard-
ized betas.
Results
As predicted, there was a significant effect of context,␤⫽.26,
p⬍.0001, 95% CIs [.23, .49], such that there was an estimated .26
unit increase in participants’ pictograph ratings on negative con-
text trials compared to neutral context trials (see Table 1). How-
ever, hunger alone did not increase participants’ negative ratings of
the pictographs. There was no significant main effect of hunger,
␤⫽⫺.02, p⫽.403, 95% CIs [⫺.10, .04]. Instead, as predicted,
a Context ⫻Hunger interaction, ␤⫽.06, p⫽.014, 95% CIs [.01,
.11] revealed that hungrier individuals only experienced ambigu-
ous pictographs as more negative in the context of a preceding
negative image (see Figure 2). A follow-up probe of the interaction
(see Table 2) indicated that hunger’s effect in negative trials was
significantly different than neutral trials.
Self-reported mean engagement was negatively correlated with
hunger,r⫽⫺.211, p⫽.002 such that the hungrier individuals
were, the more likely they were to report feeling like they strug-
gled to concentrate on the task. This finding may reflect hunger’s
impact on perceptions of subjective negative affect (i.e., feelings of
Figure 1. The affect misattribution procedure. An example trial from the affect misattribution procedure,
adapted from Payne et al. (2005, 2010). After each trial, participants rated the Chinese pictograph on a bipolar
scale (1 ⫽extremely pleasant, 4⫽neither pleasant nor unpleasant, 7⫽extremely pleasant). Pictograph
presented here is from the AMP task as validated in Payne et al. (2005).
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.
4MACCORMACK AND LINDQUIST

struggle). Nonetheless, when adding in mean engagement as a
covariate (Model 3 in Table 1), the results from Model 2 still hold
and there is no significant effect of mean engagement on picto-
graph ratings.
Discussion
In Study 1, we found that the hungrier an individual reported
feeling, the more likely they were to rate ambiguous pictographs as
negative in the presence of a negative, but not a neutral context.
These findings suggest that people may use the affect engendered
by hunger (vs. satiation) as evidence that stimuli are negative when
the context elicits such conceptualizations. According to some
psychological models, judging a stimulus as negative (i.e., a “per-
ception” or an “attitude”) is a different psychological state from
feeling unpleasant. However, a constructionist approach assumes
that affect and conceptualization are ingredients in many different
mental states (Cunningham et al., 2013; Lindquist, 2013; Lindquist
& Barrett, 2012); whether affect is attributed to an external stim-
ulus (e.g., an attitude about a pictograph) or one’s own body (e.g.,
Table 1
Study 1 Models for Hunger ⫻Context Effects on Pictograph Ratings
Effects b␤SE t 95% CI Lower 95% CI Upper
Model 1
Fixed effects
Intercept 4.051 — .044 91.945
ⴱⴱⴱ
3.964 4.138
Random effects
Residual variance (␴
2
) 2.663 — .040 — 2.584 2.744
Random intercept variance (␶
00
) .356 — .040 — .285 .445
Model 2
Fixed effects
Intercept 3.866 — .090 42.639
ⴱⴱⴱ
3.687 4.044
Context .362 .264 .066 5.462
ⴱⴱⴱ
.232 .492
Hunger ⫺.030 ⫺.028 .036 ⫺.838 ⫺.101 .040
Context ⫻Hunger .064 .061 .026 2.463
ⴱ
.013 .116
Random effects
Residual variance (␴
2
) 2.597 — .039 — 2.520 2.676
Random intercept variance (␶
00
) .358 — .040 — .286 .447
Model 3
Fixed effects
Intercept 4.300 — .400 10.737
ⴱⴱⴱ
3.511 5.090
Context .362 .264 .066 5.459
ⴱⴱⴱ
.232 .492
Hunger ⫺.038 ⫺.036 .036 ⫺1.042 ⫺.110 .034
Engagement ⫺.094 ⫺.036 .084 ⫺1.113 ⫺.261 .072
Context ⫻Hunger .064 .061 .026 2.465
ⴱ
.013 .116
Random effects
Residual variance (␴
2
) 2.597 — .039 — 2.520 2.676
Random intercept variance (␶
00
) .357 — .040 — .286 .447
ⴱ
p⬍.05.
ⴱⴱⴱ
p⬍.001.
Figure 2. Study 1 Hunger ⫻Context interaction. Participants who self-
reported as being hungrier were more likely to rate an ambiguous picto-
graph as unpleasant in negative versus neutral contexts. Although the
Likert scale anchors ranged from 1 ⫽not at all hungry to 6 ⫽extremely
hungry, actual responses ranged from 1–5, as represented in this graph.
Error bars computed ⫹/⫺1SE.
Table 2
Studies 1 and 2 Simple Slopes Tests for Model 2 With Neutral
Context as the Reference Category
Effects Estimate (SE)tp
Study 1
Intercept 4.23 (.091) 46.54 ⬍.0001
Slope .03 (.036) .96 .337
Study 2
Negative context
Intercept 4.23 (.099) 42.63 ⬍.0001
Slope .11 (.036) 3.09 .002
Positive context
Intercept 3.27 (.099) 32.93 ⬍.0001
Slope .01 (.036) .23 .821
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.
5
CONCEPTUALIZING HUNGER AS EMOTION

a feeling of unpleasantness) depends on the focus of attention in
that context (Lee, Lindquist, & Payne, 2017; Lindquist & Barrett,
2008). As we predicted, hunger does not automatically create more
negative affect in any context because hungry individuals did not
rate Chinese pictographs preceded by neutral images as more
negative.
As Study 1 is the first to demonstrate that hungry individuals
conceptualize their affect as negative feelings (i.e., negative judg-
ments of ambiguous stimuli) in the presence of a negative context,
we sought to replicate and extend this effect. Study 1 cannot rule
out the role of arousal in driving the findings, as the negative
images significantly differed from the neutral images in both
valence and arousal dimensions. Thus, it remains unclear whether
hunger would also interact with highly arousing positive contexts,
such that individuals would rate ambiguous pictographs as more
pleasant when feeling hungry following a pleasant context. How-
ever, given the literature suggesting that hunger results in unpleas-
ant, high arousal affect (e.g., Cryer, 1999), we did not predict
positive contexts to interact with hunger to shift affective percep-
tions; a positive context would be incongruent with participants’
hunger-induced affective state and they would thus be less likely to
make meaning of their state as being emotional in these contexts.
We thus conducted Study 2 to clarify the specificity of our hunger
effect.
Study 2
Although Study 1 provides initial evidence that hunger can be
conceptualized as negative judgments of ambiguous stimuli when
made meaningful in negative contexts, it could not rule out that
hunger is conceptualized as “hanger” in any high arousal context.
Study 2 thus built upon and extended Study 1 to rule out the
possibility that any high arousal context (negative or positive)
would allow participants to make meaning of their hunger as
affective feelings. Like Study 1, Study 2 employed the AMP but
this time included negative, positive, and neutral images as con-
text. Based on evidence that hunger is associated with self-reported
unpleasant affect (e.g., Cryer, 1999), we predicted that hunger
would impact affective judgments of ambiguous Chinese picto-
graphs in the presence of negative, but not positive or neutral
contexts.
Method
Participants. One-hundred and 92 Mechanical Turk workers
from the United States participated for monetary compensation. As
in Study 1, five participants were excluded from analyses because
they reported familiarity or fluency with Mandarin Chinese. Ad-
ditionally, 18 participants were excluded from final analyses be-
cause they were not blind to the purpose of the AMP (e.g., one
participant reported that it “tests my automatic affective biases”)
and 29 participants were excluded due to failed attention checks.
Thus, the final sample was 140 participants (46.4% female;
M
age
⫽35 years, SD
age
⫽10.54 years, 20- to 62-years-old). As in
Study 1, the sample size was determined ahead of time based on an
a priori power analysis and data were not analyzed until data
collection was complete.
Using Study 1 as a guide, we estimated that there would be
moderate-to-small effect sizes for our main effects (e.g., context
was ␤⫽.26; hunger was ␤⫽.04) and small effect size for
interactions (e.g., Context ⫻Hunger ␤⫽.06). Additionally, prior
simulation studies (Scherbaum & Ferreter, 2009) suggest that with
a Level 1 sample (trials) greater than 30 nested within a Level 2
sample (participants) greater than 40, we would have 90% power
to observe an effect (p⬍.05–.01). However, as we were not
manipulating hunger, and anticipating data loss due to our MTurk
sample, we aimed well above this sample size including Level 1
n⫽60 trials, Level 2 n⫽192 participants. We arrived at this
number of participants because we had allotted a set amount of
funds to the study, which resulted in 192 participants. Given that
this sample size would give us ample power, we collected data
until it was met.
Procedure. Study 2 exactly replicated Study 1 in using the
Qualtrics QRTEngine except for the addition of positive trials. All
participants first completed a practice block of 12 trials before
completing 60 experimental trials. Trials were presented in coun-
terbalanced blocks based on image type. After completing the
AMP, just as in Study 1, participants rated their engagement and
attention during the AMP. Participants then reported how hungry
they felt during the task. The study ended with demographics and
debriefing.
Materials.
Affective images. Participants saw a total of 24 neutral, 24
negative, and 24 positive images selected from the International
Affective Picture System to serve as the affective context (IAPS;
Lang et al., 2008). We used the same negative and neutral images
from Study 1, but chose additional positive images that ranged
from 6.5–7.5, M
valence
⫽6.93 (SD ⫽.33); M
arousal
⫽5.28 (SD ⫽
.48). It was impossible to match neutral images on arousal with
negative images, t(38) ⫽12.24, p⬍.0001, 95% CIs [1.98, 2.77]
or positive images, t(38) ⫽11.98, p⬍.0001, 95% CIs [1.90,
2.68], but negative and positive images were chosen so they did
not significantly differ in terms of arousal, t(38) ⫽.52, p⬎.250,
95% CIs [⫺.23, .39].
Chinese pictographs. Pictographs were randomly chosen
from the standard, previously validated AMP pictograph set
(Payne et al., 2010) and randomly assigned for pairing with the
affective contexts. As in Study 1, participants rated each picto-
graph on a bipolar Likert scale from 1 ⫽extremely pleasant to 4 ⫽
neither pleasant nor unpleasant to7⫽extremely unpleasant (M⫽
3.80, SD ⫽1.83). Participants’ trial-by-trial ratings of the ambig-
uous pictograph targets were our dependent measures.
Self-reported engagement. All participants responded to six
questions assessing how engaged they were in the task, how easy
it was to focus and stay on task, and how successful they thought
they were at ignoring the affective primes using a Likert scale from
1to5(1⫽strongly disagree,5⫽strongly agree). A mean score
was created from these six items (M⫽4.49, SD ⫽.50, range
2.80–5.00, ␣⫽.70).
Self-reported hunger. Participants reported their degree of hun-
ger by responding to “How hungry were you during the rating task?”
on a Likert scale from 1 to 6 (1 ⫽not at all hungry,4⫽somewhat
hungry,6⫽extremely hungry; range 1–5; M⫽2.34; SD ⫽1.40).
Analyses. Again, as in Study 1, inspection of all variable
histograms and scatterplots revealed no statistical outliers (greater
than two standard deviations from the mean). As per Bulmer
(1979), where skew between ⫺.50 and .50 is slightly or approxi-
mately skewed, ratings for self-reported hunger exhibited slight
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.
6MACCORMACK AND LINDQUIST

skew (.45, SE ⫽.21). Again, a greater proportion of participants in
our sample reported that they were not hungry to moderately
hungry as compared with highly hungry. We used multilevel
modeling with a random intercept, with dummy-coded context
(negative vs. neutral; positive vs. neutral) variables as predictors at
Level 1 and participants’ self-reported hunger as a predictor at
Level 2. We also examined the predicted cross-level interaction
between context and hunger. As part of model-building, Model 1
was a random effects ANOVA with no predictors to examine
dependence in the data. Model 1 for Study 2 is presented in Table
3 but not discussed further. In Model 2, we used a random
intercepts model to examine the cross-level interaction between
context and hunger. Standardized betas are presented throughout
as these allow for effect size comparison, but see Table 3 for
unstandardized betas.
Results
Replicating Study 1, the context influenced participants’ ratings
of ambiguous pictographs. Similar to Study 1, there was an esti-
mated .36 unit increase in participants’ pictograph ratings in neg-
ative contexts as compared to neutral contexts ␤⫽.36, p⬍.0001,
95% CIs [.51, .83] (see Table 3). Consistent with the broader AMP
literature, there was also an estimated .16 unit decrease in partic-
ipants’ pictograph ratings following positive contexts, indicating
that pictographs were rated more pleasantly on positive compared
to neutral contexts, ␤⫽⫺.16, p⬍.0001, 95% CIs [⫺.45, ⫺.13].
Critically, as in Study 1 there was no significant main effect for
hunger,␤⫽.03, p⫽.165, 95% CIs [⫺.01, .115], suggesting once
again that hunger on its own does not appear to drive affective
perceptions of the pictographs. Critical to our hypothesis and
replicating Study 1, there was only a significant Negative Con-
text ⫻Hunger interaction, ␤⫽.04, p⫽.031, 95% CIs [.01, .12]
(see Figure 3). A follow-up probe of the interaction found that
negative pictograph ratings were greater for hungry individuals in
the context of negative images, but not positive or neutral
images (see Table 2). Additionally, there was no significant
interaction for Positive Context ⫻Hunger, ␤⫽⫺.02, p⫽.186,
95% CIs [⫺.09, .01].
As in Study 1, individuals’ self-reported Mean Engagement was
again negatively correlated with hunger,r⫽⫺.211, p⫽.002.
However, when adding in mean engagement as a covariate (Model
3 in Table 3), the results from Model 2 still held with no significant
effect of mean engagement on pictograph ratings.
Discussion
Study 2 not only replicated the findings from Study 1, but also
ruled out that the effects of Study 1 were general to any highly
arousing context. Instead, our findings suggest that a negative
context may be key for transforming hunger into feeling negative,
high arousal emotions or colloquially, “hanger.” The unpleasant,
highly aroused affective feelings engendered by hunger appear to
be attributed to ambiguous Chinese pictographs only when the
context is congruent with the hedonic tone of those feelings. These
Table 3
Study 2 Models for Hunger ⫻Context Effects on Pictograph Ratings
Effects b␤SE t 95% CI Lower 95% CI Upper
Model 1
Fixed effects
Intercept 3.805 — .044 86.841
ⴱⴱⴱ
3.718 3.891
Random effects
Residual variance (␴
2
) 3.139 — .046 — 3.050 3.229
Random intercept variance (␶
00
) .225 — .032 — .169 .298
Model 2
Fixed effects
Intercept 3.565 — .094 38.094
ⴱⴱⴱ
3.380 3.749
Negative context .670 .366 .082 8.210
ⴱⴱⴱ
.510 .830
Positive context ⫺.293 ⫺.160 .082 ⫺3.595
ⴱⴱⴱ
⫺.453 ⫺.133
Hunger .048 .036 .034 1.395 ⫺.019 .115
Negative Context ⫻Hunger .064 .048 .030 2.158
ⴱ
.006 .122
Positive Context ⫻Hunger ⫺.039 ⫺.029 .030 ⫺1.323 ⫺.097 .019
Random effects
Residual variance (␴
2
) 2.907 — .042 — 2.826 2.992
Random intercept variance (␶
00
) .224 — .032 — .169 .296
Model 3
Fixed effects
Intercept 3.582 — .477 7.504
ⴱⴱⴱ
2.638 4.526
Negative context .670 .366 .082 8.210
ⴱⴱⴱ
.510 .830
Positive context ⫺.293 ⫺.160 .082 ⫺3.595 ⫺.453 ⫺.133
Hunger .047 .035 .035 1.360 ⫺.021 .116
Engagement ⫺.004 ⫺.001 .099 ⫺.036 ⫺.199 .192
Negative Context ⫻Hunger .064 .048 .030 2.158
ⴱ
.006 .122
Positive Context ⫻Hunger ⫺.039 ⫺.029 .030 ⫺1.323 ⫺.097 .019
Random effects
Residual variance (␴
2
) 2.907 — .042 — 2.826 2.992
Random intercept variance (␶
00
) .226 — .032 — .170 .299
ⴱ
p⬍.05.
ⴱⴱⴱ
p⬍.001.
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.
7
CONCEPTUALIZING HUNGER AS EMOTION

findings thus provide important boundary conditions for the expe-
rience of “hanger.”
Despite providing preliminary evidence that hunger can become
conceptualized as emotional in nature, Studies 1 and 2 had several
limitations. First, both studies measured rather than manipulated
hunger. Second, they used a cognitive paradigm that may lack
ecological validity. Third, we took for granted the fact that indi-
viduals would misattribute their hunger-induced affect based on
the design of the AMP, but we did not explicitly manipulate the
likelihood of such misattributions. Fourth, Studies 1 and 2 exam-
ined how conceptualizations of hunger as emotion might result in
negative judgments of stimuli, but did not examine how concep-
tualizations of hunger as emotion resulted in emotional experi-
ences (e.g., the self-reported experience of specific unpleasant
emotions). Finally, an alternate hypothesis is that hungry partici-
pants in Studies 1–2 merely lacked the self-regulation to inhibit
more negative ratings of the Chinese pictographs. This interpreta-
tion is less likely given that we did not observe a main effect of
hunger on pictograph ratings, but given the literature linking
glucose to self-control failures (e.g., Gailliot et al., 2007) and
hypothesizing the effect of hunger on self-control in aggression
(Bushman et al., 2014), we sought to rule out this alternate hy-
pothesis in Study 3.
Study 3
Study 3 was a laboratory experiment designed to address the
additional questions raised by the findings of Studies 1–2. Study 3
builds upon Studies 1–2 in four ways. First, we experimentally
manipulated hunger versus satiation. Second, our paradigm in-
volved social interactions and a real negative context to increase
the ecological validity of our findings. Third, we manipulated the
likelihood that participants would misattribute their hunger-
induced affect to the situation. Following research on affect mis-
attribution (Schwarz & Clore, 1983), we hypothesized that making
participants relatively more aware of their emotional conceptual-
izations would reduce their tendency to conceptualize hunger as
“hanger.” Drawing on the affect-labeling literature (e.g., Kassam
& Mendes, 2013; Lieberman et al., 2007; Niles, Craske, Lieber-
man, & Hur, 2015), we did so by asking participants to write about
emotion concepts (“anger,” “sadness”) in some conditions, or
asking them to write about neutral, nonemotional information in
another condition. Fourth, Study 3 measured the impact of hunger
on self-reported emotional experiences and interpersonal judg-
ments. Finally, to rule out other mechanisms for our findings,
Study 3 addressed the alternate hypothesis that regulatory deple-
tion primarily drives “hanger” by assessing whether hungry people
exhibited less self-regulation than satiated people. Self-regulation
was assessed as the length of time participants persevered on a
tedious mental rotation task (Shepard & Metzler, 1971; Williams
& DeSteno, 2008).
Method
Participants. Two-hundred and 36 PSYC101 students (58%
female; M
Age
⫽19 years old, SD
Age
⫽2.48, 17- to 45-years-old)
were recruited from the University of North Carolina at Chapel
Hill study pool and participated in the laboratory experiment for
research credit. Based on the small to medium effect size for main
effects and interactions observed in Lindquist and Barrett (2008)
which similarly manipulated affect and attention to emotion,
power analyses in GPower (Faul, Erdfelder, Lang, & Buchner,
2007) suggested that we would have 80% power to observe a
significant interaction effect (p⬍.05–.01) with 180 participants
and 90% power with 230 participants. We aimed to collect 240
participants (n⫽40 per condition) but stopped short by four
participants due to the end of the semester. No data were analyzed
until after data collection finished.
Procedure. Upon signing up for the experiment, participants
were prescreened so that any individuals who were unable or
unwilling to change their normal eating schedule prior to lab
arrival were excluded. We also explicitly excluded participants
who might be adversely impacted by our manipulations (i.e., with
diabetes, eating disorders, or mood disorders). After prescreening,
participants were randomly assigned to a condition in a 2 (Body
State: Hungry vs. Satiated) ⫻3 (Attentional Focus: Anger-
Focused vs. Sadness-Focused vs. No Emotion-Focused) between-
subjects design. Participants in the hunger condition fasted for 5 or
more hours prior to the lab visit and participants in the satiated
condition ate a full meal or large snack less than one hour prior to
lab visit. As a cover story, participants were told that the study was
about “visual performance” and that they needed to fast versus eat
prior to arrival so that we could control for the impact of glucose
on visual performance. Refer to Figure 4 for study timeline.
Fasting manipulation check. Upon arrival, participants com-
pleted informed consent and a “Food Questionnaire” which en-
sured that they had actually fasted or eaten as instructed. The
questionnaire consisted of three items indexing when the partici-
pant last ate or drank something other than water (Likert scale
ranging from 1 ⫽less than 1 hr ago,7⫽more than 6 hr ago), how
hungry the participant felt (Likert scale ranging from 1 ⫽not at all
hungry,7⫽extremely hungry), and what type of meal they had
last eaten (a full meal, moderate snack, small snack, or caloric
beverage such as a protein shake, coffee, juice, soda, etc.). After
Figure 3. Study 2 Hunger ⫻Context interactions. Participants who
self-reported as being hungrier were more likely to rate an ambiguous
pictograph as unpleasant in negative versus positive and neutral contexts.
Although the Likert scale anchors ranged from 1 ⫽not at all hungry to 6 ⫽
extremely hungry, actual responses ranged from 1–5, as represented in this
graph. Error bars computed ⫹/⫺1SE.
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.
8MACCORMACK AND LINDQUIST

completing the questionnaire, the experimenter checked that the
assigned hungry versus satiated condition had been met. Partici-
pants in the hunger condition who had eaten less than 5 hr prior to
arrival or participants in the satiation condition who had eaten
more than 1 hr prior to arrival and had not eaten a full meal or
moderate snack were rescheduled and reinstructed according to
their assigned condition.
Measure of self-regulation. All participants next completed a
“spatial reasoning” task, which was our measure of self-regulation.
This was a mental rotation task that consisted of geometric shapes
taken from Shepard and Metzler (1971, see example in Figure 4).
As per Williams and DeSteno (2008), this task was used to assess
perseverance and thus served as a measure of self-regulation.
Specifically, we examined whether hungry individuals were less
able to persevere at the tedious task than satiated individuals. On
each trial, participants compared the images of two geometrical
figures on a computer screen and determined if these figures were
able to be rotated in space to match one another or not. Participants
were told that the figure combinations were infinite and that the
task would take longer than the experiment allowed. Participants
were shown how to press a key to exit out of the task when they
felt they had completed as many trials as they could. The number
of minutes that participants spent on this task served as our
measure of self-regulation.
Attentional focus manipulation. Next, all participants com-
pleted a writing task meant to direct their explicit focus on specific
emotion concepts or neutral information (adapted from Lindquist
& Barrett, 2008). For this task, participants were randomly as-
signed to view a male face posed to display a prototypically angry,
sad, or neutral facial expression (Figure 4 shows the “anger” face).
It was reasoned that the facial expression would serve as a visual,
symbolic representation for that emotion concept (Barrett, 2011;
Doyle & Lindquist, 2017; Fridlund, 1994; Lindquist & Gendron,
2013) and thus a good cue for accessing concept-relevant associ-
ations. Participants were told that the individual was named Jon
and that “Jon feels angry (sad/neutral).” They were then instructed
to write a vignette detailing: (a) “How does Jon feel? Describe his
thoughts and bodily sensations in as much detail as possible;” (b)
“What actions might Jon take?;” and (c) “Why does Jon feel angry
(sad/neutral)—what happened to make him feel this way?” Partic-
ipants were told they could write as little or as much as they
wished, as long as they were detailed under the cover story that this
was a measure of “cognitive complexity.” As a manipulation
check, participants’ vignettes were later transcribed and coded
using the Linguistic Inquiry and Word Count program (Penne-
baker, Booth, & Francis, 2015).
Context manipulation. Next, all participants underwent the
same negative interpersonal situation to create an ecologically
valid context in which they could conceptualize their hunger as
“hanger.” We used a displaced aggression paradigm in which
participants completed a tedious computerized “visual dexterity”
task that crashed part way through. In the task, participants saw a
series of concentric colored circles on the computer screen (see
Figure 4 for an example) and were asked to decide as quickly and
accurately as possible whether the number of circles present was
odd or even. Participants were told the cover story that this task
measured the speed and accuracy of their visual perception. After
100 trials, participants received an error message that simulated a
Windows computer crash. This prompted the participant to find the
experimenter outside the testing room and inform him/her of the
crash. All experimenters were trained to deliver the same negative
reaction. The experimenter entered the testing room, looking con-
fused and upset. The experimenter attempted to fix the crash by
typing on the keyboard and clicking the mouse. The experimenter
then said “This has never happened before,” then asked the par-
ticipant: “What did you do? What keys did you press?” Finally, the
experimenter said “I don’t know how to fix this, but I’m going to
go contact my supervisor to find out. Once I get the task fixed,
you’re going to have to do the whole task over again if you want
your study credit.” After this, the experimenter left the room for a
brief period (2 min), allowing the participant to consider the
situation alone.
Dependent measures. After the brief period was over, the
experimenter reentered the testing room with a manila envelope
filled with questionnaires, and asked the participant to fill out the
questionnaires while s/he contacted the study supervisor in another
room. Inside the envelope were the two dependent measures: one
questionnaire was called a “Participant Satisfaction” questionnaire
and inquired about the participants’ emotions. The other was called
the “Rate Your Experience” questionnaire and inquired about the
participants’ perceptions of the quality of the experimenter and the
study. The cover story was that the lab had been randomly selected
by the Psychology and Neuroscience Department for a routine
quality control study on research within the department. Partici-
pants were informed that the questionnaires would be used to
evaluate whether the experimenters were performing their jobs
well. Participants were ensured that questionnaires were com-
pletely anonymous and that s/he should seal the envelope after
finishing the questionnaires to ensure confidentiality.
For the “Participant Satisfaction” questionnaire, we used the
modified Differential Emotion Scale (mDES; Fredrickson, Tu-
gade, Waugh, & Larkin, 2003) for participants’ self-reported emo-
Figure 4. Study 3 procedural order across participants. Names of tasks as
told to participants are in quotes, as part of our cover story that this was a
visual perception study. Geometric shape presented here is from Shepard
and Metzler’s (1971) mental rotation task. Facial image presented is from
the IASLab Set (Gendron, Lindquist, & Barrett, 2011).
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.
9
CONCEPTUALIZING HUNGER AS EMOTION

tion experiences. Participants rated how many times they had
experienced that emotion during the lab visit today using a 5-point
Likert scale (0 ⫽not at all to 4 ⫽extremely). The mDES covers
20 questions, with each question containing three synonymous
emotion terms. In total, there were 10 positive emotion questions
(e.g., “What is the most grateful, appreciative, or thankful you
felt?”) and 10 negative emotion questions (e.g., “What is the most
angry, irritated, or annoyed you felt?”). For the “Rate Your Expe-
rience” questionnaire, participants rated their personal impressions
of how aggressive, helpful, lacking in empathy, professional, and
judgmental the experimenter was. All items were rated for agree-
ment using a 7-point Likert scale (0 ⫽not at all to 6 ⫽extremely).
Demographics and debriefing. After the participant com-
pleted the dependent measures, the experimenter reentered the
testing room and told the participant that the study supervisor had
decided the participant did not need to redo the “Visual Dexterity”
task after all. Participants then completed a standard demographics
questionnaire before completing a funneled debriefing. During
debriefing, participants were first interviewed about their general
thoughts and reactions to the experiment and also their impressions
about what the experiment assessed. No participants reported that
they thought the study was about hunger’s influence on their
emotions and interpersonal judgments, nor did any participants
report suspicions about the authenticity of the computer crash.
After participants provided their hypotheses about the study’s
purpose, they were debriefed as to the true nature of the experi-
ment and offered snacks.
Analyses. To compute our dependent variables, we used the
items on the mDES (i.e., the “Participant Satisfaction” question-
naire), to first create a mean score for participants’ self-reported
negative, high arousal emotions (mean score included ratings for
the anger, contempt, disgust, embarrassment, fear, guilt, hate,
shame, and stress items; ␣⫽.83). We considered these emotions
to be negative and high arousal based on multidimensional scaling
and factor loadings from prior literature (e.g., Alvarado & Jame-
son, 2011; Feldman, 1995; Russell, 1980; Russell & Bullock,
1986). We did not include sadness as this is prototypically con-
sidered a low arousal emotion (Shaver, Schwartz, Kirson, &
O’Connor, 1987), although we recognize that the valence and
arousal associated with each emotion category can vary by in-
stance (e.g., Wilson-Mendenhall, Barrett, & Barsalou, 2013). We
chose to look specifically at negative, high arousal items because,
as mentioned, the literature consistently shows that hunger induces
unpleasant, highly aroused affect (e.g., Cryer, 1999; Heller et al.,
1987). Notably, our findings hold regardless of whether we assess
negative, high arousal emotions specifically or focus on all nega-
tive emotions (see online supplemental materials), likely because
there are relatively few negative low arousal emotions in the
mDES. In future research, it would be interesting to systematically
assess the extent to which hunger impacts high versus low arousal
negative emotions. Notably, as follow-up exploratory analyses, we
chose to specifically examine the emotion categories of anger and
hate, as these would most approximate the colloquial experience of
feeling “hangry.”
As we were interested in participants’ interpersonal judgments
of the experimenter, we also used the items on the “Rate Your
Experience” questionnaire to create a mean score for participants’
ratings of the experimenter consisting of ratings on how helpful,
professional, empathetic, difficult, aggressive, and judgmental the
experimenter was. Positive items were reverse scored resulting in
an index of unpleasant interpersonal judgments (␣⫽.72). Nota-
bly, as follow-up exploratory analyses, we examined a subset of
the items that we deemed a priori to be most likely to relate to the
negative interpersonal situation participants had experienced
(how aggressive, lacking empathy, and judgmental the experi-
menter was).
Prior to analyzing our data, we performed a manipulation check
(Blood Glucose Questionnaire) to ensure that fasting versus sati-
ated participants were indeed hungry versus not hungry. In our
sample, 119 participants were assigned to the hunger condition
versus 117 participants were assigned to the satiation condition.
However, 54 participants in the hunger condition who fasted for 5
or more hours still reported being somewhat full (3) to not at all
hungry (1) on the “How hungry are you right now?” scale (1 ⫽not
at all hungry to7⫽extremely hungry). similarly, seven partici-
pants in the satiated condition who just ate a full meal 1 hr or less
still reported being somewhat hungry (4) to extremely hungry (7).
These findings likely represent normal variation in satiety between
meals, as individuals can vary in length of satiety depending on
metabolism, recent meal content and physical activity, hormonal
cycles such as menstruation, and even morphological factors such
as the length of the small intestine (Blundell, Stubbs, Hughes,
Whybrow, & King, 2003; Jeroen Maljaars, Peters, Kodde, &
Geraedts, 2011; Lawton, Delargy, Brockman, Smith, & Blundell,
2000; Pohle-Krauza, Carey, & Pelkman, 2008). Indeed, there is
evidence that even at longer intervals of short-term fasting (⬎24
hr) there is variance in subjective hunger intensity (e.g., as in
Herbert et al., 2012).
To balance concerns about reliability and validity, we performed
analyses in three ways (see online supplemental materials for
additional analyses). First, we included all participants, regardless
of whether they passed our manipulation check or not in an
“intent-to-treat” analysis (see Bouwmeester et al., 2017). This type
of analysis assesses the impact of the experimentally manipulated
independent variable on the dependent variable, irrespective of
individual differences in manipulation success. Second, we ex-
cluded any participant who explicitly failed our manipulation
check (by reporting less than the Likert midpoint of 4 in the hunger
condition and greater than 4 in the satiated condition), resulting in
65 hungry participants and 110 satiated participants, with a final
N⫽175 (ⱖ20 participants per condition). Third, we performed
follow-up analyses in which we preserved power by using the full
sample but reassigned participants to the hungry versus satiated
condition based on their self-reported hunger. Importantly, the
pattern of findings for self-reported emotions replicated regardless
of the analysis method used, bolstering the claim that manipulating
participants’ blood glucose levels via fasting had the predicted
effect on emotion experience (see online supplementary materials).
We report the analyses in which we excluded participants who
were not hungry because this analysis most closely addressed the
effect of the psychological state of hunger on our outcomes. We
recognize, however, that this choice leaves open the interpretation
that the participants for whom the hunger manipulation was effec-
tive might have systematically differed from others in some man-
ner (Bouwmeester et al., 2017). Our pattern of findings in the
intent-to-treat analyses (see online supplemental materials) mini-
mizes this concern, but do not fully mitigate it as it is clear from
our manipulation check that fasting did not impact all participants
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.
10 MACCORMACK AND LINDQUIST

equally. Future research should examine which factors moderate
the effect of fasting on hunger, as addressing these moderators
would give even more insight into the mechanisms of “hanger.”
One candidate may be individual differences in interoception, that
is, the ability to perceive changes in the visceral body (Critchley &
Garfinkel, 2017).
As a manipulation check for the attentional focus factor, we
content-coded responses on the writing task using the LIWC
(Pennebaker et al., 2015) to ensure that participants in the anger-
focused condition were using more anger words than those in the
sadness-focused and no emotion-focused conditions and vice versa
in the sadness-focused condition. In the no emotion-focused con-
dition we ensured that participants were not explicitly focusing on
anything emotional by comparing the number of affective words
they used, in general (i.e., any positive or negative affect words,
e.g., good, bad), compared with the other two conditions. One-way
ANOVAs confirmed that our manipulation was successful, such
that anger-focused participants wrote about anger more than those
in the sadness-focused and no emotion-focused conditions (ps⬍
.001), sadness-focused participants wrote about sadness more than
those in the anger-focused and no emotion-focused conditions
(ps⬍.001), and no emotion-focused participants wrote about
emotion and affect less than those in the anger-focused and
Sadness-focused conditions (ps⬍.001). See Table 4 for manip-
ulation check results.
To test our primary hypotheses, we first established that no
outliers bore undue leverage on our findings for any variables (i.e.,
establishing that no individual’s means were greater than two
standard deviations above the sample mean for any variable). To
test the primary hypothesis that hunger is conceptualized as emo-
tion when participants are not explicitly paying attention to their
feelings, we then examined main effects and interactions using a
series of 2 Body State (Hunger, Satiated) ⫻3 Attentional Focus
(Anger-Focused, Sadness-Focused, No Emotion-Focused) between-
subjects factorial ANOVAs. The dependent variables were partici-
pants’ self-reported emotions and perceptions of the experimenter.
Second, to rule out that hunger induces self-regulatory depletion, we
used an independent-samples ttest to compare how long hungry
versus satiated participants persisted at the mental rotation task and
also used our measure of self-regulation as a covariate in a series of
2 Body State (Hunger, Satiated) ⫻3 Attentional Focus (Anger-
Focused, Sadness-Focused, No Emotion-Focused) between-subjects
factorial ANCOVAs.
Results
Self-reported emotions. A 2 (Hunger vs. Satiation) ⫻3
(Anger-Focused, Sadness-Focused, No Emotion-Focused) ANOVA
for mean negative, high arousal emotions revealed a significant main
effect of body state, F(1, 166) ⫽4.47, p⫽.036, ␩
2
⫽.02 such that
hungry individuals were more likely to report feeling high arousal,
negative emotions compared with satiated individuals. We found no
significant main effect of attentional focus, F(2, 166) ⫽2.12, p⫽
.112, ␩
2
⫽.02. However, as predicted, there was a significant inter-
action between Body State ⫻Attentional Focus, F(2, 166) ⫽5.51,
p⫽.005, ␩
2
⫽.06. A doubly centered planned contrast (Abelson &
Prentice, 1997) confirmed our a priori prediction that hungry partic-
ipants whose attention was not directed toward emotional information
(M⫽.88) were more likely to report negative, high arousal emotions
as compared to other hungry participants who explicitly focused their
attention on emotions such as anger (M⫽.42) or sadness (M⫽.49),
or as compared with satiated participants who focused on anger (M⫽
.48), sadness (M⫽.47), or no emotional information (M⫽.37), F(1,
166) ⫽11.52, p⬍.001 (see Figure 5).
As follow-up exploratory analyses, we focused on a two a priori
emotion adjectives: anger and hate, given their link to the colloquial
use of “hanger.” Using a 2 (Hunger vs. Satiation) ⫻3 (Anger-
Focused, Sadness-Focused, No Emotion-Focused) ANOVA, we
again found significant effects for self-reports of hate, but not for
anger (ps⬎.250).
For self-reports of hate, there was a significant main effect for
body state, F(1, 165) ⫽4.90, p⫽.028, ␩
2
⫽.03, such that
individuals who were hungry were more likely to report feeling
hate compared with individuals who were satiated. The main effect
of attentional focus was not significant, F(2, 165) ⫽1.44, p⫽
.241, ␩
2
⫽.02. Finally, we observed a significant interaction
between Body State ⫻Attentional Focus, F(2, 165) ⫽5.77, p⫽
.004, ␩
2
⫽.06. A doubly centered planned contrast confirmed the
previous pattern of findings: That hungry participants who did not
focus on emotional information (M⫽.65) were more likely to
report experiencing hate as compared with other hungry partici-
pants who focused on emotions such as anger (M⫽.09) or sadness
(M⫽.25), or as compared with satiated participants who focused
on anger (M⫽.23), sadness (M⫽.14), or no emotional informa-
tion (M⫽.03), F(1, 165) ⫽11.05, p⫽.001 (see Figure 6).
Self-reported perceptions of the experimenter. Next, to ex-
amine participants’ interpersonal judgments of the experimenter,
we ran a 2 (Hunger vs. Satiation) ⫻3 (Anger-Focused, Sadness-
Focused, No Emotion-Focused) ANOVA using the mean experi-
menter ratings score as the dependent variable (i.e., how helpful,
professional, empathetic, difficult, aggressive, and judgmental the
experimenter was). There were neither significant main effects
of the body state nor attentional focus, nor a significant inter-
action (ps⬎.20). As follow-up exploratory analyses, we ex-
amined a subset of the items that were a priori most likely to
Table 4
Mean Anger, Sadness, and Affective Words Written in the Attentional Focus Writing Task
Outcome
Attentional focus condition Test of significance
Anger-focused Sadness-focused No emotion-focused F(2, 169) p
Anger words 5.39 (.234)
s,n
.83 (.239) .30 (.270) 133.55 ⬍.0001
Sadness words .77 (.200) 5.67 (.205)
a,n
.54 (.231) 192.99 ⬍.0001
Affective words 9.78 (.404) 10.53 (.412) 6.73 (.465)
a,s
20.39 ⬍.0001
Note. Post-hoc simple effect significant differences (ps⬍.05) between the three conditions are denoted with a ⫽anger-focused, s ⫽sadness-focused,
and n ⫽no emotion-focused. Standard errors are provided in parentheses next to the word count means.
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.
11
CONCEPTUALIZING HUNGER AS EMOTION

relate to the negative interpersonal situation participants had
experienced (how aggressive, lacking empathy, and judgmental
the experimenter was) to ascertain whether hunger interacted
with awareness to influence individuals’ more specific inter-
personal judgments. Again, we ran 2 (Hunger vs. Satiation) ⫻
3 (Anger-Focused, Sadness-Focused, No Emotion-Focused)
ANOVAs with experimenters’ rated aggressiveness, lack of
empathy, and judgmental behavior as the dependent variables.
We did not find effects on participants’ ratings of aggressive-
ness or lack of empathy (all ps⬎.10).
There were, however, significant effects for participants’ ratings
of the experimenter as judgmental. There was no significant main
effect for body state, F(1, 168) ⫽1.25, p⫽.265, ␩
2
⬍.01, but
there was a significant main effect for attentional focus, F(2,
168) ⫽4.05, p⫽.019, ␩
2
⫽.05. Simple effects revealed that
within the attentional focus factor, participants in the no emotion-
focused conditions were significantly more likely to rate the ex-
perimenter as judgmental compared with participants in the anger-
focused (p⫽.006) and sadness-focused conditions (p⫽.012).
Critical to our hypotheses, there was a marginally significant
interaction between Body State ⫻Attentional Focus, F(2, 168) ⫽
2.29, p⫽.105, ␩
2
⫽.03. A doubly centered planned contrast
confirmed the previous pattern of findings: that hungry partici-
pants who did not focus on emotional information (M⫽.70) were
more likely to rate the experimenter as judgmental as compared
with other hungry participants who focused on emotions such as
anger (M⫽.05) or sadness (M⫽.13), or as compared with
satiated participants who focused on anger (M⫽.05), sadness
(M⫽.22), or no emotional information (M⫽.21), F(1, 168) ⫽
4.65, p⫽.032 (see Figure 7).
Notably, this finding should be taken as preliminary and spec-
ulative, given that it did not replicate in the “intent-to-treat”
analysis with the full sample, although it did replicate when we
reassigned participants based on their hunger status (see online
supplemental materials). This pattern of findings leaves open
the alternative hypothesis that participants who were assigned to
the hunger and satiation conditions and who passed the manipu-
lation check differ from participants who failed the check in some
meaningful way, at least with regards to the effects of hunger on
interpersonal judgments. We address these ideas in the Limitations
section below.
Regulatory depletion. Contrary to prior hypotheses (e.g.,
Bushman et al., 2014), hunger did not induce self-regulatory
depletion: hungry participants persevered as long as satiated par-
ticipants on the mental rotation task (M
hunger
⫽8.16 min vs.
M
satiated
⫽8.90 min), t(149) ⫽⫺.75, p⫽.457, 95% CIs [⫺2.72,
1.23], Cohen’s d⫽.13 (see Figure 8). Finally, to determine if
self-regulation interacted with body state or attentional focus to
drive self-reported emotions or interpersonal judgments, we reran
the ANOVA models above with time spent on the mental rotation
task as a covariate representing how much regulatory reserve
hungry versus satiated participants possessed. Self-regulation was
not significant in any model, neither as a main effect nor interac-
tion and the pattern of the previously reported ANOVAs remained
the same (see online supplementary materials).
Discussion
In Study 3, we found evidence that hunger can become concep-
tualized as emotion in a negative interpersonal situation, but only
when individuals are not explicitly focused on emotions. Consis-
Figure 5. Study 3 mean differences for mean negative, high arousal
emotions. Hungry individuals in the no emotion-focused condition re-
ported significantly greater feelings of negative, high arousal emotions
(anger, contempt, disgust, embarrassment, fear, guilt, hate, shame, and
stress). Error bars computed ⫹/⫺1SE.
Figure 6. Study 3 mean differences for hate. Hungry individuals in the no
emotion-focused condition reported significantly greater feelings of hate.
Error bars computed ⫹/⫺1SE.
Figure 7. Study 3 mean differences in ratings of the experimenter as
judgmental. Hungry individuals in the no emotion-focused condition re-
ported significantly greater perceptions of the experimenter as judgmental.
Error bars computed ⫹/⫺1SE.
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.
12 MACCORMACK AND LINDQUIST

tent with our predictions, hungry individuals reported greater un-
pleasant, high arousal emotions when their attention was not
specifically drawn to emotion. In exploratory analyses, we found
that individuals were also more likely to indicate that they were
experiencing “hate” and to view the research assistant as judgmen-
tal if they were hungry but not directed to attend to emotion. Thus,
although it is clear that hunger interacts with awareness to impact
multiple unpleasant, high arousal emotions, we do have some
preliminary evidence that if the context affords it, hunger and
awareness can translate into unpleasant, antisocial interpersonal
feelings and behaviors.
Critically, contrary to the regulatory depletion hypothesis, hun-
gry versus satiated individuals did not differ in self-regulation, nor
did self-regulation impact participants’ self-reported emotions and
interpersonal judgments when controlled for in analyses. This
finding suggests that regulatory depletion, as measured in the
present study, did not play a powerful role in shaping individuals’
emotions and judgments. Of course, it remains a possibility that
our measure of perseverance did not adequately measure partici-
pants’ self-regulatory abilities. However, the fact that hunger in-
teracted with participants’ awareness about emotions and the fact
that we did not consistently find a main effect of hunger further
underscores our interpretation that negative emotion can be con-
structed out of hunger when people make meaning of hunger-
induced negative affect in context.
Study 3 thus built on Studies 1–2 to demonstrate that specifi-
cally manipulating hunger and participants’ attention toward emo-
tional information altered “hanger.” Furthermore, these hunger-
induced emotions manifested as both experiences of negative
emotions (i.e., greater ratings of negative, high arousal emotions)
as well as perceptions that the researcher was judgmental.
General Discussion
Across three studies, we found evidence that hunger alters
individuals’ affective perceptions and experiences. Specifically,
we demonstrated that hunger can be experienced as a negative,
high arousal state—or “hanger”—when made meaningful as an
instance of emotion in a negative context. Consistent with the
classic misattribution literature, this effect occurred only when
participants were not explicitly focused on emotions. In Studies
1–2, hungrier participants were more likely to judge an ambiguous
stimulus as negative when it was preceded by a negative, but not
neutral or positive, context. Hunger on its own did not automati-
cally lead to negative affective judgments; instead, the negative
context guided how individuals automatically made meaning of
the hunger-induced affect.
Study 3 extended Studies 1–2 by explicitly manipulating hunger
and focus on emotions and examining how hunger translates into
emotions and judgments in a more ecologically valid interpersonal
context. We found that participants experienced hunger as negative
emotions, but only when not explicitly directed to focus attention
on emotional information. We have preliminary evidence that
individuals were also more likely to experience negative, antisocial
states when hungry but not explicitly focusing on emotions; par-
ticipants who were hungry but not focused on emotions were more
likely to report feeling “hate” and that the researcher was “judg-
mental.” Study 3 also provides preliminary evidence that regula-
tory depletion may not be the primary mechanism by which hunger
impacts emotion. Hungry versus satiated individuals did not differ
in self-regulation, nor did self-regulation impact the effects of
hunger and attentional focus on emotions when included as a
covariate in our models. Our findings are thus contrary to the
hypothesis that hunger primarily impacts our social and affective
lives by depleting self-control (e.g., Bushman et al., 2014; DeWall
et al., 2011).
Study 3 instead suggests that feeling “hangry” stems from
making meaning of hunger-induced affective sensations as feel-
ings of negative, high arousal emotion in an unpleasant context.
Although the colloquialism “hangry” implies that people who are
hungry may feel specifically angry, we did not predict that people
would exclusively experience anger for several reasons. First,
following the literature on the affective states induced by hunger as
well as previous psychological constructionist and affect misattri-
bution theories, we hypothesized that hunger itself would give rise
to a general feeling of unpleasant, highly aroused affect. Further-
more, following evidence that people can make meaning of their
affect in multiple ways based on the context and their goals
(Ellsworth & Scherer, 2003; Kirkland & Cunningham, 2012;
Lindquist & Barrett, 2008), as well as evidence that people rarely
endorse a single negative emotion category to describe their ex-
perience (Barrett et al., 2001; Feldman, 1995), we did not predict
that conceptualizations of hunger would result in a single emotion
(e.g., anger). This prediction was borne out in the data, where we
found in Study 3 that participants who were hungry and not
focused on their emotions reported more intense experiences of all
unpleasant, highly aroused emotions. That said, we did find that
participants reported feeling “hate” and that they found the re-
searcher to be judgmental, which may reflect participants’ nega-
tive, antisocial experiences toward the researcher in that interper-
sonal context, more specifically.
Limitations
As a first direct test of the psychological mechanisms underly-
ing “hanger,” our findings are not without limitations. For in-
stance, hunger was measured, but not manipulated in Studies 1 and
2, leaving open alternate interpretations of our findings (i.e., peo-
Figure 8. Study 3 mean differences in minutes spent on the self-
regulation task. Hungry versus satiated individuals did not significantly
differ in perseverance on the mental rotation task, t(149) ⫽⫺.75, p⫽.457,
95% CIs [⫺2.72, 1.23]. Error bars computed ⫹/⫺1SE.
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.
13
CONCEPTUALIZING HUNGER AS EMOTION

ple who report more intense responses on Likert scales for hunger
are also more generally affectively reactive to negative contexts).
Additionally, hunger was self-reported after the AMP task was
completed in both Studies 1 and 2 and thus relied on retrospective
ratings of hunger. However, we deemed this procedural choice
necessary as measuring hunger prior to the AMP might have
altered the effects, as demonstrated in Study 3, where attentional
focus impacted the resulting emotions.
To overcome limitations inherent with measuring hunger, we
manipulated hunger versus satiation in Study 3, but still found
considerable between-subject variation in the subjective experi-
ence of hunger even amid our objective manipulations of fasting.
To ensure the robustness of our findings and balance concerns
about reliability and validity, we analyzed our data in several
ways, all with similar results. In the “intent-to-treat” analyses
reported in our online supplemental materials (Bouwmeester et al.,
2017), we analyzed all participants who were assigned to fast
versus eat, regardless of their self-reported hunger and found the
predicted interaction between hunger and awareness on emotional
self-reports. In a second set of analyses, we removed participants
who did not pass our manipulation check for hunger and still found
the predicted interaction. Finally, we reassigned participants to the
hunger versus satiation condition with the logic that regardless of
fasting status, those participants who reported being hungrier
should be assigned to a hunger condition; here, we again replicated
the same pattern of findings. Granted, the latter two sets of find-
ings open up the possibility that participants who were excluded or
who were reassigned differ systematically from ones that do not
and this limits the interpretation that fasting and its interaction with
awareness had a causal effect on our dependent variable.
To address the possibility that we did not causally manipulate
hunger, future research should take two approaches. First, future
research could replicate and extend our manipulations by inducing
longer fasts and including more objective, markers of hunger such
as blood glucose and circulating active ghrelin, in addition to
self-reported hunger. However, it remains likely that there are
significant individual differences in the subjective experience of
hunger following fasting (e.g., Blundell et al., 2003; Jeroen Mal-
jaars et al., 2011). Future research should also measure and model
the individual differences that moderate the relationship between
fasting and hunger, because our findings suggest that fasting status
and subjective hunger are related, but can be uncoupled. One
candidate moderator is interoception or the awareness of one’s
bodily changes (Critchley & Garfinkel, 2017; for a review on
interoceptive contributions to eating and obesity, see Simmons &
DeVille, 2017).
Another potential moderator is how frequently individuals fast
or whether they have experience with fasting in the past. For
example, longitudinal evidence suggests that dieting and restricted
caloric intake increase appetite and may reduce how long an
individual can go before feeling hungry, even after 1 year of
dieting (Sumithran et al., 2011). However, individuals who under-
take intermittent fasting or restricted-calorie but high-protein diets
exhibit reduced appetite (e.g., Johnstone, Horgan, Murison, Brem-
ner, & Lobley, 2008; Wadden, Stunkard, Day, Gould, & Rubin,
1987). Nonetheless, it remains a possibility that the subjective
experience of hunger, rather than length of time since eating or
objective blood glucose levels, is ultimately what dictates whether
hunger will be conceptualized as emotion when the context
prompts such conceptualizations. This interpretation would remain
consistent with our constructionist predictions that subjective ex-
periences of body states become experienced as emotions in cer-
tain contexts when they are made meaningful as emotions rather
than body states.
These limitations notwithstanding, our findings held across
three studies using two different methodologies: two cognitive–
behavioral studies drawn from an online sample of adults ranging
in age from 18- to 71-years-old and a social psychological labo-
ratory experiment of college students. Although previous research
examines hunger’s impact on aggression, risky decisions, moral
judgments, and mood or affect more generally, the present article
provides the first evidence that hunger shifts self-reported experi-
ences of emotions, posing several important implications for how
our social and affective lives can be impacted by the homeostatic
processes that are continuously operating within our bodies.
Implications
Perhaps one of the most interesting implications of the present
findings is that hunger has the potential to adversely impact
affective judgments and experiences. This could help explain why
people dislike their spouses more when their blood sugar is low
(Bushman et al., 2014) or why they are more morally punitive
when hungry (Williams et al., 2016). However, our findings sug-
gest that hunger does not automatically lead to more negative
emotions and interpersonal judgments. Context plays a central role
in whether hunger is conceptualized as emotions, as does the focus
of a person’s attention. Given that situated inferences about the
meaning of affect tend to be relatively automatic and unconscious
(Barrett, Ochsner, & Gross, 2007; Schwarz & Clore, 1983;
Winkielman, Berridge, & Wilbarger, 2005), these studies suggest
that individuals only conceptualize hunger as emotion when not
explicitly focused on the emotional nature of their feelings.
Our findings also suggest that having an emotion label (e.g.,
“anger”) accessible could lead to the implicit regulation of emo-
tion, reducing the likelihood that hunger results in the experience
of negative, high arousal emotions or judgments that people and
objects in the world are unpleasant. Much research suggests that
drawing attention to affective feelings by putting feelings into
words (“affect labeling”) can regulate or reduce the intensity of
affect. For example, labeling one’s affective state in the moment
has been shown to reduce the intensity of physiological responses
(Kassam & Mendes, 2013; Niles et al., 2015) and self-reported
emotion (Lieberman, Inagaki, Tabibnia, & Crockett, 2011). Indi-
viduals who are better able to label their emotions in a discrete and
specific manner (i.e., who are high in emotion differentiation) are
better able to regulate their emotions (Barrett et al., 2001) and
show more chronic activation of brain networks involved in exec-
utive control (Lee, Lindquist, & Nam, 2017). A recent meta-
analysis of 356 neuroimaging studies found that the mere presence
of emotion labels in experimental tasks reduced amygdala activa-
tion, which is associated with marshaling bodily changes to affec-
tively evocative stimuli (Brooks et al., 2017).
It is sometimes assumed that this “muting” effect of affect
labeling may be because engaging in conscious reflection (e.g.,
telling a story about why Jon is angry vs. sad) causes detachment,
disrupting the more automatic, first-person flow of subjective
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.
14 MACCORMACK AND LINDQUIST

experience (cf., Lieberman, 2011). Although this account is plau-
sible in the present studies, it is also possible that drawing attention
to one’s affective state via emotion labels allowed participants to
make meaning of hunger as emotions (e.g., anger or sadness; see
Lindquist & Barrett, 2008), resulting in the subsequent regulation
of their experiences and behavior (see Lindquist, Satpute, & Gend-
ron, 2015 for a discussion). Whereas Studies 1 and 2 did not ask
individuals to label their feelings, in Study 3, participants were
randomly assigned to focus attention on and write about the
categories anger,sadness, versus no emotion in particular. As only
individuals in the Hunger ⫻No Emotion-Focused condition re-
ported significant differences in emotion experience, it may be that
the presence of “anger” and “sadness” labels disrupted the impact
of hunger on emotion because those labels made participants
realize that they were conceptualizing their hungry affect as emo-
tions and they were thus able to regulate their feelings and behav-
iors toward the situation and the researcher. Further research
should more explicitly examine the impact of affect labeling on
hunger and emotions. Future research might also examine how
conceptualizing one’s affective state as hunger or another body
state, as opposed to an emotion, influences emotional outcomes.
This may be one pathway for reducing hunger’s impact on rela-
tionships moral behaviors, and impulsivity (e.g., Anderberg et al.,
2016; Bushman et al., 2014; Williams et al., 2016).
The present findings also have important implications for emotion
theory. First, they support the constructionist view that affect is made
meaningful in context to generate instances of emotion. They also
underscore that homeostatic processes in the body can be a source of
that affect. Other emotion theories (basic emotion and causal appraisal
theories) tend to view bodily changes such as hunger and emotions as
emerging from two distinct biological systems (e.g., Ekman & Cor-
daro, 2011; Panksepp, 1998). It is thus assumed that hunger and
emotions could only have interacting and reciprocal influences on one
another, such that hunger activates discrete emotions. This interpre-
tation would not readily explain the opposite effect however: that
people sometimes confuse emotions for hunger. For instance, in cases
of “emotional eating,” individuals misinterpret emotions such as anx-
iety for hunger (Herman, Polivy, Lank, & Heatherton, 1987; McK-
enna, 1972). This finding is much more consistent with the construc-
tionist hypothesis that hunger and emotion emerge from the same
basic psychological “ingredient”—core affect.
Our findings can thus also refine the construct of core affect. It
is often assumed that affective representations (whether something
is good or bad, highly or lowly arousing) is only computed
centrally in the brain, in turn altering body states (e.g., Scherer,
2001; Zelazo & Cunningham, 2007). However, there is growing
acknowledgment for the opposite effect, such that affective repre-
sentations in the brain incorporate ongoing bodily changes (Barrett
& Simmons, 2015; Critchley & Nagai, 2012; MacCormack &
Lindquist, 2017; Russell, 2003). In this sense, bodily phenomena
that seem distally linked to the present context (e.g., hunger in the
presence of an offense) may still ultimately influence behavior in
that context. This idea has important implications for how we think
of “accurate” and “inaccurate” affective perceptions of the world
and what constitutes “attribution” and “misattribution.”
For instance, it is often assumed that when affect related to a less
proximal cause (e.g., due to hunger) influences judgments of the
context (e.g., the presence of an offense) that it is an “inaccurate”
perception of the world. Affect is thus said to be misattributed in
such situations. However, the constructionist conception of affect
breaks down the boundaries between attribution and misattribu-
tion. Constructionist accounts do not assume conscious access to
this meaning-making process, but rather that the brain is predicting
and drawing on prior knowledge and features of the situation in the
moment to construct meaning about what the body is feeling. The
brain does make prediction errors (e.g., Clark, 2013; Iglesias et al.,
2013), but assuming that all instances of “misattribution” (e.g.,
feeling “hangry”) are prediction errors overlooks the relevance of
core affect for maintaining homeostasis (i.e., allostasis). Even if
the unpleasant, highly aroused affect induced by hunger seems
irrelevant to the current situation (e.g., encountering a threatening
person), it is ultimately relevant because the current context might
have even more import for the well-being of your glucose-
deprived body. In this sense, unpleasant feelings may be amplified
when homeostasis is threatened by multiple sources (e.g., social
threat while glucose-depleted).
More broadly, given that the third study provides preliminary
evidence that hunger may interact with awareness to shift social
judgments, future research should implement more targeted study
designs to assess whether and how hunger shapes perceptions of
social others and the world more generally. For example, in
nonhuman animal models, fasting wood frog tadpoles (Lithobates
sylvaticus) are more active and less risk-averse when they hear
alarm calls from other frogs than are satiated wood frog tadpoles
(Carlson, Newman, & Langkilde, 2015). Similarly, when com-
pared with satiated fish, food-deprived fish will forage for food
further from home, even when in predator-laden waters (Damsgird
& Dill, 1998; Godin & Crossman, 1994). Schooling fish exhibit
reduced group cohesion when hungry but maintain group cohesion
when satiated, providing initial evidence that hunger impacts so-
cial behaviors in other species besides humans (Sogard & Olla,
1997). These animal studies suggest that hunger likely impacts
approach-avoidance behaviors, risk-taking, and social behaviors
such as group cohesion and synchrony more broadly. Future work
might extend on these findings to humans, who can also be more
or less aware of their body states. As hunger induces affective
changes, it is possible that hunger could influence any social–
cognitive process reliant on affect (e.g., attitudes toward out-group
vs. in-group members, affective forecasts about the future, risk
perceptions, and affect-based decisions).
Future work should also delve deeper into the biological path-
ways by which hunger changes affect. For example, the peptide
hormone ghrelin is a primary hunger signal, and early evidence
demonstrates that ghrelin administration increases sympathetic
nervous system reactivity during a stressor relative to placebo
administration (Lambert et al., 2011). If ghrelin increases sympa-
thetic activation, then this may be one pathway by which hunger
impacts momentary affect and in turn, emotional states. Other
growing work in both nonhuman animals and humans demonstrate
that resistance to leptin, a hormone produced by adipose cells
throughout the body to signal satiety, is implicated in depression
etiology (e.g., review in Lu, 2007).
The actual food that individuals ingest may be another mechanism
by which metabolic processes impact mood. For example, tryptophan,
an amino acid necessary to synthesize serotonin, can become depleted
with fasting (Altman, Shankman, & Spring, 2010). Tryptophan-
depleted individuals demonstrate greater negativity bias on the emo-
tional Stroop task, exhibit enhanced memory for negative stimuli
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.
15
CONCEPTUALIZING HUNGER AS EMOTION

(relative to neutral), and are worse at evaluating intimacy and ro-
mance in pictures of couples, relative to individuals who are not
tryptophan-depleted (Bilderbeck et al., 2011; Pringle, Cooper, Brown-
ing, & Harmer, 2012; Wang et al., 2009).
Finally, although the present studies focused on hunger, these
results may extend to other body states that induce negative affect
such as fatigue or inflammation (see MacCormack & Lindquist,
2017). Even more longstanding changes to peripheral body represen-
tations from illness, trauma, and normative development (i.e., puberty
and old age) likely have an influence on how people construct emo-
tions. For example, although diseases such as cancer certainly impact
mood via appraisals (e.g., uncertainty, existential threat), the way that
the disease itself alters homeostatic functioning and peripheral phys-
iology may also contribute to changes in mood and emotion—such as
systemic inflammation predicting depression (see review in MacCor-
mack & Lindquist, 2017). Future research should pursue how shifts in
putatively “nonemotional,” homeostatic processes—be they in daily
life, development, or disease—can drive emotions and interpersonal
processes. Such a research program would help reveal the body’s
power to shape emotions and the mind more generally.
References
Abelson, R. P., & Prentice, D. A. (1997). Contrast tests of interaction
hypotheses. Psychological Methods, 2, 315–328. http://dx.doi.org/10
.1037/1082-989X.2.4.315
Altman, S. E., Shankman, S. A., & Spring, B. (2010). Effect of acute
tryptophan depletion on emotions in individuals with personal and
family history of depression following a mood induction. Neuropsycho-
biology, 62, 171–176. http://dx.doi.org/10.1159/000319358
Alvarado, N., & Jameson, K. A. (2011). Shared knowledge about emotion
among Vietnamese and English bilingual and monolingual speakers.
Journal of Cross-Cultural Psychology, 42, 963–982. http://dx.doi.org/
10.1177/0022022110381125
Anderberg, R. H., Hansson, C., Fenander, M., Richard, J. E., Dickson,
S. L., Nissbrandt, H.,...Skibicka, K. P. (2016). The stomach-derived
hormone ghrelin increases impulsive behavior. Neuropsychopharmacol-
ogy, 41, 1199–1209. http://dx.doi.org/10.1038/npp.2015.297
Arnold, M. B. (1960). Emotion and personality. New York, NY: Columbia
University Press.
Balcetis, E., & Dunning, D. (2006). See what you want to see: Motivational
influences on visual perception. Journal of Personality and Social Psy-
chology, 91, 612–625. http://dx.doi.org/10.1037/0022-3514.91.4.612
Barnhoorn, J. S., Haasnoot, E., Bocanegra, B. R., & van Steenbergen, H.
(2015). QRTEngine: An easy solution for running online reaction time
experiments using Qualtrics. Behavior Research Methods, 47, 918–929.
http://dx.doi.org/10.3758/s13428-014-0530-7
Barrett, L. F. (2011). Was Darwin wrong about emotional expressions?
Current Directions in Psychological Science, 20, 400–406. http://dx.doi
.org/10.1177/0963721411429125
Barrett, L. F. (2017). What is an emotion? In A. S. Fox, R. C. Lapate, A. J.
Shackman, & R. J. Davidson (Eds.), The nature of emotion: Fundamen-
tal questions (2nd ed.). New York, NY: Oxford University Press.
Barrett, L. F., & Bar, M. (2009). See it with feeling: Affective predictions
during object perception. Philosophical Transactions of the Royal Soci-
ety of London Series B, Biological Sciences, 364, 1325–1334. http://dx
.doi.org/10.1098/rstb.2008.0312
Barrett, L. F., & Fossum, T. (2001). Mental representations of affect
knowledge. Cognition and Emotion, 15, 333–363. http://dx.doi.org/10
.1080/02699930125711
Barrett, L. F., Ochsner, K. N., & Gross, J. J. (2007). On the automaticity
of emotion. In J. Bargh (Ed.), Social psychology and the unconscious:
The automaticity of higher mental processes (pp. 173–217). New York,
NY: Psychology Press.
Barrett, L. F., & Simmons, W. K. (2015). Interoceptive predictions in the
brain. Nature Reviews Neuroscience, 16, 419–429. http://dx.doi.org/10
.1038/nrn3950
Baumeister, R. F. (2003). Ego depletion and self-regulation failure: A
resource model of self-control. Alcoholism: Clinical and Experimental
Research, 27, 281–284. http://dx.doi.org/10.1097/01.ALC.0000060879
.61384.A4
Baumeister, R. F. (2014). Self-regulation, ego depletion, and inhibition.
Neuropsychologia, 65, 313–319. http://dx.doi.org/10.1016/j.neuropsy-
chologia.2014.08.012
Benton, D. (2002). Carbohydrate ingestion, blood glucose and mood.
Neuroscience and Biobehavioral Reviews, 26, 293–308. http://dx.doi
.org/10.1016/S0149-7634(02)00004-0
Bilderbeck, A. C., McCabe, C., Wakeley, J., McGlone, F., Harris, T., Cowen,
P. J., & Rogers, R. D. (2011). Serotonergic activity influences the cognitive
appraisal of close intimate relationships in healthy adults. Biological Psy-
chiatry, 69, 720–725. http://dx.doi.org/10.1016/j.biopsych.2010.12.038
Blundell, J. E., Stubbs, R. J., Hughes, D. A., Whybrow, S., & King, N. A.
(2003). Cross talk between physical activity and appetite control: Does
physical activity stimulate appetite? The Proceedings of the Nutrition
Society, 62, 651–661. http://dx.doi.org/10.1079/PNS2003286
Bouwmeester, S., Verkoeijen, P. P. J. L., Aczel, B., Barbosa, F., Bègue, L.,
Brañas-Garza, P.,...Wollbrant, C. E. (2017). Registered replication
report: Rand, Greene, and Nowak (2012). Perspectives on Psychological
Science, 12, 527–542. http://dx.doi.org/10.1177/1745691617693624
Brooks, J. A., Shablack, H., Gendron, M., Satpute, A. B., Parrish, M. H.,
& Lindquist, K. A. (2017). The role of language in the experience and
perception of emotion: A neuroimaging meta-analysis. Social Cognitive
and Affective Neuroscience, 12, 169–183.
Bulmer, M. G. (1979). Principles of statistics. New York, NY: Dover
Publications.
Bushman, B. J., Dewall, C. N., Pond, R. S., Jr., & Hanus, M. D. (2014).
Low glucose relates to greater aggression in married couples. Proceed-
ings of the National Academy of Sciences of the United States of
America, 111, 6254–6257. http://dx.doi.org/10.1073/pnas.1400619111
Carlson, B. E., Newman, J. C., & Langkilde, T. (2015). Food or fear:
Hunger modifies responses to alarm cues in tadpoles. Hydrobiologia,
743, 299–308. http://dx.doi.org/10.1007/s10750-014-2048-5
Carter, E. C., Kofler, L. M., Forster, D. E., & McCullough, M. E. (2015).
A series of meta-analytic tests of the depletion effect: Self-control does
not seem to rely on a limited resource. Journal of Experimental Psy-
chology: General, 144, 796–815. http://dx.doi.org/10.1037/xge0000083
Chen, J., Papies, E. K., & Barsalou, L. W. (2016). A core eating network
and its modulations underlie diverse eating phenomena. Brain and
Cognition, 110, 20–42. http://dx.doi.org/10.1016/j.bandc.2016.04.004
Christensen, N. J., Alberti, K. G., & Brandsborg, O. (1975). Plasma
catecholamines and blood substrate concentrations: Studies in insulin
induced hypoglycaemia and after adrenaline infusions. European Jour-
nal of Clinical Investigation, 5, 415–423.
Clark, A. (2013). Whatever next? Predictive brains, situated agents, and the
future of cognitive science. Behavioral and Brain Sciences, 36, 181–
204. http://dx.doi.org/10.1017/S0140525X12000477
Clore, G. L., & Ortony, A. (2013). Psychological construction in the OCC
model of emotion. Emotion Review, 5, 335–343. http://dx.doi.org/10
.1177/1754073913489751
Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. (2003). Applied multiple
regression/correlation analysis in the behavioral sciences (3rd ed.).
Mahwah, NJ: Erlbaum.
Coker, R. H., & Kjaer, M. (2005). Glucoregulation during exercise: The
role of the neuroendocrine system. Sports Medicine, 35, 575–583. http://
dx.doi.org/10.2165/00007256-200535070-00003
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.
16 MACCORMACK AND LINDQUIST

Corrall, R. J., Frier, B. M., Davidson, N. M., Hopkins, W. M., & French,
E. B. (1983). Cholinergic manifestations of the acute autonomic reaction
to hypoglycaemia in man. Clinical Science, 64, 49–53. http://dx.doi.org/
10.1042/cs0640049
Critchley, H. D., & Garfinkel, S. N. (2017). Interoception and emotion.
Current Opinion in Psychology, 17, 7–14. http://dx.doi.org/10.1016/j
.copsyc.2017.04.020
Critchley, H. D., & Nagai, Y. (2012). How emotions are shaped by bodily
states. Emotion Review, 4, 163–168. http://dx.doi.org/10.1177/1754
073911430132
Cryer, P. E. (1999). Symptoms of hypoglycemia, thresholds for their
occurrence, and hypoglycemia unawareness. Endocrinology and Metab-
olism Clinics of North America, 28, 495–500, v–vi. http://dx.doi.org/10
.1016/S0889-8529(05)70084-0
Cunningham, W. A., Dunfeld, K. A., & Stillman, P. (2013). Emotional
states from affective dynamics. Emotion Review, 5, 344–355. http://dx
.doi.org/10.1177/1754073913489749
Dagher, A. (2009). The neurobiology of appetite: Hunger as addiction.
International Journal of Obesity, 33, S30–S33. http://dx.doi.org/10
.1038/ijo.2009.69
Damsgird, B., & Dill, L. M. (1998). Risk-taking behavior in weight-
compensating coho salmon, Oncorhynchus kisutch Behavioral Ecology,
9, 26–32. http://dx.doi.org/10.1093/beheco/9.1.26
Denson, T. F., Pedersen, W. C., Friese, M., Hahm, A., & Roberts, L.
(2011). Understanding impulsive aggression: Angry rumination and
reduced self-control capacity are mechanisms underlying the
provocation-aggression relationship. Personality and Social Psychology
Bulletin, 37, 850–862. http://dx.doi.org/10.1177/0146167211401420
DeWall, C. N., Deckman, T., Gailliot, M. T., & Bushman, B. J. (2011).
Sweetened blood cools hot tempers: Physiological self-control and aggres-
sion. Aggressive Behavior, 37, 73–80. http://dx.doi.org/10.1002/ab.20366
DeWall, C. N., Pond, R. S., Jr., & Bushman, B. J. (2010). Sweet revenge:
Diabetic symptoms predict less forgiveness. Personality and Individual
Differences, 49, 823–826. http://dx.doi.org/10.1016/j.paid.2010.06.030
Doyle, C. M., & Lindquist, K. A. (2017). Language and emotion: Hypoth-
eses on the constructed nature of emotion perception. In J. M.
Fernandez-Dols & J. A. Russell (Eds.), The Science of Facial Expression
(pp. 415–432). New York, NY: Oxford. http://dx.doi.org/10.1093/
acprof:oso/9780190613501.003.0022
Dutton, D. G., & Aron, A. P. (1974). Some evidence for heightened sexual
attraction under conditions of high anxiety. Journal of Personality and
Social Psychology, 30, 510–517. http://dx.doi.org/10.1037/h0037031
Ekman, P., & Cordaro, D. (2011). What is meant by calling emotions basic.
Emotion Review, 3, 364–370. http://dx.doi.org/10.1177/1754073911
410740
Ellsworth, P. C., & Scherer, K. R. (2003). Appraisal processes in emotion. In
R. J. Davidson, K. R. Scherer, & H. H. Goldsmith (Eds.), Handbook of
affective sciences (pp. 572–595). Oxford, UK: Oxford University Press.
Fairclough, S. H., & Houston, K. (2004). A metabolic measure of mental
effort. Biological Psychology, 66, 177–190. http://dx.doi.org/10.1016/j
.biopsycho.2003.10.001
Faul, F., Erdfelder, E., Lang, A.-G., & Buchner, A. (2007). G
ⴱ
Power 3: A
flexible statistical power analysis program for the social, behavioral, and
biomedical sciences. Behavior Research Methods, 39, 175–191. http://
dx.doi.org/10.3758/BF03193146
Feldman, L. A. (1995). Variations in the circumplex structure of mood.
Personality and Social Psychology Bulletin, 21, 806–817. http://dx.doi
.org/10.1177/0146167295218003
Fredrickson, B. L., Tugade, M. M., Waugh, C. E., & Larkin, G. R. (2003).
What good are positive emotions in crises? A prospective study of
resilience and emotions following the terrorist attacks on the United
States on September 11th, 2001. Journal of Personality and Social
Psychology, 84, 365–376. http://dx.doi.org/10.1037/0022-3514.84.2.365
Fridlund, A. (1994). Human facial expression: An evolutionary view. San
Diego, CA: Academic Press, Inc.
Gailliot, M. T., & Baumeister, R. F. (2007). The physiology of willpower:
Linking blood glucose to self-control. Personality and Social Psychol-
ogy Review, 11, 303–327. http://dx.doi.org/10.1177/1088868307303030
Gailliot, M. T., Baumeister, R. F., DeWall, C. N., Maner, J. K., Plant,
E. A., Tice, D. M.,...Schmeichel, B. J. (2007). Self-control relies on
glucose as a limited energy source: Willpower is more than a metaphor.
Journal of Personality and Social Psychology, 92, 325–336. http://dx
.doi.org/10.1037/0022-3514.92.2.325
Gendron, M., Lindquist, K. A., & Barrett, L. F. (2011). Ratings of IASLab
facial expression stimuli [Unpublished data]. Retrieved from http://
www.affective-science.org/
Godin, J. J., & Crossman, S. L. (1994). Hunger-dependent predator in-
spection and foraging behaviours in the threespine stickleback (Gaster-
osteus aculeatus) under predation risk. Behavioral Ecology and Socio-
biology, 34, 359–366. http://dx.doi.org/10.1007/BF00197006
Hall, J. B., & Brown, D. A. (1979). Plasma glucose and lactic acid
alterations in response to a stressful exam. Biological Psychology, 8,
179–188. http://dx.doi.org/10.1016/0301-0511(79)90046-2
Hangry. (2015). Oxford dictionaries. Retrieved from http://www
.oxforddictionaries.com/us/definition/american_english/hangry
Heller, S. R., Macdonald, I. A., Herbert, M., & Tattersall, R. B. (1987).
Influence of sympathetic nervous system on hypoglycaemic warning
symptoms. Lancet, 2, 359–363. http://dx.doi.org/10.1016/S0140-6736
(87)92382-8
Hepburn, D. A., Deary, I. J., & Frier, B. M. (1992). Classification of
symptoms of hypoglycaemia in insulin-treated diabetic patients using
factor analysis: Relationship to hypoglycaemia unawareness. Dia-
betic Medicine, 9, 70–75. http://dx.doi.org/10.1111/j.1464-5491
.1992.tb01718.x
Hepburn, D. A., Deary, I. J., Munoz, M., & Frier, B. M. (1994). Physio-
logical manipulations of psychometric mood factors using acute insulin-
induced hypoglycemia in humans. Personality and Individual Differ-
ences, 18, 385–391. http://dx.doi.org/10.1016/0191-8869(94)00162-L
Herbert, B. M., Herbert, C., Pollatos, O., Weimer, K., Enck, P., Sauer, H., &
Zipfel, S. (2012). Effects of short-term food deprivation on interoceptive
awareness, feelings and autonomic cardiac activity. Biological Psychology,
89, 71–79. http://dx.doi.org/10.1016/j.biopsycho.2011.09.004
Herman, C. P., Polivy, J., Lank, C. N., & Heatherton, T. F. (1987).
Anxiety, hunger, and eating behavior. Journal of Abnormal Psychology,
96, 264–269. http://dx.doi.org/10.1037/0021-843X.96.3.264
Hermanns, N., Kubiak, T., Kulzer, B., & Haak, T. (2003). Emotional
changes during experimentally induced hypoglycaemia in type 1 diabe-
tes. Biological Psychology, 63, 15–44. http://dx.doi.org/10.1016/S0301-
0511(03)00027-9
Hermanns, N., Scheff, C., Kulzer, B., Weyers, P., Pauli, P., Kubiak, T., &
Haak, T. (2007). Association of glucose levels and glucose variability
with mood in type 1 diabetic patients. Diabetologia, 50, 930–933.
http://dx.doi.org/10.1007/s00125-007-0643-y
Iglesias, S., Mathys, C., Brodersen, K. H., Kasper, L., Piccirelli, M., den
Ouden, H. E., & Stephan, K. E. (2013). Hierarchical prediction errors in
midbrain and basal forebrain during sensory learning. Neuron, 80, 519–
530. http://dx.doi.org/10.1016/j.neuron.2013.09.009
Inzlicht, M., Schmeichel, B. J., & Macrae, C. N. (2014). Why self-control
seems (but may not be) limited. Trends in Cognitive Sciences, 18,
127–133. http://dx.doi.org/10.1016/j.tics.2013.12.009
Jeroen Maljaars, P. W., Peters, H. P. F., Kodde, A., Geraedts, M., Troost,
F. J., Haddeman, E., & Masclee, A. A. M. (2011). Length and site of the
small intestine exposed to fat influences hunger and food intake. British
Journal of Nutrition, 106, 1609–1615. http://dx.doi.org/10.1017/S000
7114511002054
Job, V., Walton, G. M., Bernecker, K., & Dweck, C. S. (2013). Beliefs about
willpower determine the impact of glucose on self-control. Proceedings of
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.
17
CONCEPTUALIZING HUNGER AS EMOTION

the National Academy of Sciences of the United States of America, 110,
14837–14842. http://dx.doi.org/10.1073/pnas.1313475110
Johnstone, A. M., Horgan, G. W., Murison, S. D., Bremner, D. M., & Lobley,
G. E. (2008). Effects of a high-protein ketogenic diet on hunger, appetite,
and weight loss in obese men feeding ad libitum. The American Journal of
Clinical Nutrition, 87, 44–55. http://dx.doi.org/10.1093/ajcn/87.1.44
Kassam, K. S., & Mendes, W. B. (2013). The effects of measuring
emotion: Physiological reactions to emotional situations depend on
whether someone is asking. PLoS ONE, 8, e64959. http://dx.doi.org/10
.1371/journal.pone.0064959
Keltner, D., Ellsworth, P. C., & Edwards, K. (1993). Beyond simple
pessimism: Effects of sadness and anger on social perception. Journal of
Personality and Social Psychology, 64, 740–752. http://dx.doi.org/10
.1037/0022-3514.64.5.740
Kirkland, T., & Cunningham, W. A. (2012). Mapping emotions through
time: How affective trajectories inform the language of emotion. Emo-
tion, 12, 268–282. http://dx.doi.org/10.1037/a0024218
Kurzban, R. (2010). Does the brain consume additional glucose during
self-control tasks? Evolutionary Psychology, 8, 244–259. http://dx.doi
.org/10.1177/147470491000800208
Lambert, E., Lambert, G., Ika-Sari, C., Dawood, T., Lee, K., Chopra, R.,
. . . Schlaich, M. P. (2011). Ghrelin modulates sympathetic nervous
system activity and stress response in lean and overweight men. Hyper-
tension, 58, 43–50. http://dx.doi.org/10.1161/HYPERTENSIONAHA
.111.171025
Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (2008). International
affective picture system (IAPS): Affective ratings of pictures and instruc-
tion manual (Technical Report A-8). Gainesville, FL: University of
Florida.
Lawton, C. L., Delargy, H. J., Brockman, J., Smith, F. C., & Blundell, J. E.
(2000). The degree of saturation of fatty acids influences post-ingestive
satiety. British Journal of Nutrition, 83, 473–482.
Lee, J. Y., Lindquist, K. A., & Nam, C. S. (2017). Emotional granularity
effects on event-related brain potentials during affective picture process-
ing. Frontiers in Human Neuroscience, 11, 133. http://dx.doi.org/10
.3389/fnhum.2017.00133
Lee, K. M., Lindquist, K. A., & Payne, B. K. (2017). Constructing bias:
Conceptualization breaks the link between implicit bias and fear of
Black Americans. Emotion. Advance online publication. http://dx.doi
.org/10.1037/emo0000347
Lerner, J. S., Small, D. A., & Loewenstein, G. (2004). Heart strings and
purse strings: Carryover effects of emotions on economic decisions.
Psychological Science, 15, 337–341. http://dx.doi.org/10.1111/j.0956-
7976.2004.00679.x
Lieberman, M. D. (2011). Why symbolic processing of affect can disrupt
negative affect: Social cognitive and affective neuroscience investiga-
tions. To appear. In A. Todorov, S. T. Fiske, & D. Prentice (Eds.), Social
neuroscience: Toward understanding the underpinnings of the social
mind (pp. 188–209). New York, NY: Oxford University Press. http://
dx.doi.org/10.1093/acprof:oso/9780195316872.003.0013
Lieberman, M. D., Eisenberger, N. I., Crockett, M. J., Tom, S. M., Pfeifer,
J. H., & Way, B. M. (2007). Putting feelings into words: Affect labeling
disrupts amygdala activity in response to affective stimuli. Psychological
Science, 18, 421–428. http://dx.doi.org/10.1111/j.1467-9280.2007.01916.x
Lieberman, M. D., Inagaki, T. K., Tabibnia, G., & Crockett, M. J. (2011).
Subjective responses to emotional stimuli during labeling, reappraisal,
and distraction. Emotion, 11, 468–480. http://dx.doi.org/10.1037/
a0023503
Lindquist, K. A. (2013). Emotions emerge from more basic psychological
ingredients: A modern psychological constructionist approach. Emotion
Review, 5, 356–368. http://dx.doi.org/10.1177/1754073913489750
Lindquist, K. A., & Barrett, L. F. (2008). Constructing emotion: The
experience of fear as a conceptual act. Psychological Science, 19,
898–903. http://dx.doi.org/10.1111/j.1467-9280.2008.02174.x
Lindquist, K. A., & Barrett, L. F. (2012). A functional architecture of the
human brain: Emerging insights from the science of emotion. Trends in
Cognitive Sciences, 16, 533–540. http://dx.doi.org/10.1016/j.tics.2012
.09.005
Lindquist, K. A., & Gendron, M. (2013). What’s in a word? Language
constructs emotion perception. Emotion Review, 5, 66–71. http://dx.doi
.org/10.1177/1754073912451351
Lindquist, K. A., Satpute, A. B., & Gendron, M. (2015). Does language do
more than communicate emotion? Current Directions in Psychological
Science, 24, 99–108. http://dx.doi.org/10.1177/0963721414553440
Lindquist, K. A., Satpute, A. B., Wager, T. D., Weber, J., & Barrett, L. F.
(2016). The brain basis of positive and negative affect: Evidence from a
meta-analysis of the human neuroimaging literature. Cerebral Cortex,
26, 1910–1922. http://dx.doi.org/10.1093/cercor/bhv001
Lindquist, K. A., Wager, T. D., Kober, H., Bliss-Moreau, E., & Barrett,
L. F. (2012). The brain basis of emotion: A meta-analytic review.
Behavioral and Brain Sciences, 35, 121–143. http://dx.doi.org/10.1017/
S0140525X11000446
Loewenstein, G. (2000). Emotions in economic theory and economic
behavior. Preferences, Behavior, and Welfare, 90, 426–432.
Lu, X.-Y. (2007). The leptin hypothesis of depression: A potential link
between mood disorders and obesity? Current Opinion in Pharmacol-
ogy, 7, 648–652. http://dx.doi.org/10.1016/j.coph.2007.10.010
MacCormack, J. K., & Lindquist, K. A. (2017). The constitutive role of the
body in emotion: Schachter’s legacy for a psychological constructionist
view on emotion. Emotion Review, 9, 36–56. http://dx.doi.org/10.1177/
1754073916639664
Marks, V., & Rose, F. G. (1981). Hypoglycaemia (2nd ed.). Oxford, UK:
Blackwell Scientific Publications.
McKenna, R. J. (1972). Some effects of anxiety level and food cues on the
eating behavior of obese and normal subjects: A comparison of the
Schachterian and psychosomatic conceptions. Journal of Personality
and Social Psychology, 22, 311–319. http://dx.doi.org/10.1037/h00
32925
Miles, E., Sheeran, P., Baird, H., Macdonald, I., Webb, T. L., & Harris,
P. R. (2016). Does self-control improve with practice? Evidence from a
six-week training program. Journal of Experimental Psychology: Gen-
eral, 145, 1075–1091. http://dx.doi.org/10.1037/xge0000185
Muraven, M., & Baumeister, R. F. (2000). Self-regulation and depletion of
limited resources: Does self-control resemble a muscle? Psychological
Bulletin, 126, 247–259. http://dx.doi.org/10.1037/0033-2909.126.2.247
Niles, A. N., Craske, M. G., Lieberman, M. D., & Hur, C. (2015). Affect
labeling enhances exposure effectiveness for public speaking anxiety.
Behaviour Research and Therapy, 68, 27–36. http://dx.doi.org/10.1016/
j.brat.2015.03.004
Panksepp, J. (1998). Affective neuroscience: The foundations of human and
animal emotions. New York, NY: Oxford University Press.
Payne, B. K., Cheng, C. M., Govorun, O., & Stewart, B. D. (2005). An
inkblot for attitudes: Affect misattribution as implicit measurement.
Journal of Personality and Social Psychology, 89, 277–293. http://dx
.doi.org/10.1037/0022-3514.89.3.277
Payne, B. K., Hall, D. L., Cameron, C. D., & Bishara, A. J. (2010). A
process model of affect misattribution. Personality and Social Psychol-
ogy Bulletin, 36, 1397–1408. http://dx.doi.org/10.1177/014616721
0383440
Payne, B. K., & Lundberg, K. (2014). The affect misattribution procedure:
Ten years of evidence on reliability, validity, and mechanisms. Social
and Personality Psychology Compass, 8, 672–686. http://dx.doi.org/10
.1111/spc3.12148
Pennebaker, J. W., Booth, R. J., & Francis, M. E. (2015). Linguistic Inquiry
and Word Count: LIWC 2015. Austin, TX: LIWC.
Peters, A., Schweiger, U., Pellerin, L., Hubold, C., Oltmanns, K. M.,
Conrad, M.,...Fehm, H. L. (2004). The selfish brain: Competition for
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.
18 MACCORMACK AND LINDQUIST

energy resources. Neuroscience and Biobehavioral Reviews, 28, 143–
180. http://dx.doi.org/10.1016/j.neubiorev.2004.03.002
Pohle-Krauza, R. J., Carey, K. H., & Pelkman, C. L. (2008). Dietary
restraint and menstrual cycle phase modulated L-phenylalanine-induced
satiety. Physiology & Behavior, 93, 851–861. http://dx.doi.org/10.1016/
j.physbeh.2007.11.051
Porubská, K., Veit, R., Preissl, H., Fritsche, A., & Birbaumer, N. (2006).
Subjective feeling of appetite modulates brain activity: An fMRI study.
NeuroImage, 32, 1273–1280. http://dx.doi.org/10.1016/j.neuroimage
.2006.04.216
Pringle, A., Cooper, M. J., Browning, M., & Harmer, C. J. (2012). Effects
of low dose tryptophan depletion on emotional processing in dieters.
Eating Behaviors, 13, 154–157. http://dx.doi.org/10.1016/j.eatbeh.2011
.12.005
Raudenbush, S. W., & Bryk, A. S. (2001). Hierarchical linear models:
Applications and data analysis methods (2nd ed.). Thousand Oaks, CA:
Sage.
Russell, J. A. (1980). A circumplex model of affect. Journal of Personality
and Social Psychology, 39, 1161–1178. http://dx.doi.org/10.1037/
h0077714
Russell, J. A. (2003). Core affect and the psychological construction of
emotion. Psychological Review, 110, 145–172. http://dx.doi.org/10
.1037/0033-295X.110.1.145
Russell, J. A., & Bullock, M. (1986). Fuzzy concepts and the perceptions
of emotion in facial expressions. Social Cognition, 4, 309–341. http://
dx.doi.org/10.1521/soco.1986.4.3.309
Schachter, S., & Singer, J. E. (1962). Cognitive, social, and physiological
determinants of emotional state. Psychological Review, 69, 379–399.
http://dx.doi.org/10.1037/h0046234
Scherbaum, C. A., & Ferreter, J. M. (2009). Estimating statistical power
and required sample sizes for organizational research using multilevel
modeling. Organizational Research Methods, 12, 347–367. http://dx.doi
.org/10.1177/1094428107308906
Scherer, K. R. (2001). Appraisal considered as a process of multi-level
sequential checking. In K. R. Scherer, A. Schorr, & T. Johnstone (Eds.),
Appraisal processes in emotion: Theory, methods, research (pp. 92–
120). New York, NY: Oxford University Press.
Schwarz, N., & Clore, G. L. (1983). Mood, misattribution, and judgments
of well-being: Informative and directive functions of affective states.
Journal of Personality and Social Psychology, 45, 513–523. http://dx
.doi.org/10.1037/0022-3514.45.3.513
Shaked, A., & Clore, G. L. (2017). Breaking the world to make it whole
again: Attribution in the construction of emotion. Emotion Review, 9,
27–35. http://dx.doi.org/10.1177/1754073916658250
Shaver, P., Schwartz, J., Kirson, D., & O’Connor, C. (1987). Emotion
knowledge: Further exploration of a prototype approach. Journal of
Personality and Social Psychology, 52, 1061–1086. http://dx.doi.org/10
.1037/0022-3514.52.6.1061
Shepard, R. N., & Metzler, J. (1971). Mental rotation of three-dimensional
objects. Science, 171, 701–703. http://dx.doi.org/10.1126/science.171
.3972.701
Simmons, W. K., & DeVille, D. C. (2017). Interoceptive contributions to
healthy eating and obesity. Current Opinion in Psychology, 17, 106–
112. http://dx.doi.org/10.1016/j.copsyc.2017.07.001
Sogard, S. M., & Olla, B. L. (1997). The influence of hunger and predation
risk on group cohesion in a pelagic fish, walleye pollock Theragra
chalcogramma.Environmental Biology of Fishes, 50, 405–413. http://
dx.doi.org/10.1023/A:1007393307007
Sumithran, P., Prendergast, L. A., Delbridge, E., Purcell, K., Shulkes, A.,
Kriketos, A., & Proietto, J. (2011). Long-term persistence of hormonal
adaptations to weight loss. The New England Journal of Medicine, 365,
1597–1604. http://dx.doi.org/10.1056/NEJMoa1105816
Symmonds, M., Emmanuel, J. J., Drew, M. E., Batterham, R. L., & Dolan,
R. J. (2010). Metabolic state alters economic decision making under risk
in humans. PLoS ONE, 5, e11090. http://dx.doi.org/10.1371/journal
.pone.0011090
Tataranni, P. A., Gautier, J. F., Chen, K., Uecker, A., Bandy, D., Salbe,
A.D.,...Ravussin, E. (1999). Neuroanatomical correlates of hunger
and satiation in humans using positron emission tomography. Proceed-
ings of the National Academy of Sciences of the United States of
America, 96, 4569–4574. http://dx.doi.org/10.1073/pnas.96.8.4569
Taylor, L. A., & Rachman, S. J. (1988). The effects of blood sugar level
changes on cognitive function, affective state, and somatic symptoms.
Journal of Behavioral Medicine, 11, 279–291. http://dx.doi.org/10
.1007/BF00844433
Thomas, K. A., & Clifford, S. (2017). Validity and mechanical turk.
Computers in Human Behavior, 77, 184–197. http://dx.doi.org/10.1016/
j.chb.2017.08.038
Vadillo, M. A., Gold, N., & Osman, M. (2016). The bitter truth about sugar
and willpower: The limited evidential value of the glucose model of ego
depletion. Psychological Science, 27, 1207–1214. http://dx.doi.org/10
.1177/0956797616654911
Van Kleef, G. A. (2009). How emotions regulate social life. Current
Directions in Psychological Science, 18, 184–188. http://dx.doi.org/10
.1111/j.1467-8721.2009.01633.x
Vohs, K. D., Baumeister, R. F., Schmeichel, B. J., Twenge, J. M., Nelson,
N. M., & Tice, D. M. (2014). Making choices impairs subsequent
self-control: A limited-resource account of decision making, self-
regulation, and active initiative. Motivation Science, 1, 19–42. http://dx
.doi.org/10.1037/2333-8113.1.S.19
Wadden, T. A., Stunkard, A. J., Day, S. C., Gould, R. A., & Rubin, C. J.
(1987). Less food, less hunger: Reports of appetite and symptoms in a
controlled study of a protein-sparing modified fast. International Jour-
nal of Obesity, 11, 239–249.
Wang, L., Mullette-Gillman, O. A., Gadde, K. M., Kuhn, C. M., McCarthy,
G., & Huettel, S. A. (2009). The effect of acute tryptophan depletion on
emotional distraction and subsequent memory. Social Cognitive and
Affective Neuroscience, 4, 357–368. http://dx.doi.org/10.1093/scan/
nsp025
Williams, E. F., Pizarro, D., Ariely, D., & Weinberg, J. D. (2016). The
Valjean effect: Visceral states and cheating. Emotion, 16, 897–902.
http://dx.doi.org/10.1037/emo0000158
Williams, L. A., & DeSteno, D. (2008). Pride and perseverance: The
motivational role of pride. Journal of Personality and Social Psychol-
ogy, 94, 1007–1017. http://dx.doi.org/10.1037/0022-3514.94.6.1007
Wilson-Mendenhall, C. D., Barrett, L. F., & Barsalou, L. W. (2013).
Situating emotional experience. Frontiers in Human Neuroscience, 7,
764. http://dx.doi.org/10.3389/fnhum.2013.00764
Winkielman, P., Berridge, K. C., & Wilbarger, J. L. (2005). Unconscious
affective reactions to masked happy versus angry faces influence con-
sumption behavior and judgments of value. Personality and Social
Psychology Bulletin, 31, 121–135. http://dx.doi.org/10.1177/014616
7204271309
Zelazo, P. D., & Cunningham, W. A. (2007). Executive function: Mech-
anisms underlying emotion regulation. In J. Gross (Ed.), Handbook of
emotion regulation (pp. 135–158). New York, NY: Guilford Press.
Zillmann, D. (1971). Excitation transfer in communication-mediated ag-
gressive behavior. Journal of Experimental Social Psychology, 7, 419–
434. http://dx.doi.org/10.1016/0022-1031(71)90075-8
Received June 10, 2017
Revision received November 30, 2017
Accepted December 27, 2017 䡲
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.
19
CONCEPTUALIZING HUNGER AS EMOTION