To Conform or to Maintain Self-Consistency? Hikikomori Risk in Japan and the Deviation From Seeking Harmony

Abstract

There is an increasing number of youth in Japan who are dropping out of society and isolating themselves in their bedrooms from years to decades at a time. These so-called hikikomori or social isolates have been described by Japanese clinicians as suffering most commonly from an autism spectrum disorder. However, such claims come out of examining biased samples of hikikomori who have already been referred to mental health professionals. Alternatively, we argue that the tendency for hikikomori represents an increasing marginalization of Japanese youth who are consciously rejecting dominant cultural values of harmony-seeking and consequently deviating in their motivation to conform to others' behaviors. The current study tests for this alternative argument for hikikomori tendencies. One hundred ninety-five participants were recruited from Japanese universities and divided into high risk vs. low risk of becoming hikikomori according to an instrument that measures such risks. As expected, high risk students shared similar social perceptions about pervasive values in their society as low risk students, thereby ruling out deficits in social perception as being associated with hikiko-mori risk. Instead, high risk students were not motivated to conform to others' behaviors relative to low risk students and this difference was fully mediated by differences in preferred levels of harmony-seeking. Furthermore, high risk students scored lower on both local identity and global identity relative to low risk students, thereby reifying their marginalized identity in Japanese society.

Full text

Journal of Social and Clinical Psychology, Vol. 33, No. 10, 2014, pp. 918-935
918
© 2014 Guilford Publications, Inc.
This study was supported by grants from the Kokoro Research Center, Kyoto
University and JSPS KAKENHI Grant Number 2109215.
The authors would like to thank Keiko Ishii and Yuri Yano for their help with data
collection.
Address correspondence to Vinai Norasakkunkit at the Department of Psychology,
Gonzaga University, 502 East Boone Avenue, Spokane, WA 99258-0102; E-mail:
vinainor@me.com.
HIKIKOMORI RISK IN JAPAN
NORASAKKUNKIT AND UCHIDA
TO CONFORM OR TO MAINTAIN
SELF-CONSISTENCY? HIKIKOMORI RISK
IN JAPAN AND THE DEVIATION FROM
SEEKING HARMONY
VINAI NORASAKKUNKIT
Gonzaga University
YUKIKO UCHIDA
Kyoto University
There is an increasing number of youth in Japan who are dropping out of society
and isolating themselves in their bedrooms from years to decades at a time. These
so-called hikikomori or social isolates have been described by Japanese clinicians
as suffering most commonly from an autism spectrum disorder. However, such
claims come out of examining biased samples of hikikomori who have already
been referred to mental health professionals. Alternatively, we argue that the ten-
dency for hikikomori represents an increasing marginalization of Japanese youth
who are consciously rejecting dominant cultural values of harmony-seeking and
consequently deviating in their motivation to conform to others’ behaviors. The
current study tests for this alternative argument for hikikomori tendencies. One
hundred ninety-ve participants were recruited from Japanese universities and
divided into high risk vs. low risk of becoming hikikomori according to an instru-
ment that measures such risks. As expected, high risk students shared similar
social perceptions about pervasive values in their society as low risk students,
thereby ruling out decits in social perception as being associated with hikiko-
mori risk. Instead, high risk students were not motivated to conform to others’
behaviors relative to low risk students and this difference was fully mediated by
differences in preferred levels of harmony-seeking. Furthermore, high risk stu-
dents scored lower on both local identity and global identity relative to low risk
students, thereby reifying their marginalized identity in Japanese society.

HIKIKOMORI RISK IN JAPAN 919
You can’t pinpoint the reason, but you can pinpoint the con-
text: it’s Japan. In Japan you have to be like other people, and if
you aren’t, you have a sense of loss, of shame. So you disappear.
—Sadatsugu Kudo, Director of Youth Support Center
which fields about 1,500 calls a year from families
seeking help about their hikikomori children.
(Time Magazine, August 28, 2000)
HIKIKOMORI AND DSM DIAGNOSES
There is an increasing number of youth in Japan who are dropping
out of society and isolating themselves in their bedrooms from years
to decades at a time. According to Japan’s Ministry of Health, La-
bor and Welfare’s first official 2003 guidelines on this culture-bound
syndrome, hikikomori (social isolation syndrome) has the following
specific diagnostic criteria: (1) no motivation to participate in school
or work; (2) no signs of schizophrenia or any other known psycho-
pathologies; and (3) persistence of social withdrawal for at least six
months. Indeed, most hikikomori individuals avoid any and all so-
cial contact, including interactions with their own family members
despite most of them shutting themselves in their own bedrooms in
their parents’ home. It is an affliction that mostly affects young Jap-
anese men (Koyama et al., 2010), although there are some women
and older people who suffer from this syndrome as well. Currently,
the lifetime prevalence rate of hikikomori is about 4.6% of Japanese
youth ages 15–34 who suffer, which is estimated to be more than
1.2 million young individuals (Norasakkunkit, Uchida, & Toivonen,
2012). The afflicted have been known to shut themselves in from six
months to even decades at a time (Zielenziger, 2006).
One obvious dilemma in studying hikikomori is that most of those
suffering from hikikomori, by definition, do not seek treatment. More
importantly, social isolation itself is not even a symptom of any of
the DSM diagnosis often assigned to an individual afflicted with
hikikomori, other than for agoraphobia. Even so, the motivation for
isolating oneself for an agoraphobic is to avoid a possible panic at-
tack in public (DSM IV–TR). The motivation for isolating oneself
among a hikikomori is simply to avoid possible social interactions
with others who might know or judge them (Zielenziger, 2006).
Indeed, many hikikomori individuals have been known to venture
outside the home in the middle of the night, often to spend time at
a convenience store because they know that it is unlikely that they

920 NORASAKKUNKIT AND UCHIDA
will encounter someone familiar at that time (Zielenziger, 2006).
The second major concern with relying only on clinical samples is
that patients have been referred to a mental health facility or clinic
because they have already been identified as suffering from a DSM
disorder. Thus, such samples may be biased.
Despite such concerns about making generalizations from clinical
samples of those suffering from hikikomori, in 2010, Japan’s Minis-
try of Health, Labor and Welfare’s updated their “guidelines” to
include individual psychopathology as a primary cause. This up-
date was largely due to the work of Kazuhiko Saito. Saito (2010)
who collected data from 125 hikikomori who sought treatment at
government run Mental Health Welfare Centers in Japan. Accord-
ing to Saito’s data, hikikomori individuals often met criteria for one
or another DSM-IV diagnosis. In his hikikomori sample, he found
that 27% suffered from an autism spectrum disorder, 22% suffered
from an anxiety disorder, 18% suffered from a personality disorder,
14% suffered from a mood disorder, 8% suffered from a psychotic
disorder, 6% suffered from adjustment disorder, and only 5% was
not otherwise specific (NOS). Thus, the most common diagnoses
among clinical samples of hikikomori were autism spectrum disor-
ders and anxiety disorders. More recently, when mothers were in-
terviewed about their hikikomori children, the most common diag-
nosis again was an autism spectrum disorder with about a third of
these patients being assigned this diagnosis (Sakai et al., 2011).
Saito’s (2010) and Sakai and colleagues’ (2011) data suggest that
10% to 15% of the hikikomori population suffer from an autism spec-
trum disorder. However, such proportions may not be replicable.
For example, Norasakkunkit and Uchida (2010) recruited a sample
of hikikomori youth through a Youth Support Center whose parents’
cooperation were solicited to pass on a survey packet that included
various mental health measures to their withdrawn adult children.
This sample was compared with an elite university student sample
and no differences were found in the levels of autism spectrum quo-
tient or anxiety tendencies (i.e., social avoidance and distress, fear
of negative evaluation, taijinkyofusho—a type of social anxiety most-
ly bound to Japanese culture, and post-traumatic stress). Having
said that, somewhat elevated levels of depression and borderline
personality tendencies, as well as somewhat lower levels of global
functioning, were present among the hikikomori sample relative to
the university student sample, although these levels were not clini-
cally significant.

HIKIKOMORI RISK IN JAPAN 921
Moreover, in the first epidemiological study conducted on hikiko-
mori that was as close to a nation-wide random sample as possible,
Koyama and colleagues (2010) conducted a face-to-face household
survey, including a structured diagnostic interview, by randomly
picking households and interviewing 4,134 individuals. They con-
firmed a hikikomori lifetime prevalence rate of 1.2% in their nation-
wide sample. Among these hikikomori individuals, the researchers
found that only half suffered from a DSM-IV diagnosis. However,
and more importantly, there was no particular diagnosis that was
systematically associated with hikikomori. For the remaining half of
hikikomori individuals, no DSM diagnosis could even be assigned.
Thus, the researchers concluded that any DSM diagnosis was an
epiphenomenon to hikikomori at best and that hikikomori is rather a
“psychopathology characterized by impaired motivation” (p. 72).
HIKIKOMORI AS A SOCIAL PATHOLOGY
Recently, we (Norasakkunkit & Uchida, 2011; Norasakkunkit &
Uchida, 2012; Norasakkunkit, Uchida, & Toivonen, 2012; Toivonen,
Norasakkunkit, & Uchida, 2011) have provided some evidence
and theoretical grounds for the argument that hikikomori is part of
a larger set of youth problems that stem from a misalignment be-
tween mainstream Japanese institutional practices and postindus-
trial pressures to make structural and ideological changes so that
Japan can become more competitive in the global marketplace. Put
more simply, hikikomori and other recent social problems in Japan
(i.e., underemployment of youth, decreasing population size, wom-
en’s increasing reluctance to get married and/or have children, etc.)
may be arising from the fact that Japanese cultural practices are not
keeping up with rapid societal changes due to globalization and
postindustrialization processes. In particular, institutional resis-
tance to postindustrial pressures include protecting long-standing
institutional cultural practices such as adhering to a seniority sys-
tem or engaging in corporate hiring practices that tightly control
and curb entry into the long-term labor market to protect the job
security of senior elites. Maintaining such protective cultural prac-
tices which are clashing with the globalization pressures to be more
competitive, horizontal, and flexible in the labor market comes at
an increasing social cost born mostly by young Japanese who are
systematically prevented from attaining secure and stable social

922 NORASAKKUNKIT AND UCHIDA
spaces in the Japanese labor market (see also Brinton, 2011). Since
many youth are prevented from participating in mainstream Japa-
nese society while very little, if any, legitimately recognized alter-
native life course to success is available to them, a sense of anomie
ensues. Consequently, many Japanese youth may not see the benefit
of internalizing a motivational style that is oriented toward con-
formity for the sake of harmony. Nor have they become globalized
by adopting the values of another culture (e.g., Western culture).
Hence, they may be rejecting mainstream cultural values without
switching to other values in a coherent way. Thus any motivation
to act in culturally incongruent ways may be an outcome of their
actual or anticipated marginalization rather than a cause of it.
After having ruled out any systematic link between hikikomori
and any DSM diagnosis, Koyama and colleagues’ (2010) boldly con-
cluded that hikikomori is a psychopathology of impaired motivation.
More recently, Norasakkunkit and colleagues (2012) have argued
that hikikomori is more of a social pathology than a psychopathology
since the cause of the pathology lies more in the structural aspects
of society, which limit the institutional means to accomplishing
cultural goals for many Japanese youth (see Toivonent, Norasak-
kunkit, & Uchida, 2011). Since impaired motivation implies a deficit
in capacity as opposed to a conscious choice, Norasakkunkit and
colleagues have argued that the motivation among hikikomori is not
necessarily impaired but rather deviant from the Japanese cultural
mandate toward conformity for the sake of harmony. Yet, because
the social reality in Japan limits multiple paths to cultural success,
at least relative to other postindustrialized societies, such devi-
ant values and motivational styles may actually be dysfunctional
and distressing for those with such tendencies, even to the point of
wanting to avoid all interactions with the outside world.
RATIONALE AND RESEARCH HYPOTHESES OF THE CURRENT
STUDY
Although there is some empirical basis for the above social pathol-
ogy argument (see Norasakkunkit & Uchida, 2011), the role of being
motivated to dissociate from conformity and harmony expectations
for those with hikikomori tendencies has not been more systematical-
ly tested. The current study attempted to do just that. Specifically,
the current study attempted to investigate the motivation toward
conforming to others for social harmony reasons among university

HIKIKOMORI RISK IN JAPAN 923
students who are at risk of becoming hikikomori and students at low
risk of becoming hikikomori, according to a marginalization risk fac-
tor scale (Uchida & Norasakkunkit, 2011) that was developed for a
previous study.
With respect to measuring preferred values and social cognitive
deficits associated with autism spectrum disorders, the current
study employed a harmony seeking values scale standardized on a
Japanese sample. Each participant was asked to evaluate each item
on the harmony seeking values scale at three levels: (1) self, (2) ideal
self, and (3) other people in their society in general (i.e., perceived
consensus). Preferred values were operationalized as harmony
seeking values at the ideal self level. Also, perceiving and under-
standing common beliefs requires a well intact theory of mind (To-
massello, 1999), which is not likely if one is afflicted with an au-
tism spectrum disorder because such disorders are associated with
at least a certain degree of theory of mind deficits (Tager-Flusberg,
2007). Thus, anybody with an autism spectrum disorder should not
only have deficits in the capacity to perceive socially shared percep-
tions but would also have deficits in the capacity to be concerned
about such perceptions. Therefore, any differences between high
risk students and low risk students in levels of perceived consen-
sus of harmony-seeking values may possibly indicate deficits in
perceiving pervasive values in society while no such differences
should indicate that the role of such deficits being associated with
hikikomori risk is unlikely.
With respect to operationalizing the motivation to conform for
harmony seeking reasons, Cialdini, Wosinka, Barrett, Butner, and
Gornik-Durose (1999) found that the motivation to comply with
a request to complete a survey was greater for individuals in an
interdependent-oriented society when they believed that most of
their classmates have complied with similar requests (social proof)
than when they believed that they had generally complied to simi-
lar requests in the past (commitment consistency), while this pat-
tern was reversed for people in an independent-oriented society.
Cialdini and colleagues’ study was replicated in the current study
to compare high risk students with low risk students on their mo-
tivations to conform. It was expected that the compliance of low
risk students will be motivated by social proof (i.e., social conformi-
ty), while compliance for high risk students will deviate from such
motivations and may even be oriented more toward commitment
consistency (i.e., nonconformity) relative to low risk students. Fur-

924 NORASAKKUNKIT AND UCHIDA
thermore, the difference between high risk and low risk students
in motivation to conform under conditions of social proof was ex-
pected to be mediated by levels of harmony seeking values at the
ideal self level.
Finally, according to the acculturation and adaptation model by
Berry and Sam (1997), marginalized individuals should be disiden-
tifying with their own heritage culture as well as that of their host
culture. Since the social pathology argument suggests that hikiko-
mori individuals are marginalized in part because they are rejecting
mainstream cultural values without necessarily becoming global-
ized by adopting the values of another culture (e.g., Western cul-
ture), we have chosen to adapt the Barry and Sam model by mea-
suring local identity and global identify. It was expected that high
risk students will score lower on both local identity and global iden-
tity relative to low risk students, thereby reifying the marginalized
identities of high risk students (see Figure 1).
METHOD
PARTICIPANTS
Since there are relatively few university students who score high
on the marginalization risk factor scale, participants were recruited
in three phases in order to maximize sample size. Unfortunately,
not all the measures could be included in phase two and phase
three since some of the measures relevant to this study were simply
tagged onto other studies during those phases. The specific mea-
sures that were excluded in each phase, if any, are mentioned in
the Materials section below. In phase one, 119 participants were re-
cruited from Kyoto University (63 males, 53 females, 3 unspecified).
In phase two, another 22 participants were recruited from Kyoto
University again (11 males, 11 females). In phase 3, 54 participants
were recruited from Kobe University (33 males, 21 females). The age
ranged from 18 to 28 with an average age of 20.24 (SD = 1.56). All
participants were asked to complete a survey packet and received a
$10 gift card for participating in this study.
MEASURES
Hikikomori Risk Factor Scale (HRFS). The HRFS (Uchida & Nora-
sakkunkit, under review) was developed by compiling common

HIKIKOMORI RISK IN JAPAN 925
attitudes and values held by marginalized Japanese youth, in-
cluding hikikomori, according to other investigators on the subject.
These items used a 7-point Likert scale. Initial analyses identified 27
meaningful survey items that measured the risk of becoming mar-
ginalized in modern Japan. Furthermore, three distinct risk factors
emerged in the initial exploratory factor analysis. Factor 1 was la-
beled Freeter Lifestyle Preference, which refers to the tendency of
becoming a Freeter who consciously chooses not to work despite
opportunities and job availabilities. An example of an item from
this factor is, “After graduation, if there is something at work that I
cannot tolerate, it is better to not force myself and instead just quit.”
The next factor that emerged, Factor 2, has to do with a Lack of Self
Competence (i.e., not feeling competent in accomplishing interde-
pendent cultural tasks). An example of an item from this factor is,
“My social skills are low, and I am not good at relating to others.”
Factor 3 has to do with an “Unclear Ambition for the Future” (i.e.,
having unclear or unrealistic goals for what they want to do in the
future). An example of an item from this factor is, “I don’t quite
know what I want to do in the future.”
Although these factors are relatively distinct from each other, a
reliability analysis also revealed that an overall score, combining all
the items, could be calculated to represent overall risk of becoming
FIGURE 1. Adapting Barry and Sam’s (2007) acculturation and
adaptation model to a globalization and adaptation model.

926 NORASAKKUNKIT AND UCHIDA
marginalized. In the current data, the reliability for the items that
make up the overall risk score was alpha = .81.
Norasakkunkit & Uchida (2010) found that university students
consistently score lower on all three factors as well as on the over-
all score relative to a nationwide sample of occupationally inactive
youth or the so-called NEETs (Not in Employment, Education, or
Tra in ing) an d a sm all s am ple of thos e sufferi ng from hikikomori. Fur-
thermore, Norasakkunkit, Toivonen, and Uchida (2011) found that in
January of 2011, in a sample of fourth-year Kyoto University students
(n = 21), about 80% of students were searching for or had already se-
cured a long-term employment position three months prior to their
graduation. However, these students were considered to be at low
risk of potentially being marginalized as NEETs or hikikomori accord-
ing to HRFS. In contrast, in a sample of fourth year Kyoto University
students who were at high risk of being marginalized according to
HRFS (n = 5), the rate of searching for a job or securing a long-term
employment position three months prior to graduation was 0%.
This measure was included in all three phases of data collection.
Willingness to Comply (WTC). The WTC (Cialdini et al., 1999) is a
marketing survey that comes in two versions. Participants are asked
how much they were willing to comply to completing a marketing
survey on a scale from 0 (very low likelihood) to 8 (very high likeli-
hood). One version of the survey asked the participants to rate their
likelihood of complying to a request assuming that the majority of
their classmates have already complied (i.e., conformity or social
proof version). The other version of this survey asked participants to
rate their likelihood of complying to the request assuming that they
have complied to a similar survey in the past (commitment consis-
tency version). This measure was included in all three phases of data
collection.
Local-Global Identity Scale. The Local-Global Identity Scale (Zhang
& Khare, 2009) is a 19-item scale to measure local-global identity.
Each item is on a 7-point scale (1 = strongly disagree; 7 = strongly
agree). Nine items constitute the global identity scale (alpha = .74)
and 10 items constitute the local identity scale (alpha = .74). The
mean score for each scale was computed separately. This measure
was included in phase 1 data collection only.
Measure of Harmony-Seeking at the Ideal Self, Self, and Perceived Con-
sensus Levels (Hashimoto & Yamagishi, 2013). The harmony-seeking
measure is a subscale of a Japanese collectivism self-report instru-

HIKIKOMORI RISK IN JAPAN 927
ment which measures collectivistic values at three levels: (1) ideal
self level, (2) self level, and (3) common beliefs level. With this in-
strument, Hashimoto and his colleagues (2011) showed a pattern
whereby most Japanese endorse highest levels of collectivism at
the perceived consensus level and the lowest level of collectivism
at the ideal self level. Each item was evaluated on a 7-point scale (1
= Doesn’t describe at all; 7 = Describes very much). The Cronbach’s
Alpha reliability analyses for the three levels for which this value
was measured were: (1) self level: alpha = .76; (2) ideal self level:
alpha = .80; and (3) perceived consensus level: alpha = .74.
This measure was included in phase 1 and phase 2 only. Further-
more, harmony seeking at the perceived consensus level was not
measured in phase 2.
PROCEDURE
Participants were divided into a high-risk group and a low- (to mod-
erate) risk group according to a cut-off score on the HRFS (Uchida &
Norasakkunkit, 2011). Then, in an experimental design, half the par-
ticipants were given a survey packet in which the WTC contained
instructions asking participants to rate their willingness to comply
to a request under the social proof (i.e., conformity) condition while
the other half was given a version of the packet in which the WTC
contained instructions asking participants to rate their willingness
to comply to a request under the condition of commitment consis-
tency. The rest of the measures in the survey packet were identical
across the two compliance conditions.
RESULTS
To test the hypothesis that differences in the level of preferred
harmony-seeking values, rather than deficits in social perception,
are associated with risk, preferred values were operationalized as
harmony-seeking values at the ideal self level. Also any differences
between high-risk students and low-risk students in levels of per-
ceived consensus of harmony-seeking values may possibly indicate
deficits associated with social perception, which can index an au-
tism spectrum disorder. Relevant correlations between all measures
are indicated in Table 1. Relevant descriptive and independent sam-
ples t-test statistics are indicated in Table 2.
As expected, independent samples t-tests revealed that there was
no difference between low-risk and high-risk students in the per-

928 NORASAKKUNKIT AND UCHIDA
ceived consensus of harmony seeking. This finding suggests that
high-risk students have the samte capacity for commonly shared
social perceptions as low-risk students. A 2 × 3 risk (high-risk, low-
risk) by harmony-seeking condition (perceived consensus, self, and
ideal self) mixed design ANOVA with Risk as a between subjects
factor and Harmony Seeking as a within subjects factor revealed the
expected interaction effect between the two factors, as indicated by
F(2, 234) = 6.31, p = .002. Specifically, independent samples t-tests re-
vealed that low risk students scored higher on harmony seeking val-
ues relative to high risk students at both the self level and the ideal
self level with the ideal self level yielding the biggest difference.
A repeated measures one-way ANOVA of harmony seeking con-
dition for low risk students further revealed a main effect of har-
mony seeking, as indicated by F(2, 180) = 16.14, p < .001. Specifical-
ly, while low risk students perceive harmony seeking values at the
self level as being lower than that at the perceived consensus level
(Mean Difference = .55, p < .001), there was no difference in percep-
tion of harmony seeking at the ideal self level and the perceived
consensus level. In other words, low risk students would like to
ideally be at the level of harmony seeking as what they perceive to
be the case among the majority of Japanese people. In contrast, a
repeated measures one-way ANOVA of harmony seeking for high
risk also revealed a main effect of harmony seeking, as indicated
by F(2, 54) = 14.85, p < .001, where all three levels of harmony seek-
ing were different from each other. Specifically, high risk students
would ideally like to be lower on harmony seeking than what they
perceive to be the case in Japanese society (Mean Difference = .81, p
= .006) although this ideal is slightly higher than the level at which
they perceive themselves to have (Mean Difference = -.39, p < .05).
Together, these findings suggest that low risk students and high risk
students most differed on their preferred level of harmony seeking
TABLE 1. Correlations Among All Measured Variables
123456
1. HRFS - –.27** –.33*** .15 –.28** –.16
2. Harmony seeking – self level - .65*** –.12 .37*** .21*
3. Harmony seeking – ideal self level - .03 –.28** .04
4. Harmony seeking – perceived consensus
level
- –.007 .19*
5. Mean local identity - .37***
6. Mean global identity -
Note. *p < .05; **p < .01; ***p < .01

HIKIKOMORI RISK IN JAPAN 929
(i.e., harmony seeking at the ideal self level) with high risk students
being significantly lower than that of low risk students.
To test the hypothesis that high–risk students have a motivational
style that deviates from conformity while low–risk students will
be oriented toward conformity, the responses to WTC scale were
submitted to a 2 (risk) by 2 (compliance condition) ANOVA. The
expected interaction was found, as indicated by F (1,191) = 5.84,
p = .02 (see Figure 2). However, comparing willingness to comply
between low-risk and high-risk students in each compliance condi-
tion suggests that high-risk students were not more motivated to
comply by commitment consistency than were low-risk students.
As was expected, the difference between a low-risk and high-risk
student lied in willingness to comply under the social proof condi-
tion, as indicated by t(106) = 2.14, p < .04. In a moderated regression
analysis, the interaction term between HRFS scores and compliance
conditions was also significant, as indicated by β = .06, p < .05, al-
though simple slope analyses did not reveal significant findings.
Overall, these findings suggests that, relative to low-risk students,
the motivation style of high-risk students was more about deviating
from the culturally dominant motivation process (i.e., conforming
to others’ behaviors) and less about being motivated by something
else, like commitment consistency.
Furthermore, as expected, the difference between low-risk and
high-risk students in willingness to comply under conditions of
social proof was fully mediated by harmony seeking at the ideal
TABLE 2. Mean Differences Between High-Risk University Students and Low-Risk
University Students (Self, Ideal Self, and Perceived Consensus Refer to Harmony-Seeking
Values at Different Levels)
Measures Sample
(Risk)
n M SD t df p
Self Level Low 105 4.76 .82 2.96 139
High 36 4.27 .93 <.01
Ideal Self Level Low 105 5.25 .87 3.94 139
High 36 4.53 1.14 <.001
Perceived Consensus Level Low 91 5.29 .72 ns
High 28 5.38 .60
WTC Low 143 3.80 3.35 ns
High 52 3.69 3.13
Local Identity Low 91 3.97 .73 3.83 .73 <.001
High 28 3.39 .64 .64
Global Identity Low 91 3.81 .84 1.97 117 =.05
High 28 3.46 .81

930 NORASAKKUNKIT AND UCHIDA
self level (see Figure 3 for relevant statistics). Thus, the difference
in motivation to conform to others’ behaviors between high risk
and low risk students can be accounted for by the difference in how
much low-risk students and high-risk students want to ideally be
oriented toward harmony seeking. Importantly, high-risk students
do not want to be oriented toward harmony seeking as much as the
low-risk students do.
Finally, as expected, high-risk students scored lower than low-
risk students on both local identity and global identity (see Table 2
for relevant means). However, once HRFS was regressed on global
identity and local identity, only local identity and not global iden-
tity significantly and inversely predicted HRFS scores, as indicated
by β = -.26, p = .007. This finding suggests that high-risk students
have disidentified with both their heritage culture and global cul-
ture relative to the low-risk students, but the crucial predictor of
hikikomori risk among these predictors is local identity. Applying
Berry and Sam’s (1997) acculturation model to this finding suggests
that hikikomori risk can reasonably be associated with having a mar-
ginalized identity.
DISCUSSION
In the current study, all the hypotheses were supported, thereby
suggesting that the atypical motivation style associated with hikiko-
mori tendencies seem to be guided by culturally-deviant preferred
values and a marginalized identity. Specifically, high-risk students
FIGURE 2. Risk by Compliance (Commitment Consistency) on Mean
Willingness to Comply.

HIKIKOMORI RISK IN JAPAN 931
had a motivational style that deviated from conformity which can
be explained by the desire to deviate from a harmony-seeking ori-
entation. This pattern is consistent with previous findings that high-
risk students also deviated from mainstream Japanese tendencies in
terms of having more of an entity view of self rather than an incre-
mental view of self and therefore being more likely to give up af-
ter a failure relative to low-risk students (Norasakkunkit & Uchida,
2011). Also, high-risk students, while sharing the same perception
about the pervasive social orientation in Japanese society, also devi-
ated from wanting to personally align themselves with that social
orientation compared to low-risk students.
The fact that hikikomori is itself not a symptom of any particular
DSM diagnosis and that it was relatively rare in the past with rising
incidence over time (Zielenziger, 2006) suggests that it is probably
largely rooted in the social-structural forces of society due to glo-
balization, postindustrialization, and institutional reactions to such
pressures that are alienating many Japanese youth who may con-
sequently be choosing not to conform to the standards and norms
of their society (Brinton, 2011; Genda, 2005; Kosugi, 2004; Norasak-
kunkit et al., 2012; Saito, 1998; Kawanishi, 2009; Zielenziger, 2006).
The motivational style and preferred values of high-risk students
suggest that hikikomori is a kind of rebellion against conformity and
against other psychological processes that are well suited for func-
tioning smoothly in Japanese society, such as having a sense of local
identity, and idealizing values associated with social harmony and
so on. If indeed many of these young adults feel much pressure to
conform in a society that can no longer assure them of a secure so-
cial space to belong to, then perhaps the state of anomie that ensues
would naturally compel them to retreat and withdraw, even if there
is no reason to doubt the soundness of their personal mental health.
FIGURE 3. Mediation Model: Harmony Seeking at the Ideal Self Level
Mediates the Relationship Between Risk and Compliance.

932 NORASAKKUNKIT AND UCHIDA
It is also worth mentioning that the fact that hikikomori is consid-
ered a culture-bound syndrome does not preclude cases of hikikomo-
ri being documented in other cultures, just as other culture-bound
syndromes in Japan like taijinkyofosho can be found in the United
States (Norasakkunkit, Kitayama, & Uchida, 2012) or cases of an-
orexia which do not involve body image concerns previously com-
mon in Hong Kong have also been documented in Western cultures
(Watters, 2010). Indeed, cases of hikikomori have been documented in
South Korea, Oman, and Spain (Kato et al., 2012). Therefore, hikiko-
mori is not a culture-bound syndrome in the most literal sense. Yet,
it can be reasonably argued that what makes a syndrome culture-
bound is not its complete absence in other cultures but its relative
prevalence as well as the prevalence of the conditions that give rise
to and sustain the syndrome in a culture.
For example, maintaining a seniority system, as well as maintain-
ing a relatively impermeable long-term labor market due to rigid
hiring practices in the midst of postindustrial pressures to do pre-
cisely the opposite, may be preventing many young Japanese from
joining mainstream society because they have to bear the brunt of
the social cost of these globally uncompetitive institutional prac-
tices (Toivonen, Norasakkunkit, & Uchida, 2011).
Another condition associated with hikikomori may be related to
how Japanese parents seem to be much more willing than Ameri-
can parents to financially and logistically fully support their adult
children well in to middle age without necessarily seeking external
interventions (Kawanishi, 2009). Furthermore, public aid in Japan
is only made available if self-help and family support are not suf-
ficient (Vij, 2007).
These more or less culture-specific conditions may represent
some of the conditions that give rise to and sustain hikikomori as
a syndrome or a more pervasive social problem. These conditions
speak to why specific manifestations of a larger issue become cul-
ture-bound. Having said that, the larger issue may be less culture-
bound. For example, hikikomori may simply be one manifestation of
how many young people are finding it difficult to cope with modern
society and with increased competitiveness due to globalization. In
the United Kingdom or the United States, the same issue may mani-
fest in terms of increased homeless youth (Saito, 2013), for example.
Therefore, to say that hikikomori primarily stems from a neurological
or psychological disorder would be like saying that homelessness
is primarily caused by mental disabilities. While there are certainly

HIKIKOMORI RISK IN JAPAN 933
cases in which mental health issues trigger hikikomori or homeless-
ness, the vast majority of cases may be stemming from structural
forces in society, such as unequal opportunities to access stable jobs,
that are marginalizing such individuals or demotivating them from
embracing mainstream cultural values and goals.
Even though a mental disability may not be the cause of hikiko-
mori, there are few people, if any, who would dispute the fact that
hikikomori is dysfunctional and probably distressing as well. Indeed,
mild depression is often found among those suffering from hikiko-
mori (Koyama et al., 2010; Norasakkunkit & Uchida, 2010) but it
is very plausible that such distress and dysfunction are the conse-
quences of leading a hikikomori lifestyle rather than a cause of hikiko-
mori itself. In fact, Kato and his colleagues (2011) have found that
Japanese young people who are trying to cope with modern society
are increasingly suffering from what they call modern-type depres-
sion, which is characterized by a shift in values form collectivism
to individualism, distress and reluctance to accept prevailing social
norms, a vague sense of omnipotence, and avoidance of effort and
strenuous work. The fact that these tendencies are associated with
depression suggest that perhaps deviating from prevailing cultural
norms and values can be distressing in itself, even if those engaging
in such behaviors personally “prefer” to deviate from such norms
and values.
One limitation of this study was the reliance on an analog sample
rather than on actual people suffering from hikikomori. In the cur-
rent study, however, there were logistic issues with recruiting real
people suffering from hikikomori to participate in a social psycho-
logical study. Future research should try to obtain data on harmo-
ny-seeking tendencies at the self level, the ideal self level, and the
perceived consensus level from a nonclinical sample of those who
are suffering from real hikikomori. The second limitation of the study
was that not every relevant measure was included in each phase of
data collection. This may have limited the power of the findings
from analyses which included measures from phase one data col-
lection only. Therefore, any analysis that include harmony seeking
at the perceived self level, local identity, or global identity was an
analysis performed with the participants from phase one data col-
lection only.
In conclusion, the current study provided support for the idea
that hikikomori risk is associated with a deviation or possibly an out-
right rejection of mainstream cultural values of harmony seeking

934 NORASAKKUNKIT AND UCHIDA
in Japan, which then explains the motivation to deviate from social
conformity as a behavioral tendency. At the same time, the evidence
in the current study also suggested that hikikomori risk was not as-
sociated with deficits in commonly shared social perception. Thus,
much of the weight of the evidence in the current investigation
suggests that hikikomori risk should be conceptualized as a social
pathology that is driving many young individuals to consciously
become cultural deviants rather than as an innate deficit in the ca-
pacity to perceive socially-shared perceptions and therefore a lack
of capacity to be concerned about such perceptions.
REFERENCES
Berry, J. W., & Sam, D. (1997). Acculturation and adaptation. In J. W. Berry, M. H.
Segall, & C. Kagitcibasi (Eds.), Handbook of cross-cultural psychology (Vol. 3, pp.
291–326). Boston: Allyn & Bacon.
Brinton, M. (2011). Lost in transition: Youth, work, and instability in postindustrial Japan.
New York: Cambridge University Press.
Cialdini, R. B., Wosinska, W., Barrett, D. W., Butner, J., & Gornik-Durose, M. (1999).
Compliance with a request in two cultures: The differential influence of social
proof and commitment/consistency on collectivists and individualists. Per-
sonality and Social Psychology Bulletin, 25, 1242–1253.
Genda, Y. (2005). A nagging sense of job insecurity: the new reality facing Japanese youth.
Tokyo: International House of Japan, Inc.
Hashimoto, H., Ohashi, K., & Yamagishi, T. (2011, January). Common beliefs, behaving
by anticipating other’s response, and self-sustaining mechanism of Japanese interde-
pendence. Presented at the 12th annual meeting of the Society for Personality
and Social Psychology, San Antonio, Texas, USA.
Hashimoto, H., & Yamagishi, T. (2013). Two faces of interdependence: Harmony
seeking and rejection avoidance. Asian Journal of Social Psychology, 16, 142–151.
Kato, T. A., Shinfuku, N., Fujisawa, D., Tateno, M., Ishida, T., Akiyama, T. et al. (2011).
Introducing the concept of modern depression in Japan; An international case
vignette survey. Journal of Affective Disorders, 135, 66–76.
Kato, T. A., Tateno, M., Shinfuku, N., Fujisawa, D., Teo, A. R., Sartorius, N. et al.
(2012). Does the ‘hikikomori’ syndrome of social withdrawal exist outside Ja-
pan? A preliminary international investigation. Social Psychiatry and Psychiatric
Epidemiology, 47, 1061–1075.
Kawanishi, Y. (2009). Mental health challenges facing contemporary Japanese society: The
“Lonely People.” Kent, UK: Global Oriental Ltd.
Kosugi, R. (2008). Escape from work: Freelancing youth and challenge to corporate Japan.
Melbourne: Trans Pacific Press.
Koyama, A., Miyake, Y., Kawakami, N., Tsuchiya, M., Tachimori, H., & Takeshima, T.
(2010). Lifetime prevalence, psychiatric comorbidity and demographic corre-
lates of “hikikomori” in a community population in Japan. Psychiatry Research,
176, 69–74.

HIKIKOMORI RISK IN JAPAN 935
Norasakkunkit, V., Kitayama, S., & Uchida, Y. (2012). Social anxiety and holistic cog-
nition: Self-focused social anxiety in the U.S. and other-focused social anxiety
in Japan. Journal of Cross-Cultural Psychology, 43(5), 742-757.
Norasakkunkit, V, Toivonen, T., & Uchida, Y. (2011, October 28). Emerging adulthood
and exploration in Japan: Better strategically late than dangerously early? Fifth Con-
ference on Emerging Adulthood, Providence, RI, USA.
Norasakkunkit, V., & Uchida, Y. (June 29, 2010). A social/cultural psychological perspec-
tive on NEET/hikikomori: Implications for self, motivation, rehabilitation, and differ-
entiation. Kokoro Research Center, Kyoto University, Kyoto, Japan.
Norasakkunkit, V., & Uchida, Y. (2011). Psychological consequences of post-indus-
trial anomie on self and motivation among Japanese youth. Journal of Social
Issues, 67, 774–786.
Norasakkunkit, V. & Uchida, Y., (2012). Marginalized Japanese youth in post-indus-
trial Japan: Motivational patterns, self-perceptions, and the structural founda-
tions of shifting values. In G. Trommsdorff & X. Chen (Eds.), Values, religion,
and culture in adolescent development (pp. 211-234). Cambridge: Cambridge Uni-
versity Press.
Norasakkunkit, V., Uchida, Y., & Toivonen, T. (2012). Caught between culture, soci-
ety, and globalization: Youth marginalization in post-industrial Japan. Social
and Personality Psychology Compass, 6(5), 361–378.
Saito, K. (2010) Guideline of assessment and support for Hikikomori. Ministry
of Helath, Labour, and Welfare. http://www.ncgmkohnodai.go.jp/pdf/
jidouseishin/22ncgm_hikikomori.pdf
Saito, T. (1998). Shakaiteki Hikikomori: Owaranai Shishunki (Social withdrawal: Unending
adolescence). Tokyo: PHP Shuppan.
Saito, T. (2013). Hikikomori: Adolescence without end. Minneapolis: University of Min-
nesota Press.
Sakai, M., Horikawa, H., Nonaka, S., Matsumoto, M., Hirakawa, S., & NPO KHJ Par-
ents Association (2011). Survey report on Hikikomori Vol.8. “Hikikomori and
developmental issues”. Tokushima Univ.
Tager-Flusberg, H. (2007). Evaluating the theory of mind hypothesis of autism. Cur-
rent Directions in Psychological Science, 16, 316–320.
Toivonen, T., Norasakkunkit, V., & Uchida, Y. (2011). Unable to conform, unwilling to
rebel? Youth, culture and motivation in globalizing Japan. Frontiers in Cultural
Psychology, 2, 1–9.
Tomassello, M. (1999). Cultural origins of human cognition. Cambridge, MA: Harvard
University Press.
Uchida, Y., & Norasakkunkit, V. (under review). NEET and Hikikomori spectrum: A
socio-cultural marginalization.
Vij, R. (2007) Japanese modernity and welfare. Hampshire: Palgrave Macmillan.
Watters, E. (2010). Crazy like us: The globalization of the American psyche. New York:
Free Press.
Zhang, Y., & Khare, A. (2009). The impact of accessible identities on the evaluation of
global versus local products. Journal of Consumer Research, 36, 524–537.
Zielenziger, M. (2006). Shutting out the sun: How Japan created its own lost generation.
New York: Nan A. Talese.

Copyright of Journal of Social & Clinical Psychology is the property of Guilford Publications
Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv
without the copyright holder's express written permission. However, users may print,
download, or email articles for individual use.