Clay Art Therapy for patients with depression: Efficacy on emotion regulation and
insight into treatment (postscript)
1.1. World scenario of depression and treatment in the community
Depression is a major contributing factor to the global burden of disease. To measure
illness by Disability Adjusted Life Years (DALYs – number of years of life lost with
number of years lived with a disability), among the ten leading causes of disability in
the world, unipolar depressive disorder was ranked third in 2004, with 4.3% of total
DALYs. However, it will become the number one disability adjusted illness in the
world by 2030, with 6.2% of total DALYs (World Health Organisation, 2008).
Depression currently affects 350 million people over the globe (World Health
Mental health professionals worldwide have noticed the pressing need of treating
mental illness in the community that is more cost-effective than conventional
hospitalization. The King’s Fund of London suggests sustainable effort and
strength-based approach to help people with mental health problems return to work
during the rehabilitation process (McCrone, Dhanasiri, Patel, Knapp, & Lawton-Smith,
2008). One such example is that 50% among the 40,000 diagnosed schizophrenic
patients in Hong Kong, are treated primarily in the community (Hong Kong Hospital
Authority, 2011). Correspondingly, a brand new mode of mental health services – the
Integrated Community Center for Mental Wellness (ICCMW) has been innovated in
recent years that provides community-based, short-term supportive services for
mental health outpatients (Hong Kong Hospital Authority, 2011). Art therapy is one
of the supplementary treatment approaches, in addition to pharmacological
1.2. Psychotherapy for treating depression
To date, selective serotonin reuptake inhibitors (SSRIs) are still the most popular type
of antidepressant, among which fluoxetine is most typically used (Bastos, Guimaraes,
& trentini, 2013; National Institute of Mental Health of the United States, 2013).
However, nonadherence to medication and the social stigma of illness (van Geffen et
al., 2007) have drawn attention to alternative treatment methods, such as
psychotherapy. It has been pointed out depression patients have higher adherence
rate to psychotherapy and show more incentive to sharing with psychotherapists
than passively receiving pharmacological treatment (Lee, Tsang, & Kwok, 2007).
The literature showed efficacy of various psychotherapy approaches, including art
therapy, in treating major depressive disorder (MDD) in the respects of alleviation of
depression signs and symptoms (Bar-Sela, Atid, Danos, Gabay, & Epelbaum, 2007;
Gussak, 2009), strengthening psychosocial functioning (Puig, Lee, Goodwin, &
Sherrard, 2006), and better control of relapses (Thase, 2009). It is evidenced
psychotherapy is able to increase serotonin 5HT-1A receptor densities (Karlsson,
2012), whereas treatment with fluoxetine did not (Karlsson, Hirvonen, Salminen, &
Hietala, 2013). Nevertheless, psychotherapy is being criticized as lacking scientific
evidence, such as randomized controlled trials, to endorse the efficacy (Ebmeier,
Donaghey, & Steele, 2006; Kimport & Robbins, 2012).
Cognitive dysfunction is often reported in Major Depression Disorder (MDD) with the
usual difficulties in concentrating, indecisiveness, and memory (Trivedi & Greer,
2014). Another common disorder closely related to MDD and cognitive dysfunction is
alexithymia. Alexithymia is characterized as a deficit in the cognitive processing of
affect. It exhibits in the difficulty to separate emotion expression from sensational
processes and articulate inner processes with conscious, verbal form to
communication (Karlsson, Naatanen, & Stenman, 2008; Kronholm, Partonen,
Salminen, Mattila, & Joukamaa, 2007). Despite the fact that antidepressants can help
improve cognitive symptoms, it does not persist effectively (Trivedi & Greer, 2014).
Long-term psychodynamic psychotherapy or combined psychotherapy with
fluoxetine for depressed patients are evidenced to be more efficacious in improving
neurocognitive functioning than fluoxetine alone (Bastos et al., 2013). Short-term or
long-term psychotherapy are also able to positively enhancing daily functioning in
working abilities (Knekt, Lindfors, Sares-Jäske, Virtala, & Härkänen, 2013).
1.3. Right-brain processes & emotion regulation
Emotion system is complicated; each of the integrated emotional states of mind
involves cortical-subcortical coordination and connection to brain arousal systems
(e.g. norepinephrine and serotonin) and a variety of bodily states (Panksepp, 2009).
The left brain is good at processing verbal, conscious, and rational information,
whereas the right brain is an important region of processing emotion, nonverbal,
holistic, processes, and spiritual elements (Larsen, Brand, Bermond, & Hijman, 2003).
The right brain is also intimately involved in the processing of bodily-based affective
information associated with various motivation states (Schore, 2003a, 2003b). It also
has close connections with the involuntary Autonomic Nervous System (ANS) and the
cardiovascular system to enable a quick responding system to environmental threats
and danger for survival (Schutz, 2005). Interestingly, a study researching women with
alexithymia have distinctly left-lateralization of brain processes when they were
emotionally provoked. In comparison, those participants without alexithymia showed
balanced activities in the two hemispheric brains (Karlsson et al., 2008).
Studies have supported that affect can be dysregulated in either hyperarousal or
hypoarousal states (Schore, 2002, 2003a, 2003b). Under the hypoarousal mode,
when an individual is exposed to a prolonged period of despair or hopelessness, the
dominant parasympathetic dorsal vagal system of the ANS will be turned on that
produces a dissociate state of reducing heartrates, blood pressure, and respiration.
An extended state of depressed mood and emotional detachment pose negative
impacts on the limbic system, influence regulatory functions of the ANS, and could
cumulate eventually in various disorders, such as depression (Schore, 2009).
1.4. Research on art and the brain
The relationship between art and the brain have been mostly neglected (Liu & Miller,
2008). Art making involves complex neurological processes, as implied in the various
art making procedures that involve both the mind and the body (Lusebrink, 1990,
2004). Visual art appreciation comprises aesthetic responses in which the regulatory
processes of emotion and self-reflection are involved (Vessel, Starr, & Rubin, 2012).
Combining the use of fMRI with behavioral analysis, two distinct patterns of neural
activities exhibited from two different subnetworks during visual art appreciation are
identified. These distinctive networks go through the frontal cortex and subcortical
regions, such as the amygdala. It is interesting that activity in these regions showed a
gradual increase in response to the most emotionally moving artworks as perceived
by the particular participants of the experiment (Vessel et al., 2012). Literature also
shows how training in art could help to retain some brain functions, when comparing
between stroke patients with art training and those who were not (Alajouanine, 1948;
Studies on EEG documented the differences in the brain’s activities between artists
and non-artists. Results show significantly stronger long-range, delta band
synchronization between frontal and temporo-occipital electrodes in the artists
group as well as strong right hemispheric dominance in terms of synchronization
(Bhattacharya & Petsche, 2002, 2005). In artists, the higher synchrony in the
low-frequency band (delta-band) is probably owing to the participation of a higher
level and long-term visual art memory and extensive top-down processing hint at the
involvement of more complex activity in the neocortex. The results show that in
artists, functional cooperation between cortical regions during drawing from
memory was notably different from those in non-artists (Bhattacharya & Petsche,
1.5. Benefits of use of combined nonverbal-verbal form of communication in art
Clay work is a body-based psychotherapy that can be used in healing grief, depressed
mood, anger, and fear (Sherwood, 2004). As a form of clay work, Clay Art Therapy
(CAT) combines nonverbal and verbal elements in the psychotherapeutic processes.
The processes of clay-work integrates the experience and expression of
multi-sensory modalities (tactile, visual, proprioception (Elbrecht, 2013), auditory),
and kinesthetic activities in interacting with clay that range from gentle touch on clay
to intense input of physical energy (e.g. pounding, rolling, molding clay slump). The
processes of creating personally meaningful clay products require intense
participation of perceptual skills, affective expression, creating symbolically
meaningful clay works, cognitive functions (e.g. memory, decision making,
organization skills), and creative ability. These various processes can help raise the
ability to properly understand emotion and better control of emotional expression.
Through these processes, distorted thoughts and emotions can also be progressively
restored and organized into a holistic piece (Carr, 2008). Verbal expression of
emotions will eventually become easier in psychotherapeutic process that can
potentially benefit improvement of alexithymia (Nan & Ho, 2014).
Hypothetically, the complex, sequential processes of CAT can potentially activate
neural activities in various cortical-subcortical regions (Lusebrink, 2004) that can
benefit the attunement of psychophysiological processes (Carr, 2008) that helps
2. 1. Design
The study is a randomized controlled trial comparing Clay Art therapy (CAT) with
nondirective Visual Art control group (VA) that investigates the effects of clay art
making, guided by an art therapist, on depressed mood of MDD patients. Both
groups of participants are on medical treatment. Major assessments were at baseline
(T0), immediate postintervention (T1), and 3-weeks postintervention (T2).
Randomization to CAT group and VA group was done using computer-generated
Sample size was calculated to achieve a medium effect size of 0.15, at 80% power
and a significance level of 0.05 in repeated measures MANOVA under the proposed
experimental/control group, and three time-points design. Including the clinical and
demographic factors affecting intervention outcome and adjusting for a 25% attrition
rate, a total of 120 participants (60 per arm) is required.
The investigation has obtained ethical approval from the local institutional review
board and the informed consent of the participants was obtained.
Inclusion criteria: 1. Diagnosed MDD by psychiatrist, 2. Age 18 to 60, male or female,
3. Patients were on regular medical treatment, 4. The Chinese version of the
Beck-Depression-Inventory-II (BDI-II-C) was applied for screening. With reference to
the post-treatment test results from the pilot study (mean and minimum scores were
19/12 [Time 1); 18/3 [Time 2)); cut-off point for inclusion was set at 18 scores, the
upper limit of mild forms of depression (scores 14-19) (Beck, Steer, & Garbin, 1988;
Nan & Ho, 2014), 5. Participants had to be currently receiving services from an
ICCMW for depression outpatients, 6. Approximately 6-8 group members for each
group recruited on a convenience basis.
Exclusion criteria: Individuals with psychosis and/or suicidal attempts/tendencies
Participants were recruited from three ICCMWs located in different areas because
one ICCMW did not have enough samples available for this study within the
investigation period. All of the participants were referred by the social workers
responsible for them. Participants were informed of the objectives, duration, content,
process of the groups, and other ethical considerations prior to signing the consent
form to join the groups and filling out the questionnaires. Upon confirmation of
reaching the inclusion scores by the BDI-II-C, participants were randomized to CAT or
VA treatment groups. The treatment started in the following week.
Clay Art Therapy (CAT) takes the approach to divide the treatment mechanisms into
four major aspects – the Kinesthetic/Sensory, Perceptual/Affective,
Cognitive/Symbolic and the Creative processes. The primary goal of utilizing these
treatment mechanisms aimed at monitoring the dysregulated affective process by
enhancing and balancing activities on the physical aspect (Kinesthetic and Sensory)
and functioning on other psychological processes (Perceptual, Affective, Cognitive,
Symbolic, and Creative) (Hinz, 2009).
The CAT group was held with six 2.5 hours weekly sessions. Apart from clay art
making, in-depth verbal group sharing on the clay-work process and clay products
was included for reflection at the end of each session. An art therapist facilitated the
clay art making processes and an activity worker of the serving ICCMW was assigned
to be the co-facilitator for the group, in assisting the art therapist in-charge with
coordination and administrative work.
The VA group was also held with six 2.5 hours weekly session. The group was
conducted by social workers. The group contents were similar to those of the
ordinary interest class provided by the serving participating agencies that included a
variety of content based on the preferences of the participants. These activities
included making handcraft work, coloring mandalas, verbal sharing, listening to
relaxation music or a combination of these activities. All VA group members were
subsequently arranged to join the CAT group after the completion of the study.
The Chinese version of the Beck-Depression-Inventory-II (BDI-II-C) was adapted from
the original BDI–II and is an established and validated scale measuring levels of
depression (Beck et al., 1988; Byrne, Stewart, & Lee, 2004). The BDI-II-C has 21
questions, each answer being scored on a scale value from 0 to 3, with higher total
scores show more severe depressive symptoms.
The Body–Mind–Spirit–Well-Being Inventory (BMSWBI) is an established, validated
and reliable scale (Ng, Yau, Chan, Chan, & Ho, 2005) and was found to have strong
correlations with other different scales that measure similar domains (S. M. Y. Ho,
Chan, & Ho, 2004). The BMSWBI includes four scales, including Physical Distress,
Daily Functioning, Affect, and Spirituality. Spirituality consists of tranquility,
resistance to disorientation, and resilience subscales. The BMSWBI has a total of 56
items, of which 30 are negative, for the measurement of holistic health. The
inventory uses a 10-point scale (0-10 point), with negative items scored in the
The Twelve-Item General Health Questionnaire (GHQ-12) is an established and
validated scale, measuring current mental health (Chong & Wilkinson, 1989; Pan &
Goldberg, 1990; GL Assessment, 2013; Goldberg & Huxley, 1980; Goldberg &
Williams, 1988). GHQ-12 measures two areas – the inability to carry out normal
functions and the appearance of new and distressing experiences. Each item has 4
intervals from which to choose an answer.
The Twenty Item Toronto Alexithymia Scale – Chinese (TAS-20-C) has 20 items, with
each item measured on a 5-point Likert scale within which 5 of the items are
negatively arranged. The TAS-20 consists of 3 factors: (1) DIF—problem recognizing
feelings and differentiating them from the bodily sensations of emotions (7 items); (2)
DDF—problem depicting feelings to others (5 items); and (3) EOT—an externally
directed cognitive way of thinking (8 items). The TAS-20-C has shown acceptable
internal and retest reliability when applied to Chinese populations (Zhu et al., 2007).
All analyses were conducted with Statistical Package for the Social Sciences (SPSS
version 18.0, SPSS Inc., Chicago, IL). Continuous data were presented by mean and
standard deviation (SD). Categorical data were presented by frequency (percentage).
Between-group differences at baseline were assessed by chi-squared test for
categorical data and independent t-test for continuous data.
Repeated measures MANCOVA was applied to measure the multivariate outcome of
treatment, adjusting for TAS-20 at baseline, due to a significant between-group
difference in scores. Pillai’s trace (V) was used to present the multivariate outcome
instead of Wilks’ lambda (λ), which is used only when the independent variable has
more than two groups, which was not applicable in this study. Although Wilks’
lambda is widely used, it is not as powerful as Pillai’s trace, which can be used with
any number of independent variable groups (Mayers, 2013). Univariate outcomes
were performed to assess the interaction effect of group and time for each outcome
measure. The within-group changes were assessed by paired t-test for each group.
The effect size was determined using Cohen’s d statistics. Intention-to-treat analysis
was performed in this study and missing data were substituted by the final observed
3.1. Participants characteristics
A total of 120 were eligible for the study. After screening, 106 ((88.3%) met inclusion
criteria and agreed to participate in the study. Of these, 53 were randomized to CAT
group and 53 to the VA group. The flow of participants is shown in Figure 1.
Baseline sociodemographic and clinical variables are detailed in Table 1. The mean
age of the participants was 46.1 years (standard deviation [SD] = 10.5) in CAT group
and 44.1 years (SD = 10) in the VA group. Eighty-nine percent of the participants in
this study were female (90.4% in CAT group and 87.5% in VA group). A total of 44.2%
participants in the CAT group and 37.5% in the VA group had been diagnosed with
MDD in the previous 0–3 years, whereas 21.2% in the CAT group and 35.4% in the VA
group had been diagnosed with depression more than 10 years earlier. These results
reflected an interesting phenomenon, namely that the period since diagnosis did not
appear to affect the motivation of participants to join the study.
The average BDI-II-C score in the CAT group was 32.77 (SD = 9.18), defined as a
moderate form of depression according to BDI-II-C, whereas the average score in the
VA group was 28.77 (SD = 11.54), closer to the lower range of moderate depression
(29 – 63).
The average TAS-20 in the CAT group was 65.34 (SD = 9.21) and 59.98 (SD = 10.59) in
the VA group, with more than a total scores of 60 showing high Alexithymia. Except
in TAS-20 scores, no significant baseline differences in the three other measures were
identified between the groups and no significant differences in sociodemographic
variables between the two groups.
Individual scales and subscales:
As shown in Table 2, the BDI-II-C (d = -1.1), GHQ-12 (d = -0.9), and BMSWBI total
scores (d = 0.6) significantly improved in the CAT group immediately after treatment
(T1) (all p values < .001), and the improvements were all maintained at T2 (all p
values < 0.001). In the VA group, the changes in the BDI-II-C (d = -0.3), GHQ-12 (d =
-0.3), and BMSWB total scores (d = 0.3) were nonsignificant immediately at T1 and T2
(all p values > 0.05). Although the scores changed with time, the magnitude of
improvement was slight.
The TAS-20 score (d = -0.2) did not improve significantly in the CAT group at T1 (p >
0.05), but did at T2 (d = -0.4, p < 0.01). In the VA group, the change in the TAS-20
score (d = -0.1) was nonsignificant immediately at T1 and T2 (all p values > 0.05).
Although the score changed with time, the magnitude of improvement was slight
The scores of the BMSWBI subscales of the CAT group had significant improvement
immediately at T1, with Daily Functioning (d = 0.5, p < 0.001), Positive Affect (d = 0.7,
p < 0.001), Negative Affect (d = -0.6, p < 0.001), Spirit scale of Tranquility (d = 0.5, p <
0.001), and Spirit scale of Disorientation (d = -0.5, p < 0.01), and the improvements
were maintained at T2 (all p values < .001). Although Physical Distress (d = -0.1) and
Spirit scale of Resilience (d = 0.3) did not significantly improve at T1, significant
change was demonstrated at T2 (d = -0.3, p < 0.05; d = 0.5, p < 0.01, respectively).
The magnitudes of improvement of all BMSWBI subscales ranged from moderate to
strong and were generally constant.
In the VA group, only Positive Affect (d = 0.4, p < 0.05) and Negative Affect (d = - 0.4,
p < 0.05) significantly improved at T1, and only Positive Affect was maintained at T2
(d = 0.3, p < 0.05). The Spirit scale of Disorientation change was nonsignificant at T1
(d = -0.3, p > 0.05) but was significant at T2 (d = -0.5, p < 0.05). The scores of the
other BMSWBI subscales did not significantly improve at T1 or T2 (p > 0.05).
Multivariate effect of treatment:
Repeated measures MANCOVA analyses confirmed significant multivariate effects for
treatment, in interaction between group x time [v = 0.266, F (4, 94) = 3.4, p = .000,
partial η2 = 0.27]. The power to detect the effect was 0.989 (Table 3). The results
confirmed the impacts of CAT treatment. Given the significance of the overall test
result, univariate analyses were examined. The results revealed a significant effect for
interaction between group x time for BDI [F(1.8, 658.8) = 11.0, p = 0.000, η2 = 0.102],
GHQ-12 [F(1.9, 281) = 8.0, p = 0.001, η2 = 0.076], and BMS_WHOLE [F(2.0, 10472) =
4.4, p = 0.014, η2 = 0.043], but not on interaction between group x time for TAS-20
[F(1.9, 65.0) = 1.7, p = 0.185, η2 = 0.017]. The mean variance between groups in each
time point is plotted in Figure 2 that showed significant change over time.
Correlation analysis between depression signs & other aspects of change:
The correlations of depression signs, as indicated by BDI-II-C, with the other aspects
of change are shown in Table 4. At baseline, BDI-II-C was positively correlated with
GHQ-12 (R = 0.578, p < 0.01), and negatively corrected with BMS_WHOLE (R = -0.670,
p < 0.01) and TAS-20 (R = -0.298, p < 0.05). BDI-II-C was positively correlated with the
BMSBWI subscales of BMS_A (physical distress) (R = 0.496, p < 0.01), BMS_NeA
(negative affect) (R = 0.672, p < 0.01), and BMS_D_Dis (spirit_disorientation) (R =
0.346, p < 0.01); and BDI-II-C was negatively correlated with BMS_B (daily
functioning) (R = -0.400, p < 0.01), BMS_PoA (positive affect) (R = -0.407, p <0.01),
and BMS_D_Tr (spirit_tranquility) (R = -0.342, p < 0.05), but it was not correlated
with BMS_D_Res (spirit_resilience) (R = -0.070, p > 0.05).
The change in BDI-II-C at T1 was positively correlated with the changes in GHQ-12 (R
= 0.736, p < 0.01) and TAS-20 (R = 0.347, p < 0.05), and negatively correlated with
BMS_WHOLE (R = -.788, p < 0.01). The correlations were maintained in the changes
from T0 to T1 and T0 to T2. The correlation strength of BDI-II-C was strong with
GHQ-12 and BMS_WHOLE, and moderate with TAS-20.
The correlations of BDI-II-C with the BMSWBI subscales show that the change in
BDI-II-C (T1) was positively correlated with the changes in BMS_A (R = 0.555, p <
0.01), BMS_NeA (R = 0.681, p < 0.01), and BMS_D_Dis (R = 0.673, p < 0.01) and
negatively correlated with BMS_B (R = -.507, p < 0.01), BMS_PoA (R = -.695), and
BMS_D_Tr (R = -.460, p < 0.01). The correlations were maintained in the changes
from T0 to T1 and from T0 to T2. The correlation of BDI-II-C with BMS_D_Res was
nonsignificant from T0 to T1 (R = -.242, p > .05) but was significant from T0 to T2 (R =
-.323, p < 0.05). Generally, the strength of correlation of BDI-II-C with the BMSWBI
subscales ranged from moderate to strong.
4.1. Significance of the study
This study aimed at investigating the combined nonverbal–verbal form of Clay Art
Therapy as complementary treatment for depression outpatients in the community
setting. The findings support that CAT was more effective than nondirective visual art
control group in the benefits of reducing the level of depression and improving daily
functioning, general mental health, and holistic body–mind–spirit well-being.
The positive results demonstrated that short-term therapeutic clay work treatment
guided by art therapy can supplement pharmacological treatment, not only in
symptoms reduction, but also in positively strengthening daily functioning (Choi &
Park, 2012) and enhancing personal strength, intra-personal and inter-personal
connections (Sholt & Gavron, 2006). The results supported that short-term
psychotherapy can reduce psychiatric symptoms and enhance functional abilities
(Knekt et al., 2013). The use of creative arts may have also given mental health
outpatients a positive way to face the challenges in the social environment (Sato &
Yeh, 2013) and tackle their associated disabilities (Vázquez, Holtzman, Tondo, &
Baldessarini, 2015) with cost effective and strength-based treatment method.
4.2. Speculation in the improvement of cognitive ability to process affect
In an experiment comparing women with and without alexithymia (Karlsson et al.,
2008), the mean score of TAS-20 for those 10 women with alexithymia was 62.4 (SD =
1.58), with the mean age = 35.2 (SD = 5.51), that was close to the TAS-20 baseline
scores of this study: CAT group 65.34 (SD = 1.27) and VA group 59.98 (SD = 1.53).
Despite TAS-20 scores did not show significant improvement immediately after CAT
intervention, the measurement of TAS-20 was significant as assessed three weeks
after intervention. It gave hints to the trend of the improvement of cognitive function
or cognitive ability to process affect (Karlsson et al., 2008), which the VA control
group did not exhibit such change (Figure 2) although the VA control group was
under regular pharmacological treatment during the study period. 49 out of 52 CAT
group members could complete 6 sessions, with total attendance rate = 94%.
Although this study demonstrated the benefits of CAT among patients with
depression, the participants were all adults and most were women. The mean age of
the participants were 46.1 years (SD = 10.5) in the CAT group and 44.1 years (SD = 10)
in the VA control group respectively, in which 89% of the participants in the study
were female. The age group and gender bias may limit the power of this research
project and affect the generalizability of the study findings. This study proposes that
CAT be adopted as an intervention protocol with other age or population groups.
Another limitation that may create research bias lies in the design of the control
group. The use of nondirective art mixed media as control and led by a social worker
may arouse speculation from the participants that the real art therapy treatment was
in the CAT group and created Hawthorne effect and affect the research results. It
might be technically difficult and short of resources to locate another creative arts
therapist, such as, a music therapist, and compared the effects of CAT intervention
with music therapy. However, it is considerable to have a ceramicist to lead a clay
work control group, in comparing therapeutic and nontherapeutic use of clay work.
The selection of depression outpatients was a response to test the possibility to treat
mental illness in community setting. However, the setting of ICCMW poses potential
difficulty to effective data collection owing to the mobility of outpatients in
supportive community service setting, namely, ICCMW. With 44.2% of the CAT group
participants and 37.5% of the VA group participants being diagnosed with depression
for a duration of 0-3 years, it gave a hint that these newly diagnosed depression
patients might only receive short term supportive services in the community setting.
Once the symptoms of depression was stabilized, they would be discharged from the
ICCMW. The usual practice of posttreatment data collection was treatment sessions x
2, meaning 12 weeks after completion of CAT treatment in this study. However, 12
weeks postintervention (T2) data collection might not be feasible in the ICCMW
Another limitation was the dosage of treatment. As shown from the quantitative
results, alleviation of depressive signs in the BDI-II-C was 20.4 (T1) and 19.3 (T2). The
results showed that depressive signs were alleviated to the minimal form of
depression. It was speculated that longer treatment duration might benefit more
profound reduction of depressive signs and even possible improvement in cognitive
function of articulating affect in TAS-20 measurement. Future research direction will
be testing whether longer treatment duration (e.g. three months, six months or a
year) will be able to create optimal treatment effects of CAT. It is also recommended
to apply use of clay in different setting and investigate how clay work could be
related to possible reduction of suicidal tendency, improvement of quality of life, and
functioning in various aspects of life.
Despite the fact that clay work has been documented as a distinct therapeutic
medium (Sholt & Gavron, 2006), Clay is not a typical art medium; clay work requires
a specific set of techniques that differ intrinsically from those used in 2D art forms
such as drawing and painting. The firing of clay works also requires special
equipment, in particular a kiln. Installing and using a kiln in an art studio necessitates
special and frequent maintenance. The technical problems are not unresolvable;
however, promoting ceramic art in art education and art therapy training and
education programs is necessary. Furthermore, researchers must investigate the
therapeutic effects of a complete procedure of pottery art making in clinical
treatment, instead of testing effects of clay by interacting with it for a few minutes.
Art therapy educators must train art therapists who also are competent potters. By
doing so, the healing power of CAT can be utilized throughout the art therapy
4.4. Future research direction: Biomarkers for tracking physiological change
The somatosensory processes of involving hands in clay work take a bottom-up mode
of information processing (Society for Neuroscience, 2005) and are rich in haptic
perception (Elbrecht & Antcliff, 2014). The visual processes of guiding aesthetic
judgment and art making processes involve complex coordination of different cortical
regions (Liu & Miller, 2008) that take a top-down processing mode (Society for
Neuroscience, 2005). The experience, re-shaping and expression of emotion in group
sharing as well as in art making functions similarly to the process of emotional
learning in verbal form of psychotherapy (Bastos et al., 2013; Karlsson, 2012).
However, art making also involves autonomic nervous activities, hormonal responses,
and complex connecting work between the amygdala and the prefrontal cortex (PFC)
(Fuster, 2003) to process working memory, positive and negative emotional
expressions, and to execute decision making and the various work tasks of CAT (Carr,
2008; Davidson, Putnam, & Larson, 2000; Society for Neuroscience, 2005). The
various clay art making processes can coherently work together to aid regulating
affect and create psychophysiological attunement (Hinz, 2009) that enhances the
regulatory functions of the ANS (Schore, 2009).
For CAT involves both the mind and the body to intervention, assessing the impact by
more precisely measuring both psychological and physiological change in the future
research is essential. It is also one of the major limitations of this study that the
research results are self-reported measures. To measure physiological change, a
commonly measured biomarker for tracking the ANS is heart rate variability (HRV),
with lower HRV indicative of reduced parasympathetic cardiac control that has been
associated with disorders, such as sleep problems and difficulty in regulating
emotions (Evans et al., 2013). Another most often measured biomarker is cortisol, a
hormone indicative of change of stress response. Cortisol has been found effectively
changed in cancer patients who received Dance/Movement Therapy (R. T. H. Ho &
Fong, 2013) although it is scarcely used to measure effect of art therapy, to the
knowledge of the authors.
Another biomarker is the measurement of electroencephalography (EEG), or
specifically, event-related potential (ERP). The theoretical underpinning of this
research study, posits that creative art-making activities involve various complex
neurological processes. However, without actual neural activity measurements
obtained using neuroimaging techniques, this relationship between art making and
neurological processes can only be speculated about. The hypothesis behind using
ERP is that the impact of CAT on emotional and cognitive functioning should exert
corresponding changes in neurological processes. EEG can track change in brainwave
patterns in various brain regions and signals synchronization across various cortical
areas. The measurement in the changes in EEG signaling patterns can potentially
provide evidence to support the efficacy of CAT in alleviating emotional disturbance
or improving cognitive functioning such as concentration ability (Bhattacharya &
Petsche, 2002, 2005). (mention about pilot study prelim result?)
Figures and Tables
Figure 1: Multistage Process of Recruiting and Randomizing Group Participants
Table 1: Demographic and Clinical Characteristics of Participants (n = 100)
Demographic Intervention (n = 53) Control (n = 53) P*
Mean (SD) N (%) Mean (SD) N (%)
Age (years) 46.1 (10.5) 44.1 (10) .323
11 (23%) .061
Tertiary or above
9 (18.8%) .064
2 (4.2%) .068
Years diagnosed with depression
3.1 years-7 years
7.1 years-10 years
More than 10 years
17 (35.4%) .383
BDI-II-C 32.27 (9.18), 1.27 (SE) 28.77 (11.54), 1.67 (SE) .95
GHQ-12 22.47 (7.50), 1.04 (SE) 20.44 (6.40), 0.92 (SE) .15
BMSWBI_WHOLE 237.76 (69.17), 9.60 (SE) 263.14 (63.69), 9.19 (SE) .06
* TAS-20 65.34 (9.21), 1.27 (SE) 59.98 (10.59), 1.53 (SE) *.008
* Chi-squared test for categorical variable and T-test for continuous variables;
* SD, standard deviation; SE, standard error
* Abbreviations: BMS_WHOLE, the Body–Mind–Spirit–Well-Being Inventory total
Within Group Effect
CAT Group VA Group
T0 – T1 T0 – T2 T0 – T1 T0 – T2
T0 T1 T2 T0 T1 T2
M SD M SD d M SD d M SD M SD d M SD d
*** -1.1 19.3 11.8
*** -1.2 28.8 11.5 26.2 8.8 -0.3 25.7 8.3 -0.3
GHQ-12 22.5 7.5 15.6 8.5
*** -0.9 14.0 7.5
*** -1.1 20.4 6.4 18.5
6.4 -0.3 18.6 6.0 -0.3
WHOLE 237.8 69.2 284.5 79.7
*** 0.6 292.0 88.0
*** 0.7 263.1 63.7 279.4 66.4 0.3 277.9 65.6 0.2
TAS-20 65.3 9.2 63.3 8.6 -0.2 61.4 10.1
** -0.4 60.0 10.6 58.8 8.7 -0.1 59.2 11.5 -0.07
BMS_A 60.4 25.4 56.8 27.6 -0.1 52.9 27.4
* -0.3 57.7 23.2 59.0 22.5 0.06 59.7 24.5 0.08
BMS_B 39.4 13.5 47.7 17.7
*** 0.5 49.4 16.2
*** 0.7 44.8 14.4 48.0 16.4 0.21 48.0 15.7 0.21
BMS_PoA 28.1 15.1 38.2 15.2
*** 0.7 37.2 16.2
*** 0.6 31.9 15.6
* 0.4 36.7 15.7
BMS_NeA 72.0 21.9 58.7 21.9
*** -0.6 57.4 24.2
*** -0.6 68.6 19.9 61.3 16.6
* -0.4 63.9 18.2 -0.2
BMS_D_Tr 17.6 9.7 22.6 9.4
*** 0.5 22.3 10.1
*** 0.5 23.0 11.3 23.0 9.1 0.00 23.7 8.7 0.1
BMS_D_Dis 30.0 10.5 25.4 9.7
** -0.5 24.6 11.4
*** -0.5 27.3 9.2 25.0 8.9 -0.3 23.2 8.8
BMS_D_Res 15.1 7.2 17.0 6.1 0.3 18.0 5.2
** 0.5 17.1 6.1 16.6 6.0 0.00 16.4 5.2 -0.1
Abbreviations: BMS_WHOLE, Body-Mind-Spirit-Well-Being-Inventory total scores;
BMS_A, Physical Distress; BMS_B, Daily Functioning; BMS_PoA, BMS Positive Affect
Scale; BMS_NeA, BMS Negative Affect Scale; BMS_D_Tr, BMS Spirit Scale of
Tranquility; BMS_D_Dis, BMS Spirit Scale of Disorientation; BMS_D_Res, BMS Spirit
Scale of Resilience; SD – Standard deviation; d – effect size; * P < 0.05, ** P <0.01,
*** P < 0.001
Table 2: Measure Outcome Comparison of CAT and VA Groups
Table 3: Combined effect of CAT on treating depression over time
Measures Significance of MANCOVA
Multivariate outcome Univariate outcomes
Pillai’s V F (p) partial η2
F (p) partial η2
(0.000) *** 0.102
(0 .001) *** 0.076
(0 .014) ** 0.043
* P < 0.05, ** P <0.01, *** P < 0.001
* indicates significant p values
* p = .000 * p = .001
* p = .014 p = .185
Figure 2: Comparison of mean results of measures that showed significant change
over time comparing the two treatment groups, with P Values for Interaction Effect
of Group X Time.
Table 4: Correlations in CAT Group of Depressive Sign Changes with General Health,
Body–Mind–Spirit–Well-Being, Verbal Expression of Inner Process, and BMSWBI
Change of BDI-II-C ( p)
Primary & Secondary Measures T0 p T1 p T2 p
Change of GHQ-12 0.578 ** < 0.01 0.736 ** < 0.01 0.724** < 0.01
Change of BMS_WHOLE -0.670 ** < 0.01 -0.778** < 0.01 -0.825** < 0.01
Change of TAS-20 -0.298 * < 0.05 0.347* < 0.05 0.476** < 0.01
BMSWBI Subscales T0 p T1 p T2 p
BMS_A (physical distress) 0.496 ** < 0.01 0.555** < 0.01 0.723** < 0.01
BMS_B (daily functioning) -0.400 ** < 0.01 -0.507** < 0.01 -0.666** < 0.05
BMS_PoA (positive affect) -0.407 ** < 0.01 -0.695** < 0.01 -0.675** < 0.01
BMS_NeA (negative affect) 0.672 ** < 0.01 0.681** < 0.01 0.672** < 0.01
BMS_D_Tr (spirit_tranquility) -0.342 * < 0.05 -0.460** < 0.01 -0.499** < 0.01
BMS_D_Dis (spirit_disorientation) 0.346 ** < 0.01 0.673** < 0.01 0.415** <
BMS_D_Res (spirit_resilience) -0.070 > 0.05 -0.242 > 0.05 -0.323* <
Alajouanine, T. (1948). Aphasia and artistic realization. Brain, 71(17), 41.
Bar-Sela, G., Atid, L., Danos, S., Gabay, N., & Epelbaum, R. (2007). Art therapy
improved depression and influenced fatigue levels in cancer patients on
chemotherapy. Psycho-Oncology, 16(11), 980-984. doi:10.1002/pon.1175
Bastos, A. G., Guimaraes, L. S. P., & trentini, C. M. (2013). Neurocognitive changes in
depressed patients in psychodynamic psychotherapy, therapy with fluoxetine and
combination therapy. Journal of Affective Disorders., 151, 1066-1075.
Bhattacharya, J., & Petsche, H. (2002). Shadows of artistry: Cortical synchrony during
perception and imagery of visual art. Cognitive Brain Research, 13(2), 179-186.
Retrieved from http://search.proquest.com/docview/619857493?accountid=14548
Bhattacharya, J., & Petsche, H. (2005). Drawing on Mind's Canvas: Differences in
Cortical Integration Patterns Between Artists and Non-Artists. Human Brain Mapping,
26(1), 1-14. Retrieved from
Boller, F. (2005). Alajouanine's painter: Paul-Elie Gernez. In J. Bogousslavsky & F.
Boller (Eds.), Neurological disorders in famous artists (pp. 92-100). Basel: Karger.
Carr, R. (2008). Sensory processes and responses. In N. Hass-Cohen & R. Carr (Eds.),
Art therapy and clinical neuroscience. London & Philadelphia: Jessica Kingsley
Choi, Y. H., & Park, P. K. (2012). The effects of group art therapy on cognition,
depression and quality of life in elderly. International Journal of Nursing Practice,
18(Suppl. 1), 61.
Davidson, R. J., Putnam, K. M., & Larson, C. L. (2000). Dysfunction in neural circuitry
of emotion regulation - A possible prelude to violence. Science, 289, 591-594.
Ebmeier, K. P., Donaghey, C., & Steele, J. D. (2006). Recent developments and current
controversies in depression. Lancet, 367(9505), 153-167.
Elbrecht, C. (2013). Trauma healing at the clay field. . London and Philadelphia:
Kessica Kingsley Publishers.
Elbrecht, C., & Antcliff, L. R. (2014). Being touched through touch. Trauma treatment
through haptic perception at the clay field: A sensorimotor art therapy. International
Journal of Art Therapy., 19(1), 19-30.
Evans, S., Seidman, L. C., Tsao, J. C., Lung, K. C., Zeltzer, L. K., & Naliboff, B. D. (2013).
Heart rate variability as a biomarker for autonomic nervous system response
differences between children with chronic pain and healthy control children. Journal
of Pain Research, 6, 449-457.
Fuster, J. M. (2003). Cortex and mind: Unifying cognition. New York: Oxford
Gussak, D. (2009). Comparing the effectiveness of art therapy on depression and
locus of control of male and femaole inmates The Arts in Psychotherapy, 36, 202-207.
Hinz, L. D. (2009). Expressive Therapies Continuum: A framework for using art in
therapy. New York: Taylor & Francis Group.
Ho, R. T. H., & Fong, T. C. T. (2013). The associations between diurnal cortisol patterns,
self-perceived social support & sleep behavior in Chinese breast cancer patients. .
Psychoneuroendocrinology., 38(10), 2337-2342.
Ho, S. M. Y., Chan, C. L., & Ho, R. T. H. (2004). Posttraumatic growth in Chinese cancer
survivors. Psycho-Oncology, 13, 377-389.
Hong Kong Hospital Authority. (2011). Mental health service plan for adults
2010-2015. Retrieved from HK:
Karlsson, H. (2012). Psychotherapy increases the amount of serotonin receptors in
the brains of patients with major depressive disorder. In R. Levy, J. Ablon, & H.
Kächele (Eds.), Psychodynamic psychotherapy research:Evidence-based practice and
practice-based evidence. (pp. .233-238). NewYork, NY.
Karlsson, H., Hirvonen, J., Salminen, J., & Hietala, J. (2013). Increased Serotonin
Receptor 1A Binding in Major Depressive Disorder after Psychotherapy, but Not after
SSRI Pharmacotherapy, Is Related to Improved Social Functioning Capacity.
Psychotherapy and Psychosomatics, 82(4), 260-261.
Karlsson, H., Naatanen, P., & Stenman, H. (2008). Cortical activation in alexithymia as
a response to emotional stimuli. The British Journal of Psychiatry, 192(1), 32-38.
Kimport, E. R., & Robbins, S. J. (2012). Efficacy of creative clay work for reducing
negative mood: A Randomized controlled trial. Art Therapy: Journal of the American
Art Therapy Association, 29(2), 74-79. Retrieved from
Knekt, P., Lindfors, O., Sares-Jäske, L., Virtala, E., & Härkänen, T. (2013). Randomized
trial on the effectiveness of long- and short-term psychotherapy on psychiatric
symptoms and working ability during a 5-year follow-up. Nordic Journal Of Psychiatry,
67(1), 59-68. doi:10.3109/08039488.2012.680910
Kronholm, E., Partonen, T., Salminen, J. K., Mattila, A. K., & Joukamaa, M. (2007).
Alexithymia, Depression and Sleep Disturbance Symptoms. Psychotherapy and
Psychosomatics, 77(1), 63-65. Retrieved from
Larsen, J. K., Brand, N., Bermond, B., & Hijman, R. (2003). Cognitive and emotional
characteristics of alexithymia: A review of neurobiological studies. Journal of
Psychosomatic Research, 54(6), 533-541. doi:10.1016/0022-3999(94)90006-X
Lee, S., Tsang, A., & Kwok, K. (2007). Twelve-month prevalence, correlates, and
treatment preference of adults with DSM-IV major depressive episode in Hong Kong.
Journal of Affective Disorders, 98(1-2), 129-136.
Liu, A., & Miller, B. (2008). Visual art and brain Handbook of clinical neurology. SFO,
USA: Memory and AgingCenter, University of California.
Lusebrink, V. B. (1990). Imagery and visual expression in therapy. New York: Plenum
Lusebrink, V. B. (2004). Art therapy and the brain: An attempt to understand the
underlying processes of art expression in therapy. . Art Therapy: Journal of the
American Art Therapy Association., 21(3), 125-135.
McCrone, P., Dhanasiri, S., Patel, A., Knapp, M., & Lawton-Smith, S. (2008). Paying the
Price: The cost of mental health care in England to 2026. Retrieved from UK:
Nan, J. K. M., & Ho, R. T. H. (2014). Affect regulation and treatment for depression
and anxiety through art: Theoretical ground and clinical issues. Annals of Depression
and Anxiety, 1(2), 1-6.
National Institute of Mental Health of the United States. (2013). What medications
are used to treat depression? Retrieved from
Ng, S. M., Yau, J. K. Y., Chan, C. L. W., Chan, C. H. Y., & Ho, D. Y. F. (2005). The
Measurement of Body-Mind-Spirit Well-Being: Toward Multidimensionality and
Transcultural Applicability. Social Work in Health Care, 41(1), 33-52. Retrieved from
Panksepp, J. (2009). Brain emotion systems and qualities of mental life - From animal
models of affect to implications for psychotherapeutics. In D. Fosha, D. J. Siegal, & M.
F. Solomon (Eds.), The healing power of emotion: Affective neuroscience,
development, and clinical practice. New York/London: W. W. Norton & Company.
Puig, A., Lee, S. M., Goodwin, L., & Sherrard, P. (2006). The efficacy of creative arts
therapies to enhance emotional expression, spirituality, and psychological well-being
of newly diagnosed Stage I and Stage II breast cancer patients: A preliminary study.
The Arts in Psychotherapy, 33(3), 218-228. Retrieved from
Sato, S., & Yeh, T. L. (2013). Challenges in Treating Patients with Major Depressive
Disorder: The Impact of Biological and Social Factors. CNS Drugs, 27(1), S5-10.
Schore, A. N. (2002). Dysregulation of the right brain: A fundamental mechanism of
traumatic attachment and the psychopathogenesis of posttraumatic stress disorder.
Australian and New Zealand Journal of Psychiatry, 36, 9-30.
Schore, A. N. (2003a). Affect dysregulation and disorders of the self. . New York:
Schore, A. N. (2003b). Affect regulation and the repair of the self. New York: Norton.
Schore, A. N. (2009). Right-brain affect regulation - An essential mechanism of
development, trauma, dissociation, and psychotherapy. In D. Fosha, D. J. Siegal, & M.
F. Solomon (Eds.), The healing power of emotion: Affective neuroscience,
development & clinical practice. New York/London: W. W. Norton & Company.
Schutz, L. E. (2005). Broad-perspective perceptual disorder of the right hemisphere.
Neuropsychology Review, 15, 11-27.
Sherwood, P. (2004). The Healing Art of Clay Therapy. Melbourne, Victoria: ACER
Sholt, M., & Gavron, T. (2006). Therapeutic qualities of clay-work in art therapy and
psychotherapy: A Review. Art Therapy, 23(2), 66-72. Retrieved from
Society for Neuroscience. (2005). Brain facts - A primer on the brain and nervous
system. Retrieved from www.sfn.org
Thase, M. (2009). Psychotherapies for depression in adults: A review of recent
development. In H. Herrman, M. Maj, & N. Sartorius (Eds.), Depressive disorders (3rd
ed.). UK: WPA Series in Evidence & Experience in Psychiatry.
Trivedi, M. H., & Greer, T. L. (2014). Cognitive dysfunction in unipolar depression:
Implications. Journal of Affective Disorders., 152-154, 19-27.
Vázquez, G. H., Holtzman, J. N., Tondo, L., & Baldessarini, R. J. (2015). Efficacy and
tolerability of treatments for bipolar depression. Journal of Affective Disorders, 183,
van Geffen, E. C. G., van der Wal, S. W., van Hulten, R., de Groot, M. C. H., Egberts, A.
C. G., & Heerdink, E. R. (2007). Evaluation of patients' experiences with
antidepressants reported by means of a medicine reporting system. European
Journal of Clinical Pharmacotherapy, 63, 1193-1199.
Vessel, E. A., Starr, G. G., & Rubin, N. (2012). The brain on art: Intense aesthetic
experience activates the default mode network. Frontiers in Human Neuroscience,
World Health Organisation. (2008). The global burden of disease: 2004 update,
World Health Organization. (2012). Depression: A global crisis. World Mental Health
Day, October 10 2012. Retrieved from: