The Arts in Psychotherapy
The Arts in Psychotherapy
The Arts in Psychotherapy
a r t i c l e i n f o a b s t r a c t
Article history: COPE cards are 88 therapeutic associative cards (7 cm × 10 cm) with evocative art works, which were
Available online 24 April 2015 utilized in the last two decades both in clinical and community settings. Although some evidence exists
regarding the impact of cards integration in psychotherapy and experimental settings, it is unknown
Keywords: whether the therapeutic cards can induce distinct affective states in the observer. The aim of the study was
COPE cards to establish affective ratings of the COPE cards. Participants of the study (N = 100) were undergraduates
Self-assessment manikin
(71% females) who were exposed to each of the cards for 6 s, and were asked to rate their affective state
Affective dimensions
according to the Self-Assessment Manikin (SAM) rating system. Results indicated that the cards produced
significantly different affective states among the participants. Arousal induced by the image negatively
correlated with image’ valence (r = −0.71, p < 0.001) and dominance (r = −0.73, p < 0.001), while image’
valence and dominance were highly correlated (r = 95, p < 0.001). Furthermore, cluster analysis results
suggest potential subdivisions of the COPE cards, with practical implications for practitioners. This study is
the first to present the affective states evoked by therapeutic cards. Practical implications for therapeutic
cards integration in clinical and experimental settings are suggested and discussed.
© 2015 Elsevier Ltd. All rights reserved.
Introduction generated following exposure to images which evoke an emotional response (Lahad
et al., 2010).
Projective tests based on viewing images on cards, such as Rorschach or TAT,
have been used in psychology for decades. Although the statistical validity of such
tests has been debated projective tests are used widely incontemporary personality Emotional stimuli
assessment (Gawronski, 2009). Over the last decade growing interest and appli-
cation of these cards in therapy and education can be observed (Gagnon, 2001;
Lang and colleagues (Center for the Study of Emotion &
Hahs, 2008; Lahad, Farhi, Leykin, & Kaplansky, 2010; Lahad, Shacham, & Ayalon,
2012). Cards of association, or OH cards, are of specific interest for therapists (Abel
Attention, 1995; Lang, Bradley, & Cuthbert, 1995, 1997, 1999,
& Friedman, 2009; Ayalon, 2003, 2004, 2006a, 2006b, 2007). When the cards are 2001; Lang & Greenwald, 1981; Lang, Öhman, & Vaitl, 1988) devel-
employed as a therapeutic method or tool there is a requirement to ensure that oped the international affective picture system (IAPS), a standard
statistically supported relevance as to their specific use is available. norm setting of affect evoking images. Currently, the collection
Imagination, feelings and thoughts all help generate internal representations
is comprised of more than 700 colorful images, and each is stan-
of images to which we are exposed. Specifically, such internal representations are
dardized according to the following dimensions: arousal, valence
and dominance. This unique collection offers researchers a com-
mon norm to follow, a standard that promotes better scientific
understanding of important mental health issues, such as tempo-
∗ Corresponding author. Tel.: +972 4 6900600; fax: +972 4 6950740. rary insight regarding emotional development disorders, assessing
E-mail address: ran.m.cohen@gmail.com (R. Cohen). physiological impact of different stress situations, evaluation of
http://dx.doi.org/10.1016/j.aip.2015.04.004
0197-4556/© 2015 Elsevier Ltd. All rights reserved.
R. Cohen et al. / The Arts in Psychotherapy 45 (2015) 36–46 37
emotionally pathological conditions and determining the levels of protocol SEE FAR CBT where as part of the procedure the therapist
emotional damage occurred in brain-damaged patients. To create expose the client to positive cards in order for the client to create
a better standardization it is always important to enable repro- an Imaginal safe place. So far this is done intuitively. Finally, this
ducibility and expand earlier studies of emotional evoking images study has significant relevance to the field of art therapy, by relating
(Libkuman, Otani, Kern, Viger, & Novak, 2007). Indeed, the advan- artistic drawings to the affective subjective experience (Masterson
tage of the IAPS lies in subsequent repetitions (Bradley & Lang, et al., 2008).
2007).
The three affective dimensions of the IAPS are based upon Method
a dimensional theory of semantic meaning (Osgood, Suci, &
Tannenbaum, 1957). Factor analysis has revealed that most of the Participants
variance in semantic assessment was explained by a single factor,
hedonistic valence. This factor ranges from unpleasant (sad, frus- One hundred participants, 29 men and 71 women (N = 100),
trated, desperate etc.) to pleasant (happy, satisfied, hopeful etc.). ranging between 15 and 44 years of age (M = 27.0, SD = 4.67). 43%
The second factor, which also explained some of the variance was of the participants were first year psychology students at Tel-Hai
termed arousal. It reflects the active component of emotional expe- academic college in Upper Galilee, Israel, while the rest of the sam-
rience, ranging from low levels (calm, relaxed, drowsy etc.) to high ple included volunteers who were recruited using social media and
levels (excited, stimulated, highly alert etc.). The third and last fac- personal mailing lists. All participants were Hebrew fluent speak-
tor, still explaining some of the variance but not considerably, was ers, 94% were Israeli Jews, 85% were secular, 80% were born in
termed dominance; and it refers to subjective feelings of mastery, Israel, 78% were not married and 62% had a below average fam-
competence, power or skill (Russell, Ward, & Pratt, 1981), as well as ily income level. The student participants received course credit
to the levels of control the subject has over situation, ranging from for their attendance, while others enrolled a lottery for a 250 NIS
low levels to high levels. gift card for their participation.
The SAM has also been used to assess emotional reactivity during We tested Pearson correlations between the means of the card
exposure-based therapy (Sloan & Marx, 2006). And internal con- ratings for each of the three different emotional dimensions. Neg-
sistency reliability was found to be similar for younger and older ative significant correlations were found between arousal and
adults, 0.63 and 0.82 for valence, respectively (Backs, da Silva, & valence, r(88) = −0.71, p < 0.001, and between arousal and domi-
Han, 2005). Arousal scores were 0.98 for both groups. nance, r(88) = −0.73, p < 0.001; and a positive significant correlation
was found between dominance and valence r(88) = 0.95, p < 0.001.
Procedure
Cluster analysis
The entire research procedure was conducted online using
Qualtrics research suite (www.qualtrics.com). Some of partici- The first analysis aimed to select the optimal model accord-
pants have received a link via email to answer the questionnaire ing to Bayesian information criterion (BIC). The Gaussian finite
online, or were invited to complete the same questionnaire in the mixture model fitted by Expectation-Maximization (EM) was an
lab. Furthermore, the link was distributed in a snowball sampling EEV (ellipsoidal, equal volume and shape) model with two com-
manner via social media websites, such as Facebook, to enhance ponents, one contains 45 cards and the second cluster contains
survey attendance. The first screen of the electronic question- 43 cards (see Figs. 2 and 3 for model-based clustering plots), log
naire included basic demographic information and brief personality likelihood = −179.70, BIC = −431.03.
survey (not described in the present study). The second screen According to the plots, we can notice the two clusters them-
introduced the affective rating procedure and included throughout selves across the different combinations of the affective dimensions
explanation of the SAM rating system. Lang et al. (1999) instruc- (valence vs. dominance, valence vs. arousal, arousal vs. dominance).
tions for the affective ratings were modified and translated from The agglomerative hierarchical clustering (described in Fig. 4)
English to Hebrew (for full procedure instructions, see Lang et al., clearly shows two main clusters in the data, although it also shows
1999). Next, participants were trained on the process using random that more clustering solutions are available (i.e. 3 cluster and 5 clus-
artistic image obtained from the Internet and verified the com- ter solution). A summary of cards according to their cluster and
prehension of the task. Following brief introduction, participants affective dimensions appears in Table 1.
rated dimensions of arousal, valence and dominance for each of The optimal number of clusters estimated by optimum average
the 88 COPE cards. The 88 cards were divided into four equal series, silhouette width is 2 and the average silhouette width (i.e. silhou-
while a one minute break was given between each series to prevent ette coefficient) was found to be 0.54, which indicates a reasonable
exhaustion. During the task itself, each card was displayed for 6 s structure of the data.
and followed by the affective rating using SAM. The total duration
of the procedure was about 40 min in total for all participants.
Further analysis
Table 1
Summary of cards according to their cluster and affective dimensions.
38, 76, 57, 85, 17, 45, 6, 44 Celebration, happiness and love, depicted by humans Medium High High
and physical objects
49, 77, 67, 35, 66, 33, 42, 74, 62, 64 Landscape is described, including skies, sunsets, if Low High High
people are present they are surrounded by nature
43, 88, 70, 24, 51, 12, 61, 81, 14, 5, 29, 32, 79, 83 Ambivalent scenes of humans and animals Medium Medium Medium
48, 53, 80, 9, 15, 40 States of helplessness, sorrow, grief and immense High Low Low
danger
7, 2, 59, 84, 1, 60, 65, 54, 72, 78, 58, 19, 56 Combination of people and natural scenes that Low Medium Medium
produce ambivalent feelings toward the picture (or
character/s if relevant)
46, 82, 71, 3, 50, 52, 86, 31, 28, 55, 68 Ambiguous descriptions of humans, body parts and Medium Mid-Low Medium
natural scenes
22, 16, 21, 27, 30 Humans at natural scenes during a journey or walk Medium Mid-Low Medium
8, 73, 41, 75, 20, 69, 63, 87 Distressing and negative feelings inducing images of High Low Low
humans in negative mood or anxiety inducing images
13, 23, 37, 39, 4, 34, 26, 47, 10, 25, 36, 11, 18 Chaotic and fearful scenes of destruction and/or High Low Low
disaster.
Fig. 1. Examples of COPE cards evoking significant emotional responses. From right to left: COPE48 resulted in high arousal (t(99) = 8.54, p < 0.001), low valence (t(99) = −18.08,
p < 0.001) and low dominance (t(99) = −8.31, p < 0.001); COPE80 resulted in high arousal (t(99) = 6.56, p < 0.001), low valence (t(99) = −10.62, p < 0.001) and low dominance
(t(99) = −7.51, p < 0.001); COPE77 resulted in low arousal (t(99) = −5.12, p < 0.001), high valence (t(99) = 10.38, p < 0.001) and high dominance (t(99) = 6.93, p < 0.001); COPE45
resulted in low arousal (t(99) = −3.65, p < 0.001), high valence (t(99) = 16.34, p < 0.001) and high dominance (t(99) = 8.49, p < 0.001). Note: COPE cards numbers appear in small black
printing below each card.
between dominance and valence r(88) = 0.89, p < 0.001. For women, have been suggested before in second and third generation of Holo-
a significant negative correlation was found between arousal caust survivors (Scharf, 2007). Indeed, 94% of the participants in
and valence r(88) = −0.70, p < 0.001 and between dominance and this research were Israeli Jews. Hence, the emotional responses of
arousal r(88) = −0.70, p < 0.001, also, a significant positive correla- high arousal and low valence can represent anxiety and rejection,
tion was between dominance and valence r(88) = 0.96, p < 0.001.
Discussion
Fig. 3. Classification plot of the 2 clusters according to the three affective dimensions of arousal, valence and dominance.
respectively; and the low dominance might represent a feeling of that tend to rate unpleasant events as arousing. This is consis-
helplessness. COPE card no. 80, portraying a boat in a stormy sea, tent with the nature of the measurement of arousal (low vs. high
evoked similar responses. Once again the image ratings might rep- arousal) compared to the nature of valence measurement.
resent feelings of anxiety, helplessness and instability. For the present dataset we used up to date methods for clus-
COPE cards 77 and 45, show a different pattern. COPE card no. ter analysis to explore sub-groups in our data. One major question
77 portrays a man sitting in front of soaring mountains (Fig. 1), this that remains a great topic for research is the optimal number of
may express expectations and aspirations that fill a man with calm- clusters one should find in the data (Mirkin, 2011). Although vari-
ness and quietness (low arousal) and with happiness and optimism ous methods exist to solve this question, sometimes the decision is
(high valence), that in turn would facilitate a feeling of control (high subjective and depends on the author’s opinion. For practical rea-
dominance). Similarly, COPE card no. 45 portrays a pleasant situa- sons researchers would like to partition the data into 3 or 5 groups,
tion (Fig. 1) of a couple and a pastoral background, which may cause depends on the specific needs. For instance, a practitioner would
expressions of harmony, equality and intimacy. like to subset five types of cards based on a specific affective fea-
Our findings are supported by Osgood et al. (1957). Their ture, like dominance, so later it will be applicable to conduct an
research of semantic meaning, involving the emotional dimension, activity that should sequentially impact dominance levels. Other
indicated that emotional degree develops as a result of changes practitioners might choose an alternative strategy and would pre-
in basic dimensions of arousal and pleasantness. Our findings also fer to sub-divide the data into two broad categories (low arousal
align with Ayalon’s (2007) argument, associative cards stimulate vs. high arousal, or low dominance vs. high dominance). The subdi-
the imagination as they symbolize previous traumatic events. vision may also rely on the content of the cards. In the dendrogram
In line with our other findings, strong significant correlations presented in this work (Fig. 4), if pruned at three-level clusters,
were found between the three emotional dimensions. The strong by observing the leaves (at the bottom – center of the image)
negative correlation between valence and arousal might stem from we can see a cluster which contains cards 48, 53, 80, 9 and 5
either a feeling of anxiety (high arousal, low valence) or a feeling (see Appendices A & C for description of the cards and affective
of calmness and pleasure (low arousal, high valence). This expla- descriptive statistics). These cards are all highly arousing, strongly
nations also agrees with the strong negative correlation found and negatively valenced and relatively low in dominance (i.e. pro-
between arousal and dominance and to the strong positive cor- duce low levels of sense of control). These cards include images
relation between dominance and valence. Osgood et al. (1957) of individuals in submissive and not in control states. If the den-
also found similar correlations between valence and dominance. drogram is pruned at five-level clusters, we can inspect an eight
Bradley and Lang (2007) provide an explanation for these findings; cards cluster which contains the cards: 38, 76, 57, 85, 17, 45, 6 and
they show that the level of dominance displays low variance for 44. Except card 44 (which depicts colorful balloons) all other cards
the SAM scale ratings, compared with dimensions of arousal and show humans (or human and animals) and arms holding a red tra-
valence. Although, for images of a more symbolic and less social ditional heart shape. As these cards are positively valenced they can
nature better variance was measured. We saw that the situations be used as a calming and relaxing stimuli during psychotherapeu-
derived from the COPE cards deal with symbolic emotional rep- tic session. For instance, the PTSD treatment protocol SEE FAR CBT
resentations, as expected from therapeutic associative cards (Lang (Lahad et al., 2010), utilizes the use of “safe place” concept, a men-
et al., 1999). Hence, this provides an explanation for the strong cor- tal and physiological state characterized by feelings of control and
relation found between the dimensions of dominance and valence relaxation. Although the establishment of a “safe place” using ther-
in the current study. apeutic cards involves the individual selection of cards (so even
A negative linear correlation between arousal and valence, sim- not so pleasant cards can serve as safe place stimuli), the thera-
ilar to the one found in the current study, is referred to by Bradley pist can direct the client in choosing specific cards based on tested
and Lang (2007) as negative bias; and its characteristic of people estimation of the affective experience these cards can induce.
R. Cohen et al. / The Arts in Psychotherapy 45 (2015) 36–46 41
6
4
2
0
38
76
57
85
17
45
6
44
49
77
67
35
66
33
42
74
62
64
43
88
70
24
51
12
61
81
14
5
29
32
79
83
48
53
80
9
15
40
7
2
59
84
1
60
65
54
72
78
58
19
56
46
82
71
3
50
52
86
31
28
55
68
22
16
21
27
30
8
73
41
75
20
69
63
87
13
23
37
39
4
34
26
47
10
25
36
11
18
Fig. 4. Hierarchical clustering of COPE cards.
In further analysis, we sought to measure gender differences The current study has theoretical and practical implications.
in card’s ratings; based on the assumption that women tend to First, this study’s innovative attitude toward COPE cards allows for
report higher stress levels than men (Kimhi, Eshel, Zysberg, & a reliable and valid application for researchers and therapists alike.
Hantman, 2009). This tendency led us to the hypothesis that sig- Second, it expands on the IAPS collection. Furthermore, it forms a
nificant differences will be found in the emotional dimensions basis for future research, to set more accurate standards and norms;
ratings between the genders. However, we did not find a signif- these may deepen our scientific understanding of issues related to
icant effect. Thus, perhaps there is no significant difference. We mental health (Bradley & Lang, 2007).
did find a significant effect for the type of card in relation to the
emotional dimensions. Therefore, one might conclude that cer- Limitations
tain cards are able to elicit significantly stronger scores of valence,
arousal and/or dominance as opposed to others. For the dimensions The current study has a few notable limitations. First, many
of valence and dominance a significant interaction effect was found of participants answered the survey via a link that appeared on
between card type and gender. That is, certain types of cards tend social media sites, this limited supervision of careless answers by
to heighten the sense of pleasantness and control among each gen- participants. However, we tried to minimize careless responses by
der separately, while other cards do not have similar effects for any utilizing dynamic and interactive environment that clearly directs
gender. the participants and also provides short pause interval for rest.
COPE card no. 45, which had a strong effect on all three dimen- Second, many of the participants claimed that they were not able
sions of emotion amongst the participants; also showed higher to keep their concentration during the entire survey (average
feelings of control and pleasantness for women as opposed to time spent was 40 min), which in turn might also cause careless
men. However, no significant difference was found for arousal answers. Third, there was a significant difference in group size
in this case. The situation in the card (as seen in Fig. 1) might between the genders, 29 & 71, for men and women, respectively.
remind a romantic situation; causing more feelings of warmth This might have been cause for uneven distributions and bias.
and pleasure in women than in men, which might lead to higher Forth, the study involved mainly Jewish students, which limits it
feelings of empowerment and efficacy for women (Chodorow, generalizability.
1974; Unger, 2001). As mentioned, a strong positive correlation Future research should include: implementing physiological
was found between the dimensions of valence and dominance for measures in order to enhance validity of subjective affective
all of the participants and within each gender. Bradley & Lang’s responses, segmentation and time intervals to prevent careless
(2007) argument which states that for pleasant images there will answers, and adding qualitative measures to validate subjective
be a significantly stronger positive correlation for arousal and reactions. Future research should further explore the suggested
valence ratings for men than women, might explain why a sig- interaction between gender and card content. A previous study has
nificant effect was not found of both gender and the interaction shown that, in a series of experiments, more automatic in-group
between gender and card type on the arousal dimension. The cur- bias for women compared to men (Rudman & Goodwin, 2004),
rent study showed that images of pleasant situations, which in thus, gender might play a role in participant’s subjective emotional
turn enhance the feeling of control (for example, COPE card no. response to certain cards. A more thorough examination of the
45), did not significantly enhance the level of arousal amongst images in question, specifically, of gender might provide a better
women. explanation for the above mentioned interaction.
42 R. Cohen et al. / The Arts in Psychotherapy 45 (2015) 36–46
ID Description ID Description
M SD M SD M SD
Appendix B (Continued)
Arousal Valence Dominance Cluster Solution
M SD M SD M SD
Appendix C. Results of one-sample t-test of COPE cards, for the three emotional dimensions and the obtained effect size
COPE01 4.78 2.08 −1.06 0.11 5.13 1.91 0.68 0.07 5.48 2.09 2.30 0.23
COPE02 4.35 1.98 −3.28** 0.33 4.96 1.78 −0.23 0.02 5 1.91 0.00 0.00
COPE03 4.82 2.00 −0.90 0.09 5.11 1.80 0.61 0.06 4.94 2.18 −0.27 0.03
COPE04 5.84 1.92 4.37*** 0.44 3.57 1.44 −9.90*** 0.99 3.92 2.00 −5.39*** 0.54
COPE05 4.57 2.10 −2.05** 0.20 5.8 1.74 4.59*** 0.46 5.53 1.89 2.79** 0.28
COPE06 4.8 2.40 −0.83 0.08 7.46 1.59 15.51*** 1.55 6.74 2.02 8.62*** 0.86
COPE07 4.06 2.01 −4.68*** 0.47 5.06 1.76 0.34 0.03 5.31 2.05 1.51 0.15
COPE08 5.59 2.06 2.87** 0.29 4.36 1.88 −3.40** 0.34 4.31 2.29 −3.01** 0.30
COPE09 5.66 2.14 3.08** 0.31 2.91 1.69 −12.38*** 1.24 3.75 2.09 −5.99*** 0.60
COPE10 5.31 2.01 1.54 0.15 3.46 1.67 −9.21*** 0.92 4.05 2.00 −4.76*** 0.48
COPE11 5.6 2.07 2.90** 0.29 3.75 1.77 −7.08*** 0.71 4.22 1.96 −3.99*** 0.40
COPE12 4.21 2.03 −3.89*** 0.39 5.45 1.58 2.85** 0.28 6 1.85 5.39*** 0.54
COPE13 5.7 1.98 3.53** 0.35 3.93 1.86 −5.77*** 0.58 4.5 1.94 −2.58* 0.26
COPE14 4.35 1.98 −3.28** 0.33 5.82 1.68 4.86*** 0.49 5.44 1.84 2.39* 0.24
COPE15 6.18 2.11 5.59*** 0.56 2.83 1.83 −11.88*** 1.19 3.68 2.16 −6.13*** 0.61
COPE16 4.77 1.90 −1.21 0.12 3.88 1.73 −6.47*** 0.65 4.28 1.94 −3.71*** 0.37
COPE17 4.92 2.54 −0.32 0.03 7.78 1.45 19.21*** 1.92 7.02 1.84 10.96*** 1.10
COPE18 5.43 2.25 1.91 0.19 3.76 1.72 −7.19*** 0.72 4.14 1.89 −4.55*** 0.46
COPE19 4.33 2.08 −3.22*** 0.32 5.4 1.68 2.38* 0.24 5.26 1.77 1.47 0.15
COPE20 5.25 1.79 1.39 0.14 3.93 1.58 −6.75*** 0.68 4.64 1.89 −1.90 0.19
COPE21 4.91 2.03 −0.44 0.04 3.78 1.61 −7.60*** 0.76 4.23 2.17 −3.55** 0.35
COPE22 4.23 1.98 −3.88*** 0.39 4.05 1.27 −7.46*** 0.75 4.9 1.55 −0.64 0.06
COPE23 6.03 1.80 5.74*** 0.57 3.75 1.62 −7.73*** 0.77 4.4 2.21 −2.72** 0.27
COPE24 4.88 1.93 −0.62 0.06 6.37 1.66 8.24*** 0.83 5.88 1.88 4.68*** 0.47
COPE25 5.45 2.24 2.01* 0.20 3.64 1.51 −9.02*** 0.90 3.99 1.87 −5.41*** 0.54
COPE26 5.51 2.08 2.46* 0.25 3.4 1.61 −9.91*** 0.99 3.85 1.80 −6.39*** 0.64
COPE27 4.93 2.00 −0.35 0.04 4.14 1.66 −5.19*** 0.52 4.38 1.83 −3.39** 0.34
COPE28 4.51 2.05 −2.39* 0.24 4.63 1.65 −2.24* 0.22 4.8 1.76 −1.13 0.11
COPE29 4.54 1.94 −2.37* 0.24 5.77 1.80 4.28*** 0.43 5.6 1.98 3.00** 0.30
COPE30 5.15 2.05 0.73 0.07 3.94 1.68 −6.31*** 0.63 4.28 1.97 −3.66*** 0.37
COPE31 4.41 1.99 −2.97** 0.30 4.6 1.86 −2.15* 0.22 4.68 1.91 −1.67 0.17
COPE32 4.1 2.04 −4.41*** 0.44 5.99 1.61 6.16*** 0.61 5.68 1.70 4.00*** 0.40
COPE33 4.15 2.19 −3.87*** 0.39 6.74 1.87 9.31*** 0.93 6.38 1.99 6.93*** 0.69
COPE34 5.85 1.94 4.38*** 0.44 3.55 1.73 −8.38*** 0.84 4.05 1.98 −4.81*** 0.48
COPE35 4.26 2.18 −3.40** 0.34 6.65 1.59 10.40*** 1.04 5.88 1.97 4.46*** 0.45
COPE36 5.58 1.85 3.14** 0.31 3.55 1.57 −9.22*** 0.92 4.28 1.81 −3.98*** 0.40
COPE37 5.78 1.83 4.25*** 0.43 3.28 1.79 −9.59*** 0.96 4.16 2.08 −4.03*** 0.40
COPE38 4.63 2.32 −1.59*** 0.16 6.95 1.93 10.13*** 1.01 6.59 1.90 8.38*** 0.84
COPE39 5.81 2.10 3.85*** 0.39 3.41 1.53 −10.43*** 1.04 4.4 2.03 −2.96** 0.30
COPE40 5.99 1.97 5.02*** 0.50 2.99 1.57 −12.83*** 1.28 3.9 2.06 −5.35*** 0.53
COPE41 5.84 2.06 4.07*** 0.41 4.21 2.06 −3.83*** 0.38 4.65 2.32 −1.51 0.15
COPE42 4.29 2.27 −3.13** 0.31 6.98 1.97 10.05*** 1.01 6.37 2.12 6.45*** 0.65
COPE43 4.48 2.01 −2.59* 0.26 5.97 1.88 5.16*** 0.52 5.99 1.97 5.02*** 0.50
COPE44 4.7 2.42 −1.24 0.12 7.54 1.62 15.70*** 1.57 6.33 2.17 6.12*** 0.61
COPE45 4.17 2.27 −3.65*** 0.37 7.39 1.46 16.34*** 1.64 6.61 1.90 8.49*** 0.85
COPE46 4.82 2.17 −0.83 0.08 5.44 2.11 2.08* 0.21 4.6 2.29 −1.75 0.17
COPE47 5.63 1.99 3.17** 0.32 3.61 1.78 −7.81*** 0.78 3.76 2.16 −5.74*** 0.57
COPE48 6.69 1.98 8.54*** 0.85 2.37 1.45 −18.09*** 1.81 3.12 2.26 −8.31*** 0.83
COPE49 3.76 2.24 −5.54*** 0.55 6.55 1.54 10.06**** 1.01 5.91 2.04 4.45*** 0.45
COPE50 4.86 1.98 −0.71 0.07 5.15 1.48 1.01 0.10 5.06 1.68 0.36 0.04
COPE51 4.69 2.21 −1.40 0.14 6.37 1.87 7.33*** 0.73 5.94 1.92 4.89*** 0.49
R. Cohen et al. / The Arts in Psychotherapy 45 (2015) 36–46 45
Appendix C (Continued)
Arousal Valence Dominance
COPE52 5.17 1.96 0.87 0.09 4.76 1.91 −1.25 0.13 5.08 2.10 0.38 0.04
COPE53 6.07 1.95 5.49*** 0.55 3.22 1.47 −12.13*** 1.21 3.6 1.83 −7.67*** 0.77
COPE54 4.46 1.96 −2.76** 0.28 5.38 1.27 2.99** 0.30 5.66 1.62 4.06*** 0.41
COPE55 4.67 1.84 −1.80 0.18 4.53 1.43 −3.28** 0.33 4.66 1.63 −2.09* 0.21
COPE56 4.33 2.01 −3.34** 0.33 5.33 1.79 1.84 0.18 5.08 1.88 0.43 0.04
COPE57 5.21 2.18 0.96 0.10 6.76 1.87 9.43*** 0.94 6.29 1.96 6.59*** 0.66
COPE58 4.32 2.14 −3.18** 0.32 5.65 1.92 3.38** 0.34 5.18 1.97 0.92 0.09
COPE59 4.38 1.88 −3.29** 0.33 4.77 1.59 −1.45 0.14 5.17 1.94 0.87 0.09
COPE60 4.74 1.97 −1.32 0.13 4.92 1.70 −0.47 0.05 5.36 1.95 1.84 0.18
COPE61 3.9 2.00 −5.51*** 0.55 4.65 1.51 4.31*** 0.23 5.98 1.82 5.40*** 0.54
COPE62 4.59 2.35 −1.74 0.17 6.93 1.68 11.50*** 1.15 6.23 2.19 5.61*** 0.56
COPE63 5.14 2.05 0.68 0.07 4.3 1.67 −4.20*** 0.42 4.59 1.97 −2.08* 0.21
COPE64 4.3 2.21 −3.16** 0.32 7.11 1.78 11.85*** 1.19 6.12 2.09 5.35*** 0.54
COPE65 4.82 2.13 −0.85 0.08 5.51 1.78 2.86** 0.29 5.73 2.07 3.52** 0.35
COPE66 4.27 2.17 −3.37** 0.34 6.46 1.65 8.82*** 0.88 5.78 1.92 4.07*** 0.41
COPE67 4.36 2.17 −2.95** 0.29 6.44 1.83 7.85*** 0.79 6.16 2.09 5.54*** 0.56
COPE68 4.64 1.89 −1.90 0.19 4.47 1.36 −3.90*** 0.39 4.92 1.73 −0.46 0.05
COPE69 5.16 1.83 0.87 0.09 4.33 1.65 −4.06*** 0.41 4.9 1.90 −0.53 0.05
COPE70 4.73 1.98 −1.36 0.14 6.18 1.57 7.53*** 0.75 6.19 2.03 5.85*** 0.59
COPE71 4.66 2.14 −1.59 0.16 4.97 1.62 −0.19 0.02 4.78 2.37 −0.93 0.09
COPE72 4.4 1.99 −3.02** 0.30 5.42 1.68 2.49* 0.25 5.54 1.63 3.31** 0.33
COPE73 5.59 1.91 3.08** 0.31 4.24 1.79 −4.24*** 0.42 4.62 1.82 −2.09* 0.21
COPE74 4.16 2.24 −3.75*** 0.38 7 1.66 12.02*** 1.20 6.37 2.05 6.68*** 0.67
COPE75 5.75 1.95 3.84*** 0.38 4.3 2.25 −3.11** 0.31 4.66 2.35 −1.45 0.14
COPE76 4.81 2.07 −0.92 0.09 7 1.59 12.58*** 1.26 6.55 1.77 8.74*** 0.88
COPE77 3.84 2.26 −5.12*** 0.51 6.77 1.71 10.38*** 1.04 6.48 2.13 6.93*** 0.69
COPE78 4.79 1.98 −1.06 0.11 5.52 1.76 2.96** 0.30 5.25 2.14 1.17 0.12
COPE79 4.03 1.92 −5.07*** 0.51 5.66 1.34 4.94*** 0.49 5.53 1.72 3.08** 0.31
COPE80 6.26 1.92 6.56*** 0.66 3.29 1.61 −10.62*** 1.06 3.44 2.08 −7.51*** 0.75
COPE81 4.06 1.94 −4.84*** 0.48 5.81 1.44 5.62*** 0.56 5.91 1.65 5.50*** 0.55
COPE82 4.64 1.88 −1.76 0.19 5.44 1.50 2.93** 0.29 4.78 2.16 −1.02 0.10
COPE83 4.04 2.14 −4.48*** 0.45 5.69 1.38 5.01*** 0.50 5.63 1.95 3.22** 0.32
COPE84 4.43 1.83 −3.11** 0.31 4.77 1.19 −1.94 0.19 5.69 1.79 3.86*** 0.39
COPE85 5.09 2.49 0.36 0.04 7.12 1.79 11.85*** 1.18 6.6 2.17 7.37*** 0.74
COPE86 4.91 2.02 −0.45 0.04 4.73 1.43 −1.89 0.19 5.31 1.90 1.64 0.16
COPE87 5.39 1.89 2.06* 0.21 4.46 1.51 −3.58** 0.36 4.66 2.06 −1.65 0.17
COPE88 4.73 2.03 −1.33 0.13 5.79 1.83 4.31*** 0.43 5.93 1.98 4.69*** 0.47
*p < 0.05 **p < 0.01 ***p < 0.001. The values of the emotional dimensions were compared aga1inst the center value 5. Estimates of effect size (Cohen’s d) for the difference
between the mean value of 5 to the attained affective means of each cards obtained by using a formula suggested by Kotrlik and Williams (2003), and interpreted after Cohen
(1988) as d > 0.20 = small effect size, d > 0.50 = medium effect size and d > 0.80 = large effect size.
References Chodorow, N. (1974). Family structure and feminine personality. In M. Z. Rosaldo, &
L. Lamphere (Eds.), Women, culture and society (pp. 43–66). Stanford: Stanford
Abel, R. M., & Friedman, H. A. (2009). Israeli school and community response to war University Press.
trauma. A review of selected literature. School Psychology International, 30(3), Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hills-
265–281. dale, NJ: Erlbaum.
Anderberg, M. R. (1973). Cluster analysis for applications. New York, NY: Academic de Vries, A., & Ripley, B. D. (2014, December 22). Package ‘ggdendro’. Retrieved from
Press. http://cran.r-project.org/web/packages/ggdendro/ggdendro.pdf
Ayalon, O. (2003). Cope cards for trauma and healing. Community Stress Prevention, Fraley, C., & Raftery, A. E. (2006, September). MCLUST version 3: An R package for nor-
5, 82–94. mal mixture modeling and model-based clustering. Retrieved from http://www.
Ayalon, O. (2004). Children’s responses to terrorist attacks. Living with Terror, Work- dtic.mil/cgi-bin/GetTRDoc?AD=ADA456562
ing with Trauma, 171–200. Gagnon, E. (2001). Power cards: Using special interests to motivate children and youth
Ayalon, O. (2006a). Appeasing the sea: Post-Tsunami training of helpers in Thailand, with Asperger syndrome and autism. Shawnee Mission, KS: Autism Asperger Pub-
Phuket 2005. Traumatology, 12(2), 162. lishing Co.
Ayalon, O. (2006b). Beyond words – Metaphoric story cards (MSC) as a therapy Gawronski, B. (2009). Ten frequently asked questions about implicit measures and
tool (in Hebrew). Retrieved from:. http://www.maagal-m.co.il/publications/ their frequently supposed, but not entirely correct answers. Canadian Psychol-
beyond-words-card-stories.html ogy, 50(3), 141–150.
Ayalon, O. (2007). Healing trauma with metaphoric cards. Therapy Today, Gentleman, R., Ihaka, R., & Bates, D. (1997). The R project for statistical computing. R
(September), 22–24. home web site:. http://www.r-project.org
Backs, R. W., da Silva, S. P., & Han, K. (2005). A comparison of younger and older Hahs, A. (2008). Cards as therapeutic documents. International Journal of Narrative
adults’ self-assessment manikin ratings of affective pictures. Experimental Aging Therapy & Community Work, 2008(1), 30.
Research, 31(4), 421–440. Hennig, C. (2014, December 22). Package ‘fpc’. Retrieved from.
Bradley, M. M., & Lang, P. J. (2007). The International Affective Picture System (IAPS) ftp://jpackage.hmdc.harvard.edu/CRAN/web/packages/fpc/fpc.pdf
in the study of emotion and attention. In J. A. Coan, & J. J. B. Allen (Eds.), Handbook Horlings, R., Datcu, D., & Rothkrantz, L. J. (2008). Emotion recognition using brain
of emotion elicitation and assessment (pp. 29–46). Oxford University Press. activity. In Proceedings of the 9th international conference on computer systems
Bunn, A., & Korpela, M. (2012). Time series analysis in dplR. Retrieved from. http:// and technologies and workshop for PhD students in computing (p. 6). ACM.
cran.case.edu/web/packages/dplR/vignettes/timeseries-dplR.pdf Ito, T. A., Cacioppo, J. T., & Lang, P. J. (1998). Eliciting affect using the International
Center for the Study of Emotion and Attention (CSEA-NIMH). (1995). The interna- Affective Picture System: Trajectories through evaluative space. Personality &
tional affective picture system (Photographic slides). Gainesville: University of Social Psychology Bulletin, 24, 855–879.
Florida, Center for Research in Psychophysiology. Kaufman, L., & Rousseeuw, P. J. (2009). . Finding groups in data: An introduction to
Chanel, G., Kronegg, J., Grandjean, D., & Pun, T. (2006). Emotion assessment: Arousal cluster analysis (Vol. 344) New York, NY: John Wiley & Sons.
evaluation using EEG’s and peripheral physiological signals. In Multimedia con- Kimhi, S., Eshel, Y., Zysberg, L., & Hantman, S. (2009). Getting a life: Gender differ-
tent representation, classification and security. Berlin/Heidelberg: Springer. ences in postwar recovery. Sex Roles, 61, 554–565.
46 R. Cohen et al. / The Arts in Psychotherapy 45 (2015) 36–46
Kotrlik, J. W. K. J. W., & Williams, H. A. W. H. A. (2003). The incorporation of effect size Langfelder, P., Zhang, B., & Horvath, S. (2008). Defining clusters from a hierarchical
in information technology, learning, information technology, learning, and per- cluster tree: The dynamic tree cut package for R. Bioinformatics, 24, 719–720.
formance research and performance research. Information Technology, Learning, http://dx.doi.org/10.1093/bioinformatics/btm563
and Performance Journal, 21(1), 1–7. Libkuman, T. M., Otani, H., Kern, R., Viger, S. G., & Novak, N. (2007). Multidimen-
Lahad, M., Farhi, M., Leykin, D., & Kaplansky, N. (2010). Preliminary study of a new sional normative ratings for the International Affective Picture System. Behavior
integrative approach in treating post-traumatic stress disorder: SEE FAR CBT. Research Methods, 39(2), 326–334.
The Arts in Psychotherapy, 37(5), 391–399. Maechler, M., Struyf, A., & Maechler, M. M. (2008 February 29). The clus-
Lahad, M., Shacham, M., & Ayalon, O. (2012). The “BASIC Ph” model of coping and ter package. Retrieved from. http://download.nextag.com/cran/web/packages/
resiliency: Theory, research and cross-cultural application. Jessica Kingsley Pub- cluster/cluster.pdf
lishers. Masterson, J. T., Findlay, J. C., Kaplan, F., Bridgham, T., Christian, D., Gal-
Lang, P. J. (1980). Behavioral treatment and bio-behavioral assessment: Computer braith, A., et al. (2008). Art therapy and clinical neuroscience. Jessica Kingsley
applications. In J. B. Sidowski, J. H. Johnson, & T. A. Williams (Eds.), Technology Publishers.
in mental health care delivery systems (pp. 119–137). Norwood, NJ: Ablex. Mirkin, B. (2011). Choosing the number of clusters. Wiley Interdisciplinary Reviews:
Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1995). International Affective Picture Data Mining and Knowledge Discovery, 1(3), 252–260.
System (IAPS): Technical manual and affective ratings. Gainesville: University of OH Publishing. (2012). COPE Cards. Retrieved from. http://www.oh-cards.com/
Florida, Center for Research in Psychophysiology. index.php?article id=64&clang=1
Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1997). International Affective Picture Osgood, C., Suci, G., & Tannenbaum, P. (1957). The measurement of meaning. Urbana,
System (IAPS): Technical manual and affective ratings. NIMH Center for the Study IL: University of Illinois Press.
of Emotion and Attention. Rousseeuw, P. J. (1987). Silhouettes: A graphical aid to the interpretation and vali-
Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (1999). International Affective Picture dation of cluster analysis. Journal of Computational and Applied Mathematics, 20,
System (IAPS): Technical manual and affective ratings. NIMH Center for the Study 53–65.
of Emotion and Attention. Rudman, L. A., & Goodwin, S. A. (2004). Gender differences in automatic in-group
Lang, P. J., Bradley, M. M., & Cuthbert, B. N. (2001). International Affective Picture bias: Why do women like women more than men like men? Journal of personality
System (IAPS): Technical manual and affective ratings. NIMH Center for the Study and social psychology, 87(4), 494.
of Emotion and Attention. Russell, J. A., Ward, L. M., & Pratt, G. (1981). Affective quality attributed to environ-
Lang, P. J., & Greenwald, M. K. (1981). The International Affective Picture System stan- ments: A factor analytic study. Environment and Behavior, http://dx.doi.org/10.
dardization procedure and initial group results for affective judgments (Tech. Rep. 1177/0013916581133001
No. 1A). Gainesville: University of Florida, Center for Research in Psychophysi- Scharf, M. (2007). Long-term effects of trauma: Psychosocial functioning of the
ology. second and third generation of Holocaust survivors. Development and psy-
Lang, P. J., Greenwald, M. K., Bradley, M. M., & Hamm, A. O. (1993). Looking at chopathology, 19(02), 603–622.
pictures: Affective, facial, visceral, and behavioral reactions. Psychophysiology, Sloan, D. M., & Marx, B. P. (2006). Exposure through written emotional
30(3), 261–273. disclosure: Two case examples. Cognitive and Behavioral Practice, 13(3),
Lang, P. J., Öhman, A., & Vaitl, D. (1988). The International Affective Picture System 227–234.
(Photographic slides). Gainesville: University of Florida, Center for Research in Unger, R. K. (2001). Handbook of the psychology of women and gender. New York, NY:
Psychophysiology. Wiley.
Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.
Alternative Proxies: