Personality and Individual Differences 97 (2016) 239–244

Contents lists available at ScienceDirect

Personality and Individual Differences

journal homepage: www.elsevier.com/locate/paid

Associations between subtypes of social withdrawal and emotional

eating during emerging adulthood
Rebecca G. Etkin ⁎, Julie C. Bowker, Matthew D. Scalco
University at Buffalo, SUNY, USA

a r t i c l e i n f o a b s t r a c t

Article history: While the psychosocial difficulties associated with one specific type of social withdrawal, shyness, have been
Received 19 October 2015 extensively studied, less is known about the correlates of other subtypes, such as preference-for-solitude. Of
Received in revised form 14 March 2016 the existing studies on withdrawal subtypes, few focus on the emerging adulthood developmental period, and
Accepted 18 March 2016
none have examined possible physical health-related correlates and associated mechanisms. This study
Available online xxxx
considered whether two withdrawal subtypes (shyness, preference-for-solitude) are associated with emotional
Keywords:
eating vis-à-vis internalizing problems during emerging adulthood. Participants included 643 emerging adults
Shyness (283 males; Mage = 19.61) who completed measures of withdrawal subtypes, emotional eating, and internalizing
Preference-for-solitude problems (depression, loneliness, social anxiety). Path models revealed that the associations between both
Emotional eating shyness and preference-for-solitude and emotional eating were explained, in part, by depression and social
Internalizing problems anxiety. Findings suggest that withdrawing from peers during emerging adulthood, due to fear or preferences-
Emerging adulthood for-solitude, may have significant consequences for both physical health outcomes vis-à-vis psychological
difficulties.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction during this developmental stage is underscored by findings linking

It is well-known that social withdrawal (i.e., a behavioral tendency
to engage solitary behavior in the company of familiar and unfamiliar
peers) during childhood, adolescence, and emerging adulthood is a
strong risk factor for psychosocial maladjustment (i.e., depression,
anxiety, peer victimization; Rubin, Coplan, & Bowker, 2009). Little is
known, however, about the physical health-related concomitants of
social withdrawal. There is some evidence that youth who are rejected
and excluded by their peers (and as a consequence, spend considerable
time alone) are at risk for the development and maintenance of obesity-
and physical health-related difficulties (Salvy, de la Haye, Bowker, &
Hermans, 2012). When alone and away from peers, such youth appear
to overeat and “miss out” on opportunities for physical activity with
peers. These recent findings raise the intriguing possibility that youth
who are socially withdrawn and actively avoid their peers, not because
of peer difficulties but due to shyness or a preference-for-solitude, may
also be at risk for obesity-related health difficulties, including emotional
eating (i.e., eating in response to negative emotions; Bruch, 1973).
In this investigation, we consider whether two types of withdrawal
(shyness, preference-for-solitude) are related to emotional eating during
emerging adulthood (18–25 years). Investigators have not yet evaluat-
ed these associations but the importance of studying emotional eating
⁎ Corresponding author at: University at Buffalo, The State University of New York,
Department of Psychology, 306 Park Hall, Buffalo, NY 14260, USA
E-mail address: retkin@buffalo.edu (R.G. Etkin).
emotional eating with numerous physical health problems, including
obesity and eating disorder symptomatology, which often persist into
later adulthood. Mediation models in which the withdrawal subtypes
lead to emotional eating vis-à-vis internalizing problems are also
evaluated in light of theory and research suggesting that emotional
eating often serves as a (non-adaptive) means of regulating or coping
with unpleasant or threatening emotions such as loneliness and sadness
(Macht, 2008). Understanding if and why withdrawn emerging adults
are at risk for engaging in maladaptive eating behaviors has the
potential to extend etiological risk models of social withdrawal, which
to date, focus exclusively on social, social-cognitive, and psychological
outcomes (Rubin et al., 2009).
1.1. Social withdrawal subtypes and adjustment difficulties
Most developmental research has focused on withdrawn youth who
are shy (Rubin et al., 2009); these youth want to be with their peers but
are too fearful or anxious to do so. More recent work, however, has
broadened to consider two additional withdrawal subtypes, unsociabil-
ity and avoidance. Whereas unsociable youth have non-fearful prefer-
ences for solitude, avoidant youth withdraw from peers because they
actively dislike the company of others (Rubin et al., 2009). It is argued
that these two latter types of withdrawal may be best studied together
and as reflecting a preference-for-solitude, as they both involve an
affinity for being alone (Wang, Rubin, Laursen, Booth-LaForce, & Rose-
Krasnor, 2013). Whether unsociability and avoidance are examined

http://dx.doi.org/10.1016/j.paid.2016.03.059
0191-8869/© 2016 Elsevier Ltd. All rights reserved.
240 R.G. Etkin et al. / Personality and Individual Differences 97 (2016) 239–244
independently or jointly (i.e., preference-for-solitude), growing
evidence indicates that these constructs are distinct from (albeit associ-
ated with) shyness (e.g. Bowker & Raja, 2011; Nelson, 2013; Wang et al.,
2013). However, the extant research suggests that the risks are greater
for shyness than preference-for-solitude, at least during childhood and
early adolescence (Rubin et al., 2009).
One notable limitation of recent research on withdrawal subtypes is
that most prior research has focused on childhood and early adoles-
cence (10–14 years) to the neglect of withdrawal during emerging
adulthood. However, certain types of withdrawal, such as shyness,
may be especially problematic during emerging adulthood because
such behaviors likely interfere with the successful completion of
stage-salient developmental tasks, such as forming intimate relation-
ships, going to college, and obtaining employment (e.g., Bowker,
Nelson, Markovic, & Luster, 2014). Indeed, Nelson (2013) found that
shy (and also avoidant) emerging adults endorsed higher levels of a
number of different psychosocial problems (e.g., depression, relation-
ship difficulties, emotion dysregulation) relative to non-withdrawn
emerging adults. Unsociable emerging adults reported only elevated
levels of depression. However, Wang et al. (2013) found that shyness,
but not preference-for-solitude, was associated with internalizing diffi-
culties (e.g., depression, anxiety) and emotion dysregulation during late
adolescence (which partially overlaps with emerging adulthood). To
our knowledge, these are the only two studies that have compared
the concomitants of withdrawal subtypes beyond the late childhood/
early adolescent years, with inconclusive results. Thus, additional
research is clearly needed to clarify the degree of risk associated with
different types of withdrawal during emerging adulthood and
determine whether shyness also carries the greatest risk at this time.
As noted previously, another limitation of the aforementioned
studies is that the withdrawal subtypes were considered only in relation
to psychosocial outcomes. However, there is some indication that
shyness is also related to negative physical health outcomes in adoles-
cence and emerging adulthood. For instance, Page, Hammermeister,
Scanlan, and Allen (1996) found shyness to predict frequent TV viewing,
which is a risk factor for obesity-related problems such as inactivity and
overeating. Miller, Schmidt, and Vaillancourt (2008) found significant
associations between shyness and eating disorder symptoms in a sam-
ple of female undergraduate students. Aside from these studies, noth-
ing, to our knowledge, is known about the physical concomitants of
actively withdrawing from and avoiding peers due to fearful and non-
fearful motivations.
1.2. Emotional eating and social withdrawal subtypes
In this study, we evaluate whether shyness and preference-for-
solitude are related to emotional eating. Theories of emotional eating
posit that many individuals eat not due to hunger, but in an attempt
to distract from, cope with, or regulate negative or threatening emotion-
al states or experiences (Bruch, 1973; Heatherton & Baumeister, 1991;
Macht, 2008). Recent research offers compelling evidence for these
theories; for instance, Evers, Marijn Stok, and de Ridder (2010) found
that non-adaptive emotion regulation strategies, including cognitive
suppression, led to increased eating following a negative mood induc-
tion. Raspopow et al. (2013) and Spoor et al. (2007) found that coping
strategies typically conceptualized as non-adaptive due to associations
with psychological difficulties (avoidance-distraction coping, emotion-
focused coping) mediate associations between social and emotional
problems (unsupportive social interactions, negative affect) and emo-
tional eating in clinical and non-clinical samples. These authors argue
that certain individuals resort to emotional eating when they are unable
to successfully regulate or escape their emotions (due to a lack of adap-
tive strategies, such as problem-focused coping), and come to associate
emotional eating with reductions in their unpleasant feelings
(Raspopow et al., 2013; Spoor et al., 2007). Despite positive expectan-
cies that eating with help regulate emotions and facilitate coping,
emotional eating has been implicated as a correlate and precursor to se-
rious mental- and physical-health problems including depression, binge
eating disorder, and obesity (e.g. Braet et al., 2008; Ganley, 1989; Stice,
Presnell, & Spangler, 2002).
Given the reasons why individuals emotionally eat, there is some
reason to expect direct and indirect associations between emotional
eating and shyness and preference-for-solitude. Because emotional
eating and overeating are more likely to occur when individuals are
alone (e.g., Ganley, 1989), it is possible that youth who are both shy
and prefer solitude, who, by definition, spend considerable time alone,
are at risk for emotional eating. Yet, indirect effects are also plausible.
Shy emerging adults experience significant levels of internalizing
distress (in the form of anxiety, depression, and loneliness; Nelson
et al., 2008; Wang et al., 2013) and also difficulties with emotion
regulation (unlike youth who prefer solitude; Nelson, 2013), all of
which are associated with emotional eating (e.g., Macht, 2008). For
these reasons, we hypothesized that the linkage between shyness and
emotional eating might be especially strong, and that shy youth may
emotionally eat because they have difficulty regulating the internalizing
feelings they experience. As there is some evidence that preference-for-
solitude is associated with depressive feelings (Nelson, 2013), we
hypothesized that this specific type of emotional distress may also
account for associations with emotional eating. The present study is
the first, to our knowledge, to evaluate whether internalizing problems
account for linkages between these withdrawal subtypes and emotional
eating.
1.3. Study summary
The present study extends past research by examining if withdrawal
subtypes during emerging adulthood are associated with emotional
eating, and whether these associations are mediated by internalizing
difficulties (depression, loneliness, social anxiety). Main effects were
expected for both shyness and preference-for-solitude, and mediation
models were expected to be supported for shyness by way of all three
internalizing problems considered, but only supported for preference-
for-solitude vis-à-vis depression. All hypotheses were tentative, howev-
er, due to the lack of past research, and with those published studies
yielding conflicting findings. To test the unique effects of actively
avoiding peers, peer isolation was controlled. Gender differences were
also explored but no specific predictions were made, as emotional
eating is typically more common among females (e.g., Braet et al.,
2008) but evidence is mixed regarding the frequency and costs of
withdrawal for males and females during emerging adulthood (e.g.
Nelson, 2013; Nelson et al., 2008). Finally, given significant relations
between emotional eating and weight gain in several studies (e.g.
Ganley, 1989; Stice et al., 2002), BMI was considered as a covariate of
emotional eating.
2. Method
2.1. Participants and procedure
Participants were 643 undergraduate students (283 males; Mage =
19.61 years, SD = 1.25) enrolled in an introductory psychology course
at a large public university in the northeast United States. Approximate-
ly 41% of participants self-identified as Caucasian, 41% North-East or
South Asian, 8% Black/African-American, 4% Hispanic/Latino, and 6% as
“other.” Participants attended one 45-min laboratory visit conducted
by trained research assistants. Written consent was obtained from all
participants, and participants were informed that they were free to
withdraw from the study at any time. During the laboratory visit, partic-
ipants completed a series of self-report measures administered using
SurveyMonkey.com, and their height and weight were measured. All
participants were awarded course credit for their participation. Study
procedures were approved by the university Institutional Review Board.
 R.G. Etkin et al. / Personality and Individual Differences 97 (2016) 239–244 241

2.2. Measures 3. Results

2.2.1. Social withdrawal subtypes
Participants completed the 21-item revised Child Social Preference
Scale (Bowker & Raja, 2011) by indicating how much they were like
each item on a 5-point Likert scale (1 = “Not at all” to 5 = “A lot”).
For the present study, items were revised to be appropriate for emerg-
ing adulthood samples (e.g., “I'd like to play with other kids, but I'm
sometimes nervous to” was revised to read “I'd like to hang out with
others, but I'm sometimes nervous to”). In addition, two items each
descriptive of avoidance (e.g., “I often try to spend time alone because
I don't like to be with others”), shyness (e.g., “I often watch people
hanging out, but I don't try to join in”), and unsociability (e.g., “I don't
have a strong preference for being alone or with others”) were added
in an attempt to improve reliability of the factors reported by Bowker
and Raja (2011). Given these revisions, the 21 items were subjected to
exploratory factor analyses, with results reported below.
2.2.2. Depression
The Beck Depression Inventory-II (Beck, Steer, & Brown, 1996) was
used to measure depressive symptoms. This measure includes
20-items (one item assessing suicidal ideation was omitted) assessed
on a 4-point scale (e.g., ranging from 1 = “I do not feel sad” to 4 = “I
am so sad and unhappy that I can't stand it”). Internal consistency for
this measure was very good (α = .89).
2.2.3. Loneliness
The UCLA Loneliness Scale (Russell, Peplau, & Ferguson, 1978) was
utilized to assess subjective feelings of loneliness. Participants indicated
how often they felt similarly to the state described by each item using a
4-point scale (1 = “I never feel this way” to 5 = “I often feel this
way”). Sample items include “I have nobody to talk to” and “I cannot
tolerate being so alone.” Internal consistency for this scale was excellent
(α = .96).
2.2.4. Social anxiety
Anxiety, specifically relating to social interactions, was measured
using the Social Interaction and Anxiety Scale (Mattick & Clarke,
1998). For each of the 20 items, participants rated the degree to which
they felt each statement was true or characteristic of them (e.g., “I be-
come tense if I have to talk about myself or my feelings”) with answer
choices ranging from 0 = “Not at all” to 4 = “Extremely.” Reliability
for this scale was very good (α = .89).
3.1. Examining the factor structure and psychometric properties of the child
social preference scale-revised
An exploratory factor analysis was conducted for the Child Social
Preference Scale-Revised items. The scree plot suggested that a
four-factor solution was most appropriate; also, only four factors had ei-
genvalues over 1.0. Principal-axis factoring with an oblique (promax)
rotation initially yielded four factors. However, two unsociability items
were dropped because they loaded highly (N.30) on more than one fac-
tor (“I don't have a strong need to be with others,” “I rarely initiate being
with others because I don't mind spending time alone”). After dropping
these items, principal-axis factoring with oblique (promax) rotation
yielded four factors, but two additional unsociability items were
dropped because of high cross-loadings (“I don't really mind spending
time alone”, “I don't have a strong preference for being alone or with
others”). A final factor analysis yielded three factors, which we refer to
as preference-for-solitude (7-items; α = .85), shyness (6-items; α =
.88), and isolation (4-items; α = .87). The shyness factor accounted
for 40.55% of the total variance, the preference-for-solitude factor
accounted for 10.49% of the total variance, and the isolation factor
accounted for 5.13% of the total variance (56.17% altogether). There
were no cross-loadings higher than .26 for the 17 items, and factor load-
ings ranged from to .38 to .91. Factor correlations based on the promax
rotation were .50 (preference-for-solitude and isolation), .54 (shyness
and preference-for-solitude), and .70 (isolation and shyness). Final fac-
tor items and loadings are presented in Table 1. Descriptive statistics
and correlations are presented in Table 2.
3.2. Results of path models testing the indirect effects of social withdrawal
subtypes on emotional eating
Mplus version 6.12 (Muthén & Muthén, 2010) was used to estimate
a path model in which shyness and preference were related to emotion-
al eating vis-à-vis three related but distinct types of internalizing diffi-
culties, namely, loneliness, social anxiety, and depression (see Fig. 1).
All independent variables and mediators were allowed to covary. Peer
isolation was included as a control variable, and was allowed to covary
with all predictors and mediators and emotional eating. In addition, BMI
was included as a control variable but was only allowed to predict emo-
tional eating and was not allowed to covary with any other independent
variable or mediator. Given that the BMI variable was skewed and
kurtotic, indirect effects were tested using the delta method and robust
Table 1
Results of principal axis factor analyses of the child social preference scale-revised.

2.2.5. Emotional eating Factor

Participants completed the Emotional Eating subscale (13-items) of Items (abbreviated) Shyness Pref. Isolation
the Dutch Eating Behavior Questionnaire (Van Strien, Frijters, Bergers, & solitude
Defares, 1986). Frequency of the desire to eat in response to negative
Desire to be with others, but nervous to .85 −.08 .04
emotions was assessed using a 5-point Likert scale (1 = “never” to Turn down hanging out because too shy .84 .00 −.15
5 = “very often”). Sample items include “Do you have a desire to eat Like to hang out but too nervous .82 −.02 −.06
when you are bored and restless?” and “Do you have a desire to eat Like to ask others to hang out, but afraid .75 −.13 .15
when you are feeling lonely?” Internal consistency was excellent Watch people but don't try to join in .55 .13 .12
Stand near others without joining in .38 .13 .26
(α = .95). Choose to spend time alone, don't like others −.01 .76 .12
Don't like others, prefer to be alone .10 .73 −.01
Like spending time alone more than others .15 .71 −.14
2.2.6. Body mass index (BMI) Try to spend time alone because don't like others −.09 .69 .21
Participants' weight and height were measured in order to calculate When given the choice prefer others (R) −.14 .63 −.15
Happiest when with others (R) −.09 .61 −.13
BMI. Participants were asked to remove excess clothing and shoes be-
Try to avoid spending time with others .15 .56 .16
fore measurements were taken. Height was measured to the nearest Sometimes others don't want me to hang out −.11 .01 .91
one-eighth inch using a wall-mounted stadiometer (Doran Scales, Inc., Wish I could spend time with others −.04 −.08 .87
DS1100 Mechanical Height Rod), and weight was measured with a dig- Others don't want to be with me .06 −.11 .85
ital scale to the nearest 0.2 lbs. Body mass index (BMI) was determined I′d like to hang out but often excluded .24 −.05 .55

by dividing an individual's weight by height ((lbs/in2) ∗ 703). Primary factor loadings are in boldface.
242 R.G. Etkin et al. / Personality and Individual Differences 97 (2016) 239–244

Table 2
Descriptive statistics and zero-order correlations among study variables.

1 2 3 4 5 6 7 8 9

1. Shyness .50⁎⁎⁎ .66⁎⁎⁎ .20⁎⁎⁎ .32⁎⁎⁎ .46⁎⁎⁎ .75⁎⁎⁎ −.03 −.04

2. Pref. solitude .41⁎⁎ .12⁎⁎ .29⁎⁎⁎ .35⁎⁎⁎ .46⁎⁎⁎ .09⁎ −.03
3. Peer isolation .15⁎⁎⁎ .31⁎⁎⁎ .53⁎⁎⁎ .50⁎⁎⁎ −.04 .03
4. Emotional eating .26⁎⁎⁎ .17⁎⁎⁎ .25⁎⁎⁎ .01 −.02
5. Depression .48⁎⁎⁎ .43⁎⁎⁎ −.02 −.05
6. Loneliness .53⁎⁎⁎ −.03 −.04
7. Social anxiety −.05 −.08
8. Age −.05
9. BMI
M 2.26 2.22 1.80 2.31 1.48 1.93 1.68 19.61 23.51
SD 0.96 0.72 0.85 0.90 0.43 0.70 0.78 1.25 4.19
⁎ p b .05.
⁎⁎ p b .01.
⁎⁎⁎ p b .001.

maximum likelihood estimators (MLR) were used. Using bootstrapping
(5000 resamples), the model yielded nearly identical results.
The model demonstrated good fit to the data, χ2 (3) = 14.07, p =
.02; RMSEA = 0.05; CFI/TLI = 0.99/0.96; SRMR = 0.02. Both shyness
and preference-for-solitude were uniquely associated with social anxi-
ety, depression, and loneliness. Paths were also significant from social
anxiety to emotional eating and from depression to emotional eating
but not between loneliness and emotional eating (see Fig. 1).
There was a significant indirect effect of shyness on emotional eating
(β = 0.15, SE = 0.05, p b .01) through social anxiety (β = 0.10, SE =
0.04, p b .05) and depression (β = 0.04, SE = 0.01, p b .01), but not lone-
liness (β = −0.01, SE = 0.02, p = .80). Similarly, there was a significant
indirect effect of preference-for-solitude on emotional eating (β = 0.06,
SE = 0.02, p b .001). However, only depression emerged as a significant
mediator of this pathway (β = 0.04, SE = 0.01, p b .01); the pathway
through social anxiety (β = 0.02, SE = 0.01, p = .06) was marginally
significant, and the pathway through loneliness was not significant
(β = − 0.00, SE = 0.01, p = 0.80). Direct effects of both shyness and
preference-for-solitude on emotional eating were not significant,
suggesting the presence of full mediation.
Finally, to assess for gender differences, two multiple group models
were examined, one in which paths were allowed to be freely estimated
across gender and another in which paths were constrained to be the
same across gender. The nested test indicated that the freely estimated
model did not significantly differ from the constrained model (Sattorra–
Bentler Scaled χ2 (15) = 20.7, p = 1.00). Thus, the model parameters
did not significantly differ by gender.
4. Discussion
The mediation results from this investigation are novel and offer
new evidence of the risks associated with shyness during emerging
adulthood as well as possible risk mechanisms (that could be targeted
in interventions). As shyness tends to be associated with poor emotion
regulation and ineffective coping tendencies (Eisenberg, Fabes, &
Murphy, 1995), shy youth may turn to eating as a way of coping with
the depression and anxiety they experience. Furthermore, given that
shy emerging adults tend to have relationships that are lower in quality
and intimacy (Nelson, 2013; Nelson et al., 2008), they may have difficul-
ty engaging in more direct and problem-focused coping strategies, such
as seeking social support, when experiencing these emotional prob-
lems. Importantly, the current findings are the first to suggest that the
psychological difficulties experienced by shy youth may, in turn, place
them at risk for problems of a different ilk (e.g., physical health-
related problems). Unexpectedly, however, the pathway between
shyness and emotional eating was not mediated by loneliness, despite
unique associations between shyness and loneliness. Although loneli-
ness is a common trigger for emotional eating (e.g., Ganley, 1989), it

Fig. 1. Path model of shyness and preference-for-solitude predicting emotional eating vis-à-vis social anxiety, depression, and loneliness. Note. Covariates were left out of the diagram for
ease of presentation; *p b .05, **p b .01, ***p b .001.
 R.G. Etkin et al. / Personality and Individual Differences 97 (2016) 239–244
may be that shy emerging adults engage in alternative behaviors to cope
with feelings of loneliness, such as distracting themselves by engaging
in solitary-constructive activities that are more social in nature
(e.g., connecting with others online).
Perhaps most noteworthy, however, were the findings pertaining to
preference-for-solitude. First, the significant and unique associations
between preference-for-solitude and social anxiety and loneliness
were novel and provide additional evidence that preference-for-
solitude may not be entirely “benign” during emerging adulthood
(Nelson, 2013). Moreover, both types of withdrawal were related to
all indices of psychological difficulties, raising question as to whether
shyness continues to carry greater risk beyond the early adolescent
years. But why might preference-for-solitude be particularly detrimen-
tal during the college years? There is some indication that having a
preference-for-solitude leads to a lack of social initiation (Bowker &
Raja, 2011); this in turn might interfere with establishing and
maintaining new relationships, particularly in an unfamiliar college
environment, and the development of psychological distress.
Interestingly, the results further indicated that preference-for-
solitude was related to emotional eating vis-à-vis depression, but not
social anxiety or loneliness. It has been suggested that youth with
preferences-for-solitude may engage in social interactions when they
feel it will be rewarding (Bowker & Raja, 2011) and thus it is possible
that they might seek support or reassurance from peers when feeling
lonely or socially anxious. However, when experiencing depressive
symptoms, such youth may lack the energy or motivation to engage in
alternate strategies or activities to alleviate their negative feelings and
instead turn to eating. There is also evidence that depressive feelings
are more likely to be associated with greater eating urges than high-
arousal emotions such as fear or anxiety, which commonly inhibit
food consumption (Macht, 2008). Given these associations, youth who
prefer to be alone may come to expect eating to alleviate their feelings
of depression only (e.g., Stice et al., 2002).
Several limitations should be noted. First, the present study is cross-
sectional. Future studies should be longitudinal to best evaluate the
mediation results found in this study as well as the direction of effects.
Second, all measures included in the present study were self-report
(with the exception of BMI), and thus reporter-bias and shared-
method variance are possible limitations. Self-report measures are
considered the “gold standard” for assessing internalizing problems,
motivations for social withdrawal, and emotional eating. However,
experimental and observational assessments of emotional eating have
been utilized (e.g., Evers et al., 2010).
The present study provides new evidence regarding the risks associ-
ated with two types of social withdrawal during emerging adulthood.
However, one study of this age range, that utilized the same measure
as we did, found evidence for three withdrawal subtypes (shyness,
unsociability, avoidance; Nelson, 2013). And, the aforementioned
Wang et al. (2013) study used a different measure that did not allow
for the distinction between avoidance and unsociability. As such, it is
clear that additional research on the measurement and factor structure
of social withdrawal during emerging adulthood is needed. Although
gender differences were not found in the path analyses, a t-test revealed
that, consistent with past research, the mean level of emotional eating
was significantly higher for females than males, t(637) = − 4.49,
p b .001; thus, researchers should consider that associations with
other health-related outcomes not considered herein may differ for
males and females.
Despite some empirical and theoretical associations between
emotional eating and BMI, a handful of recent studies (including the
present study) have found nonsignificant associations in non-clinical
samples (e.g., Nguyen-Rodriguez, Chou, Unger, & Spruijt-Metz, 2008),
suggesting that normal weight and otherwise healthy individuals may
engage in emotional eating to regulate their emotions. Initial evidence
suggests that interventions promoting adaptive coping strategies, such
as support-seeking, are effective in reducing emotional eating behavior
243
in clinical samples (Davies, Bekker, & Roosen, 2011). Thus, it is plausible,
given the present findings, that many withdrawn emerging adults may
benefit from similar interventions or prevention efforts, regardless of
their BMI. Withdrawn emerging adults may also benefit from education
about emotional eating (e.g., that emotional eating leads to increases in
negative feelings, which in turn, can lead to further food consumption).
Learning about these cycles may motivate youth to engage in different
strategies less likely to exacerbate emotional and health problems.
In conclusion, the present study takes a step toward expanding
existing models of social withdrawal by revealing that that emerging
adults with fearful and non-fearful motivations of withdrawal are at-
risk for psychological difficulties, which may place them at further risk
for engaging in unhealthy eating behaviors. Continuing to investigate
the types of risks associated with withdrawal subtypes across develop-
ment is essential to informing ways in which to intervene and prevent
further withdrawal and health problems.
References
Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual for the Beck Depression Inventory-II.
San Antonio, TX: Psychological Corporation.
Bowker, J. C., & Raja, R. (2011). Social withdrawal subtypes during early adolescence in
India. Journal of Abnormal Child Psychology, 39, 201–212. http://dx.doi.org/10.1007/
s10802-010-9461-7.
Bowker, J. C., Nelson, L. J., Markovic, A., & Luster, S. (2014). Social withdrawal during ad-
olescence and emerging adulthood. In R. J. Coplan, & J. C. Bowker (Eds.), The handbook
of solitude: Psychological perspectives on social isolation, social withdrawal, and Being
alone. Hoboken, NJ: John Wiley & Sons.
Braet, C., Claus, L., Goossens, L., Moens, E., Van Vlierberghe, L., & Soetens, B. (2008). Differ-
ences in eating style between overweight and normal-weight youngsters. Journal of
Health Psychology, 13, 733–743. http://dx.doi.org/10.1177/1359105308093850.
Bruch, H. (1973). Eating disorders. New York: Basic Books, Inc., Publishers.
Davies, M. M., Bekker, M. J., & Roosen, M. A. (2011). The role of coping and general
psychopathology in the prediction of treatment outcome in eating disorders. Eating
Disorders: The Journal of Treatment & Prevention, 19, 246–258. http://dx.doi.org/10.
1080/10640266.2011.566147.
Eisenberg, N., Fabes, R. A., & Murphy, B. C. (1995). Relations of shyness and low sociability
to regulation and emotionality. Journal of Personality and Social Psychology, 68,
505–517. http://dx.doi.org/10.1037/0022-3514.68.3.505.
Evers, C., Marijn Stok, F., & de Ridder, D. D. (2010). Feeding your feelings: Emotion
regulation strategies and emotional eating. Personality and Social Psychology Bulletin,
36, 792–804. http://dx.doi.org/10.1177/0146167210371383.
Ganley, R. M. (1989). Emotion and eating in obesity: A review of the literature.
International Journal of Eating Disorders, 8, 343–361. http://dx.doi.org/10.1002/1098-
108X(198905)8:3b343::AID-EAT2260080310N3.0.CO;2-C.
Heatherton, T. F., & Baumeister, R. F. (1991). Binge eating as escape from self-awareness.
Psychological Bulletin, 110, 86–108. http://dx.doi.org/10.1037/0033-2909.110.1.86.
Macht, M. (2008). How emotions affect eating: A five-way model. Appetite, 50, 1–11.
http://dx.doi.org/10.1016/j.appet.2007.07.002.
Mattick, R. P., & Clarke, J. C. (1998). Development and validation of measures of social
phobia scrutiny fear and social interaction anxiety. Behaviour Research and Therapy,
36, 455–470. http://dx.doi.org/10.1016/S0005-7967(97)10031-6.
Miller, J. L., Schmidt, L. A., & Vaillancourt, T. (2008). Shyness, sociability, and eating prob-
lems in a non-clinical sample of female undergraduates. Eating Behaviors, 9, 352–359.
http://dx.doi.org/10.1016/j.eatbeh.2008.01.001.
Muthén, L. K., & Muthén, B. O. (2010). 1998–2010 mplus user's guide. Muthén and Muthén.
Nelson, L. J. (2013). Going it alone: Comparing subtypes of withdrawal on indices of
adjustment and maladjustment in emerging adulthood. Social Development, 22,
522–538. http://dx.doi.org/10.1111/j.1467-9507.2012.00671.x.
Nelson, L. J., Padilla-Walker, L. M., Badger, S., Barry, C. M., Carroll, J. S., & Madsen, S. D.
(2008). Associations between shyness and internalizing behaviors, externalizing
behaviors, and relationships during emerging adulthood. Journal of Youth and
Adolescence, 37, 605–615. http://dx.doi.org/10.1007/s10964-007-9203-5.
Nguyen-Rodriguez, S. T., Chou, C., Unger, J. B., & Spruijt-Metz, D. (2008). BMI as a moder-
ator of perceived stress and emotional eating in adolescents. Eating Behaviors, 9,
238–246. http://dx.doi.org/10.1016/j.eatbeh.2007.09.001.
Page, R. M., Hammermeister, J., Scanlan, A., & Allen, O. (1996). Psychosocial and
health-related characteristics of adolescent television viewers. Child Study Journal,
26, 319–331.
Raspopow, K., Matheson, K., Abizaid, A., & Anisman, H. (2013). Unsupportive social
interactions influence emotional eating behaviors. The role of coping styles as
mediators. Appetite, 62, 143–149. http://dx.doi.org/10.1016/j.appet.2012.11.031.
Rubin, K. H., Coplan, R. J., & Bowker, J. C. (2009). Social withdrawal in childhood. Annual
Review of Psychology, 60, 141–171. http://dx.doi.org/10.1146/annurev.psych.60.
110707.163642.
Russell, D., Peplau, L. A., & Ferguson, M. L. (1978). Developing a measure of loneliness.
Journal of Personality Assessment, 42, 290–294. http://dx.doi.org/10.1207/
s15327752jpa4203_11.
Salvy, S., de la Haye, K., Bowker, J. C., & Hermans, R. J. (2012). Influence of peers and
friends on children's and adolescents' eating and activity behaviors. Physiology &
Behavior, 106, 369–378. http://dx.doi.org/10.1016/j.physbeh.2012.03.022.
244 R.G. Etkin et al. / Personality and Individual Differences 97 (2016) 239–244

Spoor, S. P., Bekker, M. J., Van Strien, T., & van Heck, G. L. (2007). Relations between neg- Wang, J. M., Rubin, K. H., Laursen, B., Booth-LaForce, C., & Rose-Krasnor, L. (2013). Prefer-
ative affect, coping, and emotional eating. Appetite, 48, 368–376. http://dx.doi.org/10. ence-for-solitude and adjustment difficulties in early and late adolescence. Journal of
1016/j.appet.2006.10.005. Clinical Child and Adolescent Psychology, 42, 834–842. http://dx.doi.org/10.1080/
Stice, E., Presnell, K., & Spangler, D. (2002). Risk factors for binge eating onset in adoles- 15374416.2013.794700.
cent girls: A 2-year prospective investigation. Health Psychology, 21, 131–138.
http://dx.doi.org/10.1037/0278-6133.21.2.131.
Van Strien, T., Frijters, J. E., Bergers, G., & Defares, P. B. (1986). The Dutch eating behavior
questionnaire (DEBQ) for assessment of restrained, emotional, and external eating
behavior. International Journal of Eating Disorders, 5, 295–315.