Making sense of it all: The impact of sensory processing sensitivity on
daily functioning of children
Soﬁe Boterberg ⁎,PetraWarreyn
Ghent University, Ghent, Belgium
Received 8 July 2015
Received in revised form 11 December 2015
Accepted 14 December 2015
Available online xxxx
Previous research on sensory processing sensitivity and related concepts showed an association with internaliz-
ing problems. The current explorative study investigated the underlying factor structure of the parent-report
Highly Sensitive Person Scale (HSPS) and its association with problems in daily functioning. Caregivers of 235
children (3–16 years) completed the HSPS as well as questions on daily functioning. First, the factor structure
of the HSPS was explored and evaluated. Second, both differences in reported problems between a high SPS
and a control group, and in SPS factors between children with few versus many problems, were examined. Re-
sults suggested that the scores of the HSPS have good internal consistency and supported a two-factor structure
which distinguishes Overreaction to Stimuli (OS) and Depth ofProcessing (DP). Children with high SPS were re-
ported to have more internalizing problems. High OS was more common in children who cried excessively as a
baby, children with medically unexplained physical symptoms (MUPS), sleeping, eating and drinking problems
while highDP was more common in children with MUPS and sleepingproblems. This study providesthe ﬁrst em-
pirical evidence that the parent-report HSPS may add valuable information to the assessment of children with
problems in daily functioning.
© 2015 Elsevier Ltd. All rights reserved.
Sensory processing sensitivity
Highly sensitive person scale
Aron and Aron (1997) described Sensory Processing Sensitivity
(SPS) as a genetically determined temperamental or personality trait
which is present in some individuals and reﬂects an increased sensitiv-
ity of the central nervous system and a deeper cognitive processing of
physical, social and emotional stimuli (Aron, Aron, & Jagiellowicz,
2012). The terms “hypersensitivity”or “highly sensitive”,whichare
popular synonyms for the scientiﬁc concept of SPS, are increasingly
used in psychological practice both with adults and with children. How-
ever, despite the rising popularity of the concept in general society and
previous research on different genes, patterns of brain activation,
behaviors, and physiological reactions associated with high SPS (see
Aron et al., 2012 for an overview), there is still a lack of fundamental,
empirical and independent scientiﬁc evidence for the temperamental
concept of SPS. The present study has to be considered as exploratory
since it is, to our knowledge, the ﬁrst which examines SPS in children.
Aron and Aron (1997) suggested that the trait would be present
in 15 to 20% of the population. Individuals with high SPS are believed
to be easily overstimulated by external stimuli because they have a
lower perceptual threshold and process stimuli cognitively deeper
than most other people. In addition, they would respond more to cues
in the environment by comparing them to previous experiences with
similar cues. This may result in taking more time to observe and react
slower whereby they seem less prone to act when confronted with a
new situation and have more aversion towards risk-taking (Aron
et al., 2012).Further, researchin evolutionary biology provides evidence
that the trait of SPS can be observed in over 100 nonhuman species in
the form of sensitivity, responsiveness, plasticity and ﬂexibility (Wolf,
van Doorn, & Weissing, 2008).
Aron et al. (2012) state that both introversion (the inhibition of
social behaviors) and neuroticism (the reporting of intense negative
emotion) could theoretically, in some cases, be aspects of a general
sensitivity. Both Aron and Aron (1997) and Smolewska, McCabe, and
Woody (2006) undertook systematic statistical comparisons of the sen-
sitivity measure and several measures of traditional personality traits of
introversion and neuroticism to examine similarities and differences
between SPS, introversion and neuroticism. Their ﬁndings indicated
that SPS is a unique personality trait which deserves to be examined
separately. This is an important ﬁnding, since the trait of sensitivity
has often been confused with introversion and neuroticism in previous
research on personality (see also Aron et al., 2012).
A low sensory threshold, an important characteristic of high SPS, is
also present in different sensory processing patterns and disorders,
such as “Sensory Sensitivity”and “Sensory Avoiding”(Dunn, 2001),
“Sensory Defensiveness”(Ayres, 1963)and“Sensory Over-
Responsivity”(SOR; Miller, Anzalone, Lane, Cermak, & Osten, 2007). It
Personality and Individual Differences 92 (2016) 80–86
⁎Corresponding author at: Ghent University, Department of Experimental Clinical and
Health Psychology, Henri Dunantlaan 2, B-9000 Ghent, Belgium.
E-mail address: Soﬁe.Boterberg@UGent.be (S. Boterberg).
URL: http://www.ekgp.ugent.be (S. Boterberg).
0191-8869/© 2015 Elsevier Ltd. All rights reserved.
Contents lists available at ScienceDirect
Personality and Individual Differences
journal homepage: www.elsevier.com/locate/paid
is important to note that, although SPS seems to be associated with
these sensory processing patterns and disorders, it concerns a tempera-
mental trait and should therefore not be confused with these disorders.
However, the conceptual overlap between these various constructs
shows the extensive interdisciplinary interest in characteristics of hy-
persensitivity and emphasizes the theoretical and practical importance
of the concept.
The processing of sensory events, as a part of everyday life, is sug-
gested to have a signiﬁcant impact on human experience and behavior.
In adults, high SPS is associated with high levels of stress, symptoms of
ill-health, alexithymia, anxiety and depression (Benham, 2006; Liss,
Mailloux, & Erchull, 2008; Liss, Timmel, Baxley, & Killingsworth, 2005),
and in combination with a negative childhood environment, also with
negative affectivity and shyness (Aron, Aron, & Davies, 2005). Sensory
processing may interfere with the participation in daily activities, and
social, cognitive, and sensorimotor development in children as well
(Dunn, 2001). Despite the fact that no researchseems to directly exam-
ine the association between high SPS and problems in the daily func-
tioning of children, a number of studies examined the relationship
with different sensory processing patterns and disorders. Although
temperamental SPS and the different sensory processing patterns and
disorders are not the same, they do have a low sensory threshold in
common and can thus provide preliminary insight into the association
between high SPS and problems in daily functioning. Research showed
that “Sensory Sensitivity”is associated with sleeping and behavioral
problems (Reynolds, Lane, & Thacker, 2012; Shochat, Tzischinsky, &
Engel-Yeger, 2009), and ritualism and obsessive compulsive disorder
(OCD) symptoms (Dar, Kahn, & Carmeli, 2012). “Sensory Defensive-
ness”is related to eating, learning and other social, emotional and be-
havioral problems (Smith, Roux, Naidoo, & Venter, 2005; Stephens &
Royeen, 1998). “Sensory Over-Responsivity”is related to internalizing
and externalizing problems, impaired emotion regulation, and less adap-
tive social behavior, and seems to be more frequently present in children
with clinically signiﬁcant anxiety (Ben-Sasson, Carter, & Briggs-Gowan,
2009; Conelea, Carter, & Freeman, 2014). Further, research from
Gourley, Wind, Henninger, and Chinitz (2013) found that in a sample of
children with a wide range of developmental and behavioral diagnoses
the presence of sensory processing difﬁculties was related with more
internalizing and externalizing behavioral problems. Furthermore,
in children with an autism spectrum disorder (ASD), ‘Sensory Sensi-
tivity’,‘Sensory Avoiding’,‘Sensory Defensiveness’and ‘Sensory Over-
Responsivity’, are related with more negative emotional reactions and
more fear (Baranek, David, Poe, Stone, & Watson, 2006; Ben-Sasson,
Hen, et al., 2009; Green, Ben-Sasson, Soto, & Carter, 2012; Green &
Ben-Sasson, 2010; Kientz & Dunn, 1997; Pfeiffer, Kinnealey, Reed, &
Herzberg, 2005). Overall, it can be concluded that different aspects of
increased SPS seem to be mainly associated with internalizing prob-
lems. This emphasizes the need for a fundamental scientiﬁc framework
for understanding the temperamental trait of SPS in children.
To measure individual differences in SPS in adults, Aron and Aron
(1997) developed the self-report 27-item Highly Sensitive Person
Scale (HSPS), containing itemsthat measure sensitivity to a large variety
of stimuli, the extent to which an individual quickly feels overwhelmed
by intense sensory input, and artistic and emotional sensitivity. For re-
search purposes, theitems of the HSPS are rated on a 5- or 7-point Likert
scale. However, there is also a yes/no response format available in the
popular books and on the website of Elaine N. Aron. Despite the variety
of types of sensitivity in the items, the HSPS was initially reported to
have a one-dimensional structure (Aron & Aron, 1997)andwasshown
to have adequate reliability, content-oriented validity, and validity re-
garding relationships with conceptually related constructs (American
Educational Research Association, American Psychological Association,
& National Council On Measurement In Education, 2014; Aron & Aron,
1997; Benham, 2006; Evans & Rothbart, 2008; Liss et al., 2008;
Smolewska et al., 2006). To determine whether a person has high SPS
or not on a group level, Aron and Aron (1997) propose to use a relative
cut-off score of the top 20%. This cut-off score is based on previous re-
search which suggested that SPS in adults is best considered as a dichot-
omous category variable with a visible break point in the sample
distribution around the 10 to 35% (for an overview of the studies on the
sample distribution of SPS see Aron et al., 2012). The dimensionality of
the HSPS in adults was examined by three independent studies. Liss
et al. (2008) and Smolewska et al. (2006) revealed a post-hoc three-
factor structure, with a strong intercorrelation between the factors sug-
gesting a single higher order construct. Evans and Rothbart (2008) how-
ever, proposed a two-factor solution very similar to their model of adult
temperament (Evans & Rothbart, 2007). More recently, Aron theoretical-
ly redeﬁned the different facets of SPS using the acronym “DOES”(Aron
et al., 2012; Aron, 2010, 2012). “Depth of Processing”includes features
like empathy, conscientiousness, having intensive feelings for others,
having living dreams and a rich imagination, and the presence of a gener-
al thoughtfulness or awareness of long term consequences (i.e. “pause-
to-check approach”). “Overstimulation”refers to the presence of a more
frequent and stronger autonomic arousal towards situations which are
perceived as stressful. “Emotional Intensity”refers to the presence of
both more intense negative and positive emotional responses. Finally,
“Sensory Sensitivity”refers to the presence of a low pain threshold and
a low tolerance of high levels of sensory input, and noticing subtle differ-
ences. It can be assumed that the presence of these four characteristics
has a considerable inﬂuence on the daily functioning of children and is as-
sociated with different internalizing and externalizing behavioral prob-
lems. According to Aron and colleagues, these four factors would load
together on the unidimensional construct of SPS. However, until now
there has been a lack of empirical evidence to support this theoretical
four-factor model. Moreover, there is no explicit model available of
which items from the HSPS load on the different theoretical factors, and
some items seem to have a conceptual overlap which makes it impossible
to compose an a priori factor model.
In analogy with the adult questionnaire, a 23-item parent-report
questionnaire for children was developed and published in Aron's
book “The Highly Sensitive Child (HSC)”(Aron, 2002). It is important
to note that the items of the HSPS for children have a different content
and number compared to the adult HSPS. Unlike its adult counterpart,
the reliability, distribution, validity and dimensionality have not yet
been investigated. Given the increasing use of the concept of “high sen-
sitivity”in children, an instrument objectively measuring thistrait isur-
The ﬁrst goal of the present study was to explore the underlying fac-
tor structure of Aron's 23-item parent-report HSPS for children. Until
now, research only focused on the factor structure of the HSPS for
adults, resulting in a three- or two-factor model. However, based on
the fundamental differences between the HSPS for children and the
HSPS for adults, and the lack of an explicit model for the DOES-theory
in SPS, there was no a priori factor model for the HSPS in children avail-
able that could be tested, except for the one-factor structure as pro-
posed by Aron and Aron (1997). The second goal was to investigate
the association between high SPS and problems in daily functioning.
First, differences in problems in daily functioning such as antisocial be-
havior, medically unexplained physical symptoms (MUPS) and,
sleeping, eating and drinking problems between a group of children
with high SPS and a group of children with average or low SPS were
examined. Based on different studies including partial aspects of SPS
such as “Sensory Sensitivity”(Dunn, 2001), we expected that chil-
dren in the high SPS group would have more problems in their
daily functioning, especially internalizing problems. Second, differ-
ences in the factors of the HSPS and the total 23-item HSPS, as used
in clinical practice, were identiﬁed between children with few versus
many problems in daily functioning. Again, children with especially
more internalizing problems were expected to have higher SPS in
general and more speciﬁcally, were also expected to have higher
scores on the characteristic of SPS that is associated with sensory
81S. Boterberg, P. Warreyn / Personality and Individual Differences 92 (2016) 80–86
2.1. Procedure and participants
The present study was part of a broader online survey on tempera-
ment and behavioral functioning and was approved by the appropriate
ethics committee in April 2013. Participation was voluntary andanony-
mous. Caregivers of 235 children and adolescents (53.20% boys) be-
tween 3 and 16 years (M= 8.27, SD = 3.28) fully completed the 23-
item HSPS along with questions about their children's problems in
daily functioning. Initially, 258 respondents started to ﬁll out the survey
but 33 (14%) did not complete it. Based on the exploratory nature and
the relatively large sample size of the present study, we decided to not
impute the missing data. The survey was distributed in Flanders,
Belgium. Most questionnaires (n= 223; 94.89%) were ﬁlled out by
the biological mother of the child, 11 (4.68%) by the biological father
and one (0.43%) by the grandmother. Most respondents (n=189;
80.43%) and also their partners (n= 145; 61.70%) had a higher educa-
tion. To examine differences in problems in the daily functioning, two
independent samples were selected from this original sample: a sample
of childrenscoring high on SPS (i.e. high SPS group) and a control group
scoring average or low on SPS. First, the 20% children (N= 48) with the
highest total scores on the HSPS (range 98–114) were selected for the
high SPS group, as theoretically suggested by Aron and Aron (1997).
Since problems in daily functioning could be associated with the pres-
ence of a clinical diagnosis instead of temperamental sensitivity, we
excluded all children (n= 7; 14.58%) with externalizing, internalizing,
or developmental disorders.Based on parental report, clinical diagnoses
were determined by a multidisciplinary team or by accredited psychol-
ogists, speech therapists, child psychiatrists, pediatricians and pediatric
neurologists. This resulted in a high SPS group of 41 children with
19 boys and 22 girls between 4 and 15 years (M= 7.33, SD = 2.43).
The remaining 80% of children (N= 187) with an average or low total
score on the HSPS (range 39–97) comprised the control group. Again,
all children with a clinical diagnosis (n= 45; 24.06%) were excluded,
resulting in a control group of 142 children with 69 boys and 73 girls
between 3 and 16 years (M= 7.95, SD = 3.47). No signiﬁcant gender
(1) = .065, p= .799), respondent (χ
(1) = .575, p= .750), respon-
dent education (χ
(1) = 1.886, p= .930) and age (U(183) = 3000; p=
.765) differences were found between the high SPS group and the con-
trol group. Since the scores on the variable age were not normally dis-
tributed (D(183) = .160, pb.001), a non-parametric Mann–Whitney
U-test was used.
The 23-item parent-report Highly Sensitive Person Scale (HSPS; Aron,
2002).In the present study the 23-item parent-report HSPS, designed
for measuring SPS in children and adolescents, was used. The question-
naire was published in a book about high sensitivity in children (Aron,
2002) and was based on the adult version of the HSPS (Aron & Aron,
1997). The Dutch translation of the questionnaire was published by
Aron (2004) and Daele and T'Kindt (2011).Inordertoobtainsufﬁcient
variation in scores and according to previous research on the HSPS in
adults, we decided to rate each item on a ﬁve-point Likert scale (1 =
“strongly disagree”;5=“strongly agree”).
Measures of problems in daily functioning. Medically unexplained
physical symptoms (MUPS) were explored with the question: “Does
your child often complain about headaches, stomach ache/abdominal
pain or nausea without apparent medical reason?”Items that investi-
gated sleeping problems were: “When going to bed, does your child
often fall asleep within ten minutes?”and “Does your child have prob-
lems to fall back asleepwhen he/she wakes up atnight?”Further, ques-
tions regarding eating and drinking problems were: “Does your child
sometimes refuse eating different kinds of food, e.g., certain tastes or
textures, such as food with lumps?”,“Does your child sometimes eat
excessively or is it sometimes difﬁcult to stop him/her from eating?”,
“Does your child sometimes drink excessively or is it sometimes difﬁcult
to stop him/her from drinking?”and “Does your child not drink at all
during an entire day?”Furthermore, thepresence of antisocial behavior
was also explored: “Does your child often lie to others or deceive
others?”,“Does your child sometimes steal things?”and “Does your
child often argue or ﬁght with other children, or does he/she bully
them?”Finally, the question “Did your child cry often when he/she
was a baby?”was asked. In accordance with the HSPS, each item
was rated from one to ﬁve (1 = “almost never true”;5=“almost al-
2.3. Statistical analyses
First, Cronbach's alpha was computed as a measure of internal con-
sistency and the distribution of the HSPS total scores was examined to
validate if SPS in children can be considered as a dichotomous variable.
Second, to extract the underlying factor structure of the HSPS in chil-
dren, an Exploratory Factor Analysis (EFA) with oblimin oblique rota-
tion was applied to 60% of the sample. To select the number of factors
to retain, both the Scree Plot (Cattell, 1966), the Velicer (1976) mini-
mum average partial (MAP) test and Horn's Parallel Analysis (Horn,
1965) were conducted. Third, a Conﬁrmatory Factor Analysis (CFA)
was conducted on the other 40% of the sample to evaluate and compare
the ﬁt of the exploratory factor solution with a one-factor solution.
Fourth, the proportion of children in the high SPS and in the control
group with problems in daily functioning was compared using chi-
square tests. Finally, we compared the HSPS total (23-items) and the
HSPS factor scores, based on the results of the exploratory analysis, of
children with no or little problems to children with some or many prob-
lems in both groups using independent sample t-tests. As a measure of
practical signiﬁcance, Cohens' dwas reported (Cohen, 1988).
Within the group analyses, assumptions for parametric testing
(additivity and linearity, normality, homogeneity of variance and inde-
pendence, see also Field, 2013) were met. Further, since most of the
items about problems in daily functioning lacked sufﬁcient variation
of scores or suggested a bimodal distribution, it was decided to dichot-
omize the items (0 = “no or little problems”and 1 = “some or many
problems”). The item about stealing was removed from further analyses
due to a ﬂoor effect of the scores. Most analyses were executed in IBM
SPSS Statistics for Windows, Version 20.0 (IBM Corp., 2013). Both the
Velicer MAP test and Horn's Parallel Analysis were conducted in SPSS
using a syntax from O'Connor (2000). The CFA was conducted in
Amos 7.0 (Arbuckle, 2006) and to test for multivariate normality we
used the program R (R Development Core Team, 2008).
3.1. Internal consistency and distribution of the HSPS
The scoresof the 23-item parent-report HSPShad a Cronbach's alpha
of .91 (N= 235; 95% CI [.90,.93]) suggesting excellent internal consis-
tency (DeVellis, 2012). The total scores of the HSPS did not deviate
from the normal distribution (D(235) = .912, p= .376).
3.2. Exploratory factor analysis of the HSPS
The sample selected randomly for the EFA consisted of 140cases and
had an alpha coefﬁcient of .91 (95% CI [.89,.93]) for the scores of the
HSPS. Assumptions (non-multicollinearity, sampling adequacy and
factorability) for EFA were met. The Kaiser–Meyer–Olkin measure of
sampling adequacy evaluates tests of ﬁt, and ﬁndings were very good
at 0.874 (Hutcheson & Sofroniou, 1999). Bartlett's test of sphericity
was signiﬁcant (χ
(253) = 1501,133, pb.001). Using the MVN package
in R, the p-values for Mardia's multivariate skew and Mardia's multivar-
iate kurtosis were b.001 which indicated that the data are not
82 S. Boterberg, P. Warreyn / Personality and Individual Differences 92 (2016) 80–86
multivariate normally distributed. Therefore, it was decided to use the
principal axis factors method to extract the factors (Costello &
Osborne, 2005). The EFA was followed by an oblique rotation (direct
oblimin) to get a theoretically more accurate and reproducible solution
(Costello & Osborne, 2005). The Scree Plot (Cattell, 1966)indicatedthat
a two factor solution was optimal (eigenvalues: 8.06, 2.57, 1.38, 1.24,
1.13). Further, both the Velicer (1976) MAP test and Horn's Parallel
Analysis (Horn, 1965) also suggested extraction of two factors. Hence,
a two-factor solution was speciﬁed andaccounted for 41,38% of the var-
iance in theitems (32.55% and 8.83%). Individual items were retained as
indicators of a factor if their loading on that factor was larger than .364
(Stevens, 2002). When an item loaded higher than .364 on two factors,
it was also eliminated from further analysis (Costello & Osborne, 2005).
These criteria resulted in the elimination of four items: “My child no-
tices the slightest unusual odor.”(item 7), “My child notices subtleties.”
(item 20), “My child considers if it is safe before climbing high.”(item
21), and “My child feels things deeply.”(item 23). Table 1 shows the ro-
tated factors and their respective items and provides mean inter-item
correlations and alphas for each of the two factors. There was acceptable
internal consistency for both factors (DeVellis, 2012). Since the scores of
factor two were not normally distributed in this sample, Spearman's
correlation coefﬁcient was calculated (Hauke & Kossowski, 2011),
showing a moderate intercorrelation (ρ=.48,pb.001) between factor
one and two. Although the two factors accounted for nearly half of the
variance in the items, Cronbach's alpha of the remaining 19 items was
.89 (95% CI [.86,.91]), suggesting good internal consistency for the scores
on the HSPS.
3.3. Conﬁrmatory factor analysis of the HSPS
The sample selected randomly for the CFA consisted of the other 95
cases and had an alpha coefﬁcient of .92 (95% CI [.89,.94]) for the scores
of the HSPS. The ﬁt of a one-factor model was evaluated and compared
statistically to the ﬁt of a two-factor model, based on the foregoing EFA.
, root mean square error of approximation (RMSEA) and compar-
ative ﬁt index (CFI) are reported. An RMSEA of .08 or less is generally
considered an acceptable ﬁt, and ﬁts of .90 or greater are considered
acceptable for the CFI (Hu & Bentler, 1999). First, we examined the orig-
inal one-factor model as suggested by Aron and Aron (1997). According
to traditional cut-offs, the ﬁt was not acceptable: χ
(230) = 589.216
(pb.001), RMSEA = .129, CFI = .65. Next, a CFA was conducted to ex-
amine the two-factor solution: χ
(151) = 362.906 (pb.001), RMSEA =
.122, CFI = .73. Although the two-factor solution also fell short on the
test comparing the difference between
the one- and the two-factor model showed that the latter model ﬁts
the data better [χ
(79) = 226.31 (pb.001)].
3.4. The association between SPS and problems in daily functioning
Differences in reported problems betweenhigh SPS and control children.
Table 2 shows the percentages of children in the high SPS or control
group with reported problems. Within the high SPS group there were
proportionally more children with medically unexplained physical
symptoms (MUPS) (χ
(1) = 15.833, pb.005), problems to fall asleep
(1) = 7.610, pb.008) and fall back asleep (χ
(1) = 13.120, pb
.005), and eating problems (χ
(1) = 14.900, pb.005). On a 5% signiﬁ-
cance level, there were proportionally less children in the high SPS
group who were reported as sometimes or regularly lying anddeceiving
(1) = 5.312, pb.05).
Differences in SPS between children scoring low vs. high on reported
problems. Children with a high level of MUPS had signiﬁcantly higher
scores on the HSPS-total, HSPS-OS and HSPS-DP ((t(181) = 5.392,
pb.002, d=.8),(t(181) = 5.010, pb.002, d=.7),and(t(181) =
4.764, pb.002, d= .7), respectively). Regarding sleeping problems, chil-
dren with a high level of problems falling asleep had higher scores on
the HSPS-total (t(181) = 3.712, pb.002, d= .6) and HSPS-OS
(t(181) = 4.214, pb.002, d= .6), and on a 5% signiﬁcance level also
on the HSPS-DP (t(181) = 2.508, pb.05, d= .4). Children with a high
level of problems falling back asleep had also higher scores on the
HSPS-total, HSPS-OS and HSPS-DP ((t(181) = 6.467, pb.002, d=
1.0), (t(181) = 7.433, pb.002, d=1.1),and(t(181) = 3.647,
pb.002, d= .5), respectively). Further, children with a high level of eat-
ing problems had higher scores on the HSPS-total and HSPS-OS
((t(181) = 4.606, pb.002, d=.7),and(t(181) = 6.001, pb.002,
d= .9), respectively). On a 5% signiﬁcance level, children who drink ex-
cessively during the day had higher scores on the HSPS-total (t(181) =
2.052, pb.05, d= .3) and HSPS-OS (t(181) = 2.099, pb.05, d= .3).
Children who do not drink enough had higher scores on the HSPS-
total (t(181) = 3.905, pb.002, d= .6) and HSPS-OS (t(181) = 3.922,
pb.002, d= .6) and on a 5% signiﬁcance level also on the HSPS-DP
(t(181) = 2.871, pb.05, d= .4). Furthermore, on a 5% signiﬁcance
level, children with a high level of lying and deceiving had lower scores
on the HSPS-total and HSPS-DP ((t(181) = 2.643, pb.05, d= .4) and
(t(181) = 2.897, pb.05, d= .4), respectively). Children with a high
level of arguing, ﬁghting and bullying had lower scores on the HSPS-
total (t(181) = 2.953, pb.05 d= .4), HSPS-OS (t(181) = 2.219,
Proportion of the high SPS and control group with reported problems.
% High SPS (n= 41) % Control (n= 142) χ
MUPS 82.93 47.89 15.833
Problems falling asleep 56.10 32.39 7.610
Problems falling back
80.49 48.59 13.120
Eating problems 68.29 34.51 14.900
Excessive eating 56.10 58.45 .072
Excessive drinking 53.66 44.37 1.104
Not drinking enough 63.41 54.23 1.092
Lies or deceives 31.71 52.11 5.312⁎
Argues, ﬁghts or bullies 19.51 33.10 2.794
Excessive crying 48.78 42.25 .551
Note. SPS = Sensory Processin g Sensitivity; MUPS = Medically Unexplained Physical
Signiﬁcant after Bonferroni–Holm correction.
Exploratory factor analysis with oblimin direct rotation (pattern matrix).
19. My child is bothered by noisy places.
11. My child doesn't do well with big changes.
3. My child doesn't usually enjoy big surprises.
1. My child startles easily.
22. My child performs best when strangers aren't present.
2. My child complains about scratchy clothing, seams in
socks, or labelsagainst his/her skin.
12. My child wants to change clothes if wet or sandy.
16. My child prefers quiet play.
10. My child is hard to get to sleep after an exciting day.
18. My child is very sensitive to pain.
9. My child seems very intuitive.
17. My child asks deep, thought-provoking questions.
5. My child seems to read my mind.
8. My child has a clever sense of humor.
6. My child uses big words for his/her age.
15. My child notices the distress of others.
13. My child asks a lot of questions.
4. My child learns better from a gentle correction than strong
14. My child is a perfectionist.
Cronbach's alpha .86 .85
Cronbach's alpha 95% CI [.82, .89] [.81, .89]
Mean inter-item correlation .38 .39
Note. N = 140; (OS) Overreaction to Stimuli; (DP) Depth of Processing;
Depth of Processing;
Sensory Sensitivity, according to the DOES-model.
83S. Boterberg, P. Warreyn / Personality and Individual Differences 92 (2016) 80–86
pb.05, d= .3) and HSPS-DP (t(181) = 2.963, pb.05, d= .4). Children
who often cried when they were a baby had higher scores on the HSPS-
OS (t(181) = 3.055, pb.05, d= .5). See Table 3 for details.
4.1. The 23-item parent-report HSPS for children
The ﬁrst goal of the present study was to explore the underlying fac-
tor structure of Aron's 23-item parent-report HSPS (Aron, 2002). Anal-
ysis of internal consistency suggested that the scores of the parent
form of the HSPS are a reliable measure of SPS in children. Previous
studies of the latent structure of SPS in adults have suggested that the
variable is best considered as dichotomous with a break point around
10 to 35% (Aron et al., 2005, 2012; Aron & Aron, 1997). Hence, to decide
which children could be considered as having ‘high SPS’in the present
study, the recommendation of Aron and colleagues to consider the top
20% of the population as ‘highly sensitive’was followed. However, sim-
ilar to the ﬁndings from another independent study on SPS in adults
(Benham, 2006), the results of the present study demonstrated that
SPS in children is best considered as a continuous variable without a
clear cut-off. Further research in both adults and children is needed to
determine whether the trait of SPS is best considered as normally dis-
tributed or as a dichotomous (binominal) trait.
In contrast to Aron and Aron's (1997) conclusion that the HSPS mea-
sures a unidimensional construct, the present results supported a two-
factor structure. Although the two-factor solution was retrieved in a
very robust way in the exploratory analysis, it did not have a good tra-
ditional ﬁt in the conﬁrmatory analysis. However, conﬁrmatory analysis
showed that the two-factor solution ﬁts the data better than the one-
factor solution. Possible explanations for the bad ﬁtofthemodels
could be the fact that the data are not multivariate normally distributed
and/or the rather small sample size (only 40% of the total sample was
analyzed). Hence, further research on the ﬁt of the underlying factor
structure of the HSPS in children is recommended. In the present
study, the two factors could be theoretically interpreted according to
the recent DOES-conceptualization of SPS (Aron et al., 2012; Aron,
2010, 2012). The ﬁrst factor, labeled “Overreaction to Stimuli”(OS),
included 10 items and seemed to be related with the following charac-
teristics of SPS: “Overstimulation”,“Emotional Intensity”and “Sensory
Sensitivity”. The second factor included 9 items and was mostly related
to the characteristic “Depth of Processing”(DP). As expected, there was
no clear conceptual overlap with previously found factor structures in
research with adults (Evans & Rothbart, 2008; Liss et al., 2008;
Smolewska et al., 2006). Next to the fact that the items differ in number
and content between the adult and child version of the HSPS and the use
of parent-report in children versus self-report in adults, a possible expla-
nation could be that SPS in childhood is expressed in a different way
than in adulthood. It is possible that SPS in adulthood is inﬂuenced by
different factors such as education, the social environment, life stressors,
and other various aspects in an individual's life (cf. differences between
temperament and personality, see e.g., Evans & Rothbart, 2007;
Rothbart,Ahadi,Hershey,&Fisher,2001). The moderate positive inter-
correlation among the factors is consistent with a general, higher-
order construct of SPS that was also found in previous research with
adults (Liss et al., 2008; Smolewska et al., 2006).
4.2. The association between SPS and problems in daily functioning
A second goal of this study was to examine the relationship between
SPS and problems that may arise in the daily functioning of children.
Children with high SPS showed more MUPS (i.e. headaches or stomach
ache without an apparent medical reason) and more sleeping and
eating problems compared to children with average or low SPS. In ac-
cordance, children with more MUPS, sleeping, eatingand drinking prob-
lems showed high SPS in general and, more speciﬁcally, high OS. Crying
excessively as a baby was only related with high OS. Hence, this factor
seems to be related with more problems in the daily functioning com-
pared to the second factor DP. We can assume that a high OS, which in-
cludes also(hyper)sensitivity for sensory stimuli (a characteristic which
is associated with different sensory processing disorders), has a nega-
tive inﬂuence on the life of some children in making their dailyfunction-
ing more difﬁcult, even as a baby. However, children with more MUPS
and sleeping problems showed also high DP, which is consistent with
the idea that DP refers to a general thoughtfulness or a sense of long-
term consequences (Aron et al., 2012; Aron, 2010, 2012), possibly
Descriptives of HSPS factor and total scores for children scoring low versus high on the reported problems.
HSPS-OS (n= 10)
HSPS-total (n= 23)
Low High t(181) dLow High t(181) dLow High t(181) d
Problems falling asleep 33.32(7.80)
Problems falling back asleep 30.86(7.13)
Eating problems 32.42(7.44)
1.794 .3 81.92(14.30)
Excessive eating 35.53(8.03)
.525 .1 35.73(6.77)
.894 .1 87.23(15.67)
Excessive drinking 34.03(8.60)
1.278 .2 83.95(16.31)
Not drinking enough 32.65(8.25)
Lies or deceives 36.21(7.97)
1.869 .3 36.47(6.08)
Argues, ﬁghts or bullies 36.02(8.02)
Excessive crying 33.62(7.64)
−.060 .0 84.18(14.98)
Note. HSPS = Highly Sensitive Person Scale; OS = Overreaction to Stimuli; DP = Depth of Processing;MUPS = Medically Unexplained Physical Symptoms.
Signiﬁcant after Bonferroni–Holm correction.
Cohen's' d≥.8 (large).
Cohen's d≥.5 (medium).
84 S. Boterberg, P. Warreyn / Personality and Individual Differences 92 (2016) 80–86
implying that these children are more prone to worry and ruminate
about the present and the future, which may lead to internalizing prob-
lems. In general, the present results are complementary to previous
studies addressing the association between problems in daily function-
ing and high SPS in adults (Benham, 2006) or other aspects of hypersen-
sitivity in children (e.g., Reynolds et al., 2012), supporting the idea that
high SPS in children may interfere with the participation in daily
Further,in the present study, children with high SPS showed less an-
tisocial behavior. Children who were reported to lie, deceive, argue,
ﬁght or bully regularly, showed low SPS and low DP and OS. However,
these results were only signiﬁcant on a 5% level and must therefore be
interpreted with caution. Overall, it can be concluded that SPS seems
to be associated with more internalizing problems.
Finally, despite the intercorrelation of the two subscales and a high
internal consistency for the overall scale, the present results suggests
that the subscales of the HSPS are valuable to consider separately both
on the level of interpretation and especially because of their different
associations with problems in daily functioning. However, since the
two subscales only account for half of the variation in total scores, it re-
mains valuable to also include thetotal score in future studies using the
23-item parent-report HSPS in children.
There are somelimitations in the present study. First, since the scope
of the current investigation wasto examine the parent-report HSPS as it
is currently used in clinical practice and on theinternet, we decided not
to undertake an ofﬁcial translation process. However, undertaking an
ofﬁcial translation process may be useful in future studies. Second,
given the cross-sectional, correlational nature of the present study, a
causal relationship between measures of SPS and problems in the
daily functioning of children cannot be inferred. Environmental factors
such as parental warmth and exposure to stressful life events, and
child factors such as high neuroticism, may, at least partly, account for
the observed correlations (Aron et al., 2005). Further longitudinal re-
search is needed to address the possible mediating or moderating fac-
tors in the relationship between SPS and internalizing and
externalizing behavioral problems in children.
4.4. Future research directions and implications
The present study can be seen as an exploratory study since it is the
ﬁrst which examines the parent-report HSPS in children. Our main goal
is to encourage further research in temperamental sensitivity and we
hope this can be a ﬁrst step towards further investigation of SPS in chil-
dren. Hence, there are several aspects that are in need of further study.
First, to further explore the psychometric properties of the scale, an in-
vestigation of the discriminant and convergent validity against other
measures of temperament, sensitivity and behavior would be valuable.
Second, to corroborate the present ﬁndings on the association between
SPS and problems in daily functioning, additional studies could imple-
ment other measures of hypersensitivity and temperament [e.g., the Sen-
sory Proﬁle (Dunn, 1994), the SensOR (Schoen, Miller, & Green, 2008)
and the Child Behavior Questionnaire (CBQ; Rothbart et al., 2001)], stan-
dardized measures of problems in daily functioning [e.g. the Child Behav-
ior Checklist (CBCL; Achenbach & Rescorla, 2001)], and experimental
measures on the behavioral, perceptual or neurophysiological level that
are associated with high SPS. Further, a very recent 12-item child self-
report version of the HSPS (Pluess & Boniwell, 2015; Pluess et al., in
preparation) could be converged with the parent-report version in
future research. Finally, further research on the HSPS in children could
also apply some more sophisticated statistical methods such as tests
of measurement invariance. Although further research is needed, the
present results encourage addressing the temperamental trait of SPS in
children in both psychological research and practice. Recently, Pluess
and Boniwell (2015) provided evidence that high SPS could predict a
better treatment response, probably based on a deeper processing of
the content of the intervention than individuals scoring low on SPS.
Therefore, during the diagnostic process in psychological practice, it can
be recommended to conduct a personality analysis which also contains
a measure of SPS such as the HSPS. By considering both subscales
Overreaction to Stimuli and Depth of Processing of information, a
broader perspective on the daily functioning of the child or adolescent
could be obtained.
In sum, the current exploratory study provided the ﬁrst evidence for
a two-factor structure of the 23-item parent-report HSPS for children,
together with the absence of a clear cut point. High SPS was associated
with more internalizing and probably also less externalizing problems.
The ﬁrst factor OS was associated with excessive crying as a baby,
more medically unexplained physical symptoms (MUPS), more
sleeping, eating, and drinking problems while the second factor DP
was associated with more MUPS and more sleeping problems. Hence,
OS seems to be associated with more problems in the daily functioning
compared to DP. The HSPS may therefore provide valuable information
in the assessment of children and adolescents with problems in daily
We would like to express our gratitude to the participating parents
in the present study and especially to the organizations who helped
with the recruitment: vzw Victor, SIG vzw and HSP Vlaanderen. Further,
we would liketo thank Emmanuel Abatih from FIRE Statistical Consult-
ing at Ghent University for his help concerning the statistical analyses.
Finally, we would also like to thank prof. dr. Herbert Roeyers, for his
constructive feedback on a previous draft of the manuscript.
Achenbach,T. M., & Rescorla, L.A. (2001). Manual for the ASEBAschool-age forms & proﬁles.
Burlington, VT: University of Vermo nt, Research Cente r for Children, You th, and
American Educat ional Research Assocation, American Ps ychological Association and
National Council On Measurement In Educa tion (2014). Standards for educational
and psychological testing (p. 230). Washington, DC: American Educational Research
Arbuckle, J. L. (2006). Amos (version 7.0). Chicago: SPSS.
Aron, E. N. (2002). The highly sensitive child: Helping our children thrive when the world
overwhelms them (1st ed., p. 352). New York: Harmony.
Aron, E. N. (2004). Het hoogsensitieve kind: Help je kinderen op te groeien in een wereld die
hen overweldigt. (1st ed., p. 379). Amsterdam: Archipel.
Aron, E. N. (2010). Psychotherapy and the highly sensitive person: Improving outcomes for
that minority of people who are the majority of clients (1st ed., p. 288). New York:
Aron, E. N. (2012). Temperament in psychotherapy: Reﬂections on clinical practice with
the trait ofsensitivity. In M. Zentner, & R. L. Shiner (Eds.), Inhandbook of temperament
(pp. 645–672). New York: Guilford Press.
Aron, E. N., & Aron, A. (1997). Sensory-processing sensitivity and its relation to introver-
sion and emotionality. Journal of Personality and Social Psychology,73(2), 345–368.
Aron, E. N., Aron, A., & Davies, K. M. (2005). Adult shyness: The interaction of tempera-
mental sensitivity and an adverse childhood environment. Personality and Social
Psychology Bulletin,31(2), 181–197. http://dx.doi.org/10.1177/0146167204271419.
Aron, E. N., Aron, A., & Jagiellowicz, J. (2012). Sensory processing sensitivity: A review in
the light of the evolution of biological responsivity. Personality and Social Psychology
Review,16(3), 262–282. http://dx.doi.org/10.1177/1088868311434213.
Ayres, A. J. (1963). Eleanor Clarke Slagle lecture –the development of perceptual –motor
abilities: A theoretical basis for treatment of dysfunction. American Journa l of Occupation-
al Therapy,27, 221–225.
Baranek, G. T., David, F. J., Poe, M. D., Stone, W. L., & Watson, L. R. (2006). Sensory expe-
riences questionnaire: Discriminating sensory features in young children with au-
tism, developmental delays, and typ ical development. Journal of Child Psychology
and Psychiatry,47(6), 591–601. http://dx.doi.org/10.1111/j.1469-7610.2005.01546.x.
Benham, G. (2006). The highly sensitive person: Stress and physical symptom reports.
Personality and Individual Differences,40(7), 1433–1440. http://dx.doi.org/10.1016/j.
85S. Boterberg, P. Warreyn / Personality and Individual Differences 92 (2016) 80–86
Ben-Sasson, A., Carter,A. S., & Briggs-Gowan, M. J. (2009). Sensory over-responsivity in el-
ementary school: Prevalence and social–emotional correlates. Journal of Abnormal
Child Psychology,37(5), 705–716. http://dx.doi.org/10.1007/s10802-008-9295-8.
Ben-Sasson, A., Hen, L., Fluss, R., Cermak, S. A., Engel-Yeger, B., & Gal, E. (2009). A meta-
analysisof sensory modulation symptoms inindividuals with autism spectrum disor-
ders. Journal of Autism and Developmental Disorders,39(1), 1–11. http://dx.doi.org/10.
Cattell, R. B. (1966). The scree test for the number of factors. Multivariate Behavioral
Research,1(2), 245–276. http://dx.doi.org/10.1207/s15327906mbr0102_10.
Cohen,J. (1988). Statisticalpower analysis forthe behavioral sciences (2nd ed.).New Jersey:
Lawrence Erlbaum Associates.
Conelea, C.A., Carter, A. C., & Freeman, J. B. (2014). Sensory over-responsivity in a sample
of children seeking treatment for anxiety. Journal of Developmental and Behavioral
Pediatrics,35(8), 510–521. http://dx.doi.org/10.1097/DBP.0000000000000092.
Costello, A. B., & Osborne, J. W. (2005). Best practices in exploratory factor analysis: Four
recommendations for getting the most fr om your analysis. Practical Assessment,
Research& Evaluation,10(7), 1–9 Retrieved from http://pareonline.net/pdf/v10n7.pdf
Daele, I. van den, & T’Kindt, L. (2011). Mijn kind is hooggevoelig (1st ed., p. 240). Tielt, Bel-
Dar, R., Kahn, D. T., & Carmeli, R. (2012). The relationship between sensory processing,
childhood rituals and obsessive–compulsive symptoms. Journal of Behavior Therapy
and Experimental Psychiatry,43(1), 679–684. http://dx.doi.org/10.1016/j.jbtep.2011.
DeVellis, R. F. (2012). Scale development: Theory and applications (3rd ed., p. 216). USA:
SAGE: University of North Carolina, Chapel Hill.
Dunn, W. (1994). Performance of typical children on the sensoryproﬁle: An item analysis.
American Journal o f Occupational Therapy,48(11), 967–974. http://dx.doi.org/10.
Dunn, W. (2001). The sensations of everyday life: Empirical, theoretical, and pragmatic
considerations. The American Journal of Occupational Therapy ,55(6), 608–620.
Evans, D. E., & Rothbart, M. K. (2007). Development of a model for adult temperament.
Journal of Research in Personality,41, 868–888. http://dx.doi.org/10.1016/j.jrp.2006.
Evans, D. E.,& Rothbart, M. K. (2008). Temperamental sensitivity: Two constructs or one?
Personality and Individual Differences,44(1), 108–118. http://dx.doi.org/10.1016/j.
Field, A. (2013). Discovering statistics using IBM SPSS statistics (4th ed., p. 916). London,
Gourley,L., Wind, C., Henninger, E. M., & Chinitz, S. (2013). Sensoryprocessing difﬁculties,
behavioral problems, and parental stress in a clinical population of young children.
Journal of Child and Family Studies,22, 912–921. http://dx.doi.org/10.1007/s10826-
Green, S. A., & Ben-Sasson, A. (2010). Anxiety disorders and sensory over-responsivity in
children with autism spectrum disorders: Is there a causal relationship? Journal of
Autism and Develo pmental Disorders,40,1495–1504. http://dx.doi.org/10.1007/
Green, S. A., Ben-Sasson, A., Soto, T. W., & Carter, A. S. (2012). Anxiety and sensory over-
responsivity in toddlers with autism spectrum disorders: Bidirectional effects across
time. Journal of Autism and Developmental Disorders,42(6), 1112–1119. http://dx.doi.
Hauke, J., & Kossowski, T. (2011). Comparison of values of Pearson's and Spearman's
correlation coefﬁcients on the same sets of data. Quaestiones Geographicae,30(2),
Horn, J. L. (1965). A rationale and test for the n umber of factors in fact or analysis.
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for ﬁt indexes in covariance structure anal-
ysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A
Multidisciplinary Journal,6(1), 1–55. http://dx.doi.org/10.1080/10705519909540118.
Hutcheson, G., & Sofroniou, N. (1999). The multivariate social scientist. London: SAGE.
IBM Corp. (2013). IBM SPSS statistics for Windows, version 20.0. Armonk, NY: IBM Corp.
Kientz, M. A., & Dunn, W. (1997). A comparison of the performance of children with and
without autism on the sensory proﬁle. American Journal of Occupational Therapy,
51(7), 530–537. http://dx.doi.org/10.5014/ajot.51.7.530.
Liss, M., Mailloux, J., & Erchull, M. J. (2008). The relationships between sensoryprocessing
sensitivity, alexithymia, autism, depression, and anxiety. Personality and Individual
Differences,45(3), 255–259. http://dx.doi.org/10.1016/j.paid.2008.04.009.
Liss, M., Timmel, L., Baxley, K., & Killingsworth, P. (2005). Sensory processing sensitivity
and its relation to parental bonding, anxiety, and depression. Personality and
Individual Differences,39(8), 1429–1439. http://dx.doi.org/10.1016/j.paid.2005.05.007.
Miller, L. J., Anzalone, M. E., Lane, S.J., Cermak, S. A., & Osten, E. T. (2007). Concept evolu-
tion in sensory integration: A proposed nosology for diagnosis. The American Journal
of Occupational Therapy,61(2), 135–140. http://dx.doi.org/10.5014/ajot.61.2.135.
O'Connor, B. P. (2000). SPSS and SAS programs for determining the number of compo-
nents using para llel analysis and Velicer's MAP test. Behav ior Research Method s,
Instruments, & Computers,32,396–402. http://dx.doi.org/10.3758/BF03200807.
Pfeiffer, B., Kinnealey, M., Reed, C., & Herzberg, G. (2005). Sensory modulation and affec-
tive disorders in children andadolescents with Asperger disorder.American Journal of
Occupational Therapy,59(3), 335–345. http://dx.doi.org/10.5014/ajot.59.3.335.
Pluess, M., Assary, E.,Bosman, A., Bakker, J., Walda, S.,Lester, K.J., et al. Sensory-processing
sensitivity in children (in preparation).
Pluess, M., & Boniwell, I. (2015). Sensory-processing sensitivity predicts treatment re-
sponse to a school-based depression prevention program: Evidence of vantage sensi-
tivity. Personality and Individual Differences,82,40–45. http://dx.doi.org/10.1016/j.
R Development Core Team (2008). R: A language and environment for statistical computing.
Vienna, Austria: R Foundation for Statistical Computing (Retrieved from http://www.
Reynolds, S., Lane, S. J., & Thacker, L. (2012). Sensory processing, physiological stress, and
sleep behaviors in children with and withoutautism spectrum disorders. Occupation,
Participation and Health,32(1), 246–25 8. http://dx.doi.or g/10.3928/15394 492-
Rothbart, M. K., Ahadi, S. A., Hershey, K. L., & Fisher, P. (2001). Investigations of tempera-
ment at 3–7 years: The children's behavior questionnaire. Child Development,72(5),
Schoen,S. A., Miller, L. J., & Green,K. E. (2008). Pilot studyof the sensory over-responsivity
scales: Assessment and inventory. A merican Journal of Occupational Therapy,62 ,
Shochat, T., Tzischinsky, O., & Engel-Yeger, B. (2009). Sensory hypersensitivity as a con-
tributing factor in the relation between sleep and behavioral disorder s in normal
schoolchildren. Behavioral Sleep Medicine,7(1), 53–62. http://dx.doi.org/10.1080/
Smith, A. M., Roux, S., Naidoo, N. T. R., & Venter, D. J. L. (2005). Food choice of tactile de-
fensive c hildren. Nutrition,21(1), 14–19. http://dx.doi.org/10.1016/j.nut.2004.09.004.
Smolewska, K. A., McCabe, S. B., & Woody, E. Z. (2006). A psychometric evaluation of the
highly sensitive person scale: The components of sensory-processing sensitivity and
their relation to the BIS/BAS and “Big Five.”.Personality and Individual Differences,
40(6), 1269–1279. http://dx.doi.org/10.1016/j.paid.2005.09.022.
Stephens, C. L., & Royeen, C. B. (1998). Investigation of tactile defensiveness and self-
esteem in typically developing children. Occupational Therapy International,5(4),
Stevens, J. P. (200 2). Applied multivariate statistics for the social sciences (4th ed.).
Hillsdale, NJ: Erlbaum.
Velicer, W. F. (1976). Determining the number of components from the matrix of partial
correlations. Psychometrika,41(3), 321–327. http://dx.doi.org/10.1007/BF02293557.
Wolf, M., van Doorn, G. S., & Weissing, F. J. (2008). Evolutionary emergence of responsive
and unresponsive personalities. Proceedings of the National Academy of Sciences of
the United States of America,105(41), 15825–15830. http://dx.doi.org/10.1073/pnas.
86 S. Boterberg, P. Warreyn / Personality and Individual Differences 92 (2016) 80–86