An Experimental Investigation of The Effects of Perspective-Taking On Emotional Discomfort, Cognitive Fusion and Self-Compassion
An Experimental Investigation of The Effects of Perspective-Taking On Emotional Discomfort, Cognitive Fusion and Self-Compassion
An Experimental Investigation of The Effects of Perspective-Taking On Emotional Discomfort, Cognitive Fusion and Self-Compassion
A R T I C L E I N F O A B S T R A C T
Keywords: Background: Perspective-taking exercises are used in a range of therapies, such as Acceptance and Commitment
Acceptance and commitment therapy Therapy (ACT), Cognitive Therapy (CT), Dialectical Behavioural Therapy (DBT) and Compassion-Focused
Relational frame theory Therapy (CFT). Perspective-taking has been theorised in ACT to foster cognitive defusion, self-compassion and
Perspective-taking
a sense of self as distinct from and containing self-related negative thoughts and feelings. To date, few experi
Self-compassion
mental studies have investigated the impact of typical perspective-taking exercises. This study sought to
Cognitive defusion
Compassion-focused therapy investigate whether perspective-taking exercises were able to decrease state cognitive fusion and emotional
discomfort and increase state self-compassion associated with a self-related, negative thought (SRNT). It also
sought to investigate whether there are differences in effects between temporal (‘now’ vs ‘then’) and interper
sonal (‘self’ vs ‘other’) perspective-taking and between giving and receiving perspectives.
Method: A convenience sample of non-clinical participants (n = 61) generated a SRNT and then rated levels of
emotional discomfort, state cognitive fusion and state self-compassion in relation to the thought. Participants
were then guided through three within-participant conditions: a control procedure, a giving perspective and a
receiving perspective condition. Participants were allocated to one of two groups: temporal perspective-taking or
interpersonal perspective-taking. Mixed ANOVAs showed that both interpersonal and temporal exercises
significantly reduced emotional discomfort and cognitive fusion and increased self-compassion associated with a
SRNT. The effects of giving or receiving perspective differed between interpersonal and temporal groups.
Conclusion: These results provide experimental evidence that perspective-taking is a psychologically beneficial
process, therefore supporting the existing use of perspective-taking exercises in clinical practice.
1. Introduction language and cognition which posits that human behaviour involves
learned patterns of responding to stimuli based on the relations between
Acceptance and Commitment Therapy (ACT) is a modern form of them, known as relational framing.
Cognitive Behavioural Therapy (CBT) that targets increased psycho From an RFT perspective, stimuli can acquire properties to influence
logical flexibility (Hayes, Luoma, Bond, Masuda, & Lillis, 2006; Hayes, behaviour based not on physical properties of the stimulus, nor from
Strosahl & Wilson, 2012). Psychological Flexibility is our capacity to be direct contingency between stimulus and response, but from the arbi
aware of what is influencing us, open to all our experiences, even un trary derived relations between stimuli. Such relations are cued by
pleasant ones, and to continue to take steps towards the things that we environmental and language cues such as ‘same as’, ‘opposite’, ‘bigger
most care about. It has been shown to be related to a very wide range of than’, ‘before’, or ‘after’. Crucially, the stimulus functions controlled by
psychopathological outcomes (Hayes et al., 2006) and to be the primary these language cues transfer from one member of a relation to another,
mechanism of action of ACT (Stockton et al., 2019). based upon the relation specified.
ACT emerged in parallel to Relational Frame Theory (RFT; Hayes, Whilst RFT has been supported by experimental research (e.g.
Barnes-Holmes, & Roche, 2001), influencing and informing one another O’Connor, Farrell, Munnelly, & McHugh, 2017), many have found RFT
(Hayes, Strosahl, & Wilson, 2012). RFT is a behaviour analytic theory of to be too technical for easy application in clinical contexts. Applied
* Corresponding author.
E-mail address: [email protected] (D. Gillanders).
https://doi.org/10.1016/j.jcbs.2021.02.004
Received 13 July 2020; Received in revised form 20 January 2021; Accepted 16 February 2021
Available online 22 February 2021
2212-1447/© 2021 Association for Contextual Behavioral Science. Published by Elsevier Inc. All rights reserved.
L. Boland et al. Journal of Contextual Behavioral Science 20 (2021) 27–34
models like ACT have attempted to address this with their use of perspective-taking might operate, based on manipulable features of
non-technical so called ‘mid-level’ terms such as ‘defusion’ and ‘self- context that can be reliably influenced to evoke perspective-taking.
as-context’ (Hayes, Strosahl, & Wilson, 2012). Contextual Behavioural
Science, a guiding framework for both of these fields, advocates for the 1.2. Relational Frame Theory and perspective-taking
implementation of theoretically focused laboratory studies to strengthen
the link between basic RFT principles and ACT’s non-technical terms According to RFT, through repeated interactions with others, chil
(Hayes, Barnes-Holmes & Wilson, 2012; Levin, Hildebrandt, Lillis, & dren learn that their behaviour (including their own private behaviour
Hayes, 2012). such as thoughts, feelings, and urges) is different from others. By
learning to respond consistently to questions such as, “What do you
1.1. Use of perspective-taking in therapeutic work like?“, “Where did you go today?” “Where were you yesterday?” “Where
is Daddy?“, “How are you feeling?“, children derive three core relations
Perspective-taking is defined as the ability to step outside of our called deictic frames. These deictic frames specify a relationship from
selves and ‘see’ or infer aspects of a situation from another’s point of the perspective of the speaker along the dimensions of person, place and
view (Marvin, Greenberg and Mossler, 1976). This ability not only un time, the core deictic relations being I-YOU (interpersonal), HERE-
derlies the development of an understanding of others but is also crucial THERE (spatial) and NOW-THEN (temporal). Unlike other relational
to the healthy development of the self (Novak, 2012). frames, deictic frames are not based on formal dimensions of the envi
Several psychological therapies include perspective-taking in ronment and are always dependent on the perspective of the speaker.
terventions. These have different purposes, depending on the therapy. In When this repertoire of deictic relating is established, the child will
cognitive therapy, imagining what you might say to a friend having a respond to the kinds of questions above from a perspective that is
similar dysfunctional thought is used to initiate ‘cognitive distancing’ consistent and different to other people’s perspectives. A sense of self, or
(Ingram & Hollon, 1986) and reveal the client’s bias in evaluation to the perspective of ‘I’, is therefore derived from multiple examples of
wards themselves, thereby introducing corrective re-appraisal into their learning to talk about one’s own perspective and experience in relation
own thinking (Beck, Rush, Shaw, & Emery, 1979). In Dialectical to others’ perspectives (McHugh, 2015).
Behaviour Therapy (DBT) perspective-taking enhances the client’s Once this perspective of ‘I’ is derived, a person can then relate that
ability to take different perspectives on the same experience and find a deictic ‘I’ in a range of other ways with psychological content. For
‘wise synthesis’. In addition, DBT also employs mindfulness meditation example, the self can be compared to other people, can be related in
to encourage observation of experience, as well as using coordination with negative evaluations, and can participate in condi
perspective-taking skills to enhance interpersonal effectiveness (Line tional framing such as ‘It is my personality that makes me have such
han, 1990, 1993). Perspective-taking exercises are also used in problems in life’. This kind of framing leads to the experience of ‘self as
Compassion Focussed Therapy (CFT), often using mental imagery (e.g. story’ or ‘conceptualised self’ (also called ‘self as content’) in ACT. RFT
Kolts, 2016). Examples of these practices include guiding clients to postulates that in self as story, psychological content and ‘I’ are framed
construct an image of an ideal compassionate friend, to imagine in coordination and both in the here and now (Moran & McHugh, 2019).
receiving kindly support from them, also to imagine giving compassion In contrast, therapeutic use of perspective-taking exercises are the
to others, and to their own suffering. In ACT, perspective-taking exer orised to lead to the addition of frames that establish a distinction
cises are powerful elements of therapy (Villatte et al., 2016), and they relation between the self and psychological content (e.g., ‘I am having
are understood using non-technical terms such as ‘self as context’ and the thought that …). The addition of such distinction relations is
also ‘defusion’. Defusion refers to stepping back from psychological postulated to underpin the experience of cognitive defusion. RFT pos
content, not taking it literally, and reducing its dominance over tulates that this leads to the psychological content being experienced as
behaviour and experience (Gillanders et al., 2014). Self as context refers ‘YOU, THERE, THEN’ and distinct from the self which is experienced as
to a flexible repertoire of perspective taking skills that enhance defusion ‘I, HERE, NOW’. Repeated experience of this deictic distinction framing
by adding a sense of containment of psychological experience. In ACT, is assumed to lead to a more generalised awareness of the distinction
skills of flexible perspective-taking are used to detach from unhelpful between self and psychological content, which is described as ‘self as
patterns of thinking, develop a more observer-based stance, and to foster process’.
empathy and compassion towards one’s self and others (Hayes, Strosahl, Perspective-taking exercises can also lead to the derivation of coor
& Wilson, 2012; Moran, Almada, & McHugh, 2018). dination relations between one’s own struggle and other people’s
The fact that many therapy modalities have employed these kinds of suffering, leading to a transformation of empathy and compassion
interventions has led some researchers to outline commonalities and functions towards the self. Perspective-taking exercises can also lead to
distinctions between these at the level of underlying construct, referring the addition of hierarchical or containing relations between self and
to that as ‘Metacognition’ (Bernstein et al., 2015). These authors propose psychological content. These relations are thought to lead to the expe
that our capacity to shift perspective from within one’s subjective rience of ‘self as context’ in ACT.
experience onto that experience itself is composed of three overlapping Evidence supportive of the RFT account of perspective-taking and
skills: disidentification with internal experience, reduced reactivity to the derivation of self has been provided by several studies. Atkins and
internal experience and increased meta-awareness. They review and Styles (2016) used interview transcripts to code participant speech as
synthesise evidence from across different therapy modalities that shows either ‘self as story’, ‘self as process’, ‘control oriented self-rules’, ‘val
that deficits in these skills are strongly related to poorer mental health ues-oriented self-rules’ or ‘self as context’. Results showed that partici
(Bernstein et al., 2015). pant natural speech could be reliably coded, and that frequency of
Despite Bernstein and colleagues’ work in synthesising these values-oriented self-rules and self as process speech predicted better
perspective-taking concepts across modalities, the metacognitive model wellbeing and reduced distress six and 12 months later. Similar findings
of decentring that they propose remains relatively descriptive, and does were reported in a replication with healthy adolescents by Moran and
not provide a precise, experimentally manipulable framework for McHugh (2020).
investigating perspective-taking in laboratory and clinical settings. Yu, Norton, and McCracken (2017) further supported that a measure
Relational Frame Theory has, however, been used to explore of perspective-taking changed significantly due to an ACT-based pain
perspective-taking interventions conceptually and experimentally, management programme and that these changes were predictive of
leading to a more detailed and testable account of how perspective- changes in pain interference, adjustment and depression, even after
taking interventions may work. The advantage of this account is that controlling for pain intensity and pain acceptance. A limitation of this
it provides a functional, rather than descriptive account of how study is that the direction of causality could also be reversed, in that
28
L. Boland et al. Journal of Contextual Behavioral Science 20 (2021) 27–34
reductions in outcomes such as depression could also have led to the relations involved in perspective-taking of person (Self vs Other) and
improvements in self as context/perspective-taking. time (Now vs Then). The study did not seek to investigate the addition of
Moran and McHugh (2019) showed that for non-clinical adolescents, hierarchical relations as previous experimental work has done. The
framing the self hierarchically with psychological experience was experiment also addressed another research gap by testing the effect of
associated with reduced stress, depressed mood and avoidance behav giving versus receiving alternative perspectives. Consistent with previ
iour, whilst framing the self with only distinction relations was not. ous experimental studies and the theorised role of perspective-taking in
Finally, Moran, Almeida and McHugh (2018) found in a cross-sectional facilitating defusion and self-compassion, it was hypothesised that
analysis of non-clinical participants, that measures of self as content, self perspective-taking would reduce state cognitive fusion and emotional
as process and self as context all independently predicted poorer mental discomfort, and increase state self-compassion.
health in theoretically hypothesised directions.
RFT has also been used to experimentally test elements of 2. Method
perspective-taking exercises under laboratory conditions. Luciano et al.
(2011) created two defusion protocols and tested them with teens who 2.1. Participants and setting
showed problematic behaviours. One protocol had only distinction
framing (Defusion I) between the participant and their own negative A total of 61 participants (48 female, 79%) were recruited through
self-content, whilst the other also added hierarchical (i.e. containing) convenience sampling from on campus and social media advertising.
relations (Defusion II). The Defusion II protocol also contained relations Thirty-nine (64%) of the participants were students. The age of the
linking this novel perspective with greater adaptive functioning. The sample ranged from 22 to 60 (M = 27.87, SD = 0.82). The ethnic
Defusion II protocol led to significantly greater reductions in problem composition of the participants was not especially diverse, with 74% (n
behaviour and was the only protocol to lead to increased psychological = 45) identifying as “White/Caucasian”, 18% (n = 11) as “Asian”, 5% (n
flexibility. = 3) as “mixed/multiple ethnic group” and 3% (n = 2) as “Pacific
Foody, Barnes-Holmes, Barnes-Holmes, and Luciano (2013; 2015) Islander” or “Black/African”. Criteria for exclusion were expression of
replicated and extended these findings. They found that adding hierar suicidal ideation and/or current severe mental distress, and the inability
chical relations led to superior stress-reducing properties in response to to generate a SRNT rated 50 or above on a 0–100 emotional discomfort
a negative thought, compared to distinction relations alone. Other scale. The experiments took place on the university campus in private
experimental work has also investigated defusion exercises under lab rooms.
oratory or clinical analogue conditions (e.g. Larsson, Hooper, Osborne,
Bennet, & McHugh, 2016; Mandavia et al., 2015; Masuda, Twohig, 2.2. Experimental design and analytic plan
Stormo, et al, 2010, 2010a for a review see; Levin et al., 2012). From an
RFT perspective, these can be considered a form of perspective-taking in The current study followed a mixed design with two levels of a
that they create a context in which psychological content is experienced between-group condition and four levels of a within-group (repeated
in a relation of distinction to the self. These experiments differ however, measures) condition. The between-group condition was whether the
in that they achieve their effects not via a deliberate shift of perspective, participant received an interpersonal deictic manipulation (self-versus
but by disrupting usual verbal conventions (e.g., word repetition, other), or a temporal deictic manipulation (current self-versus younger
singing thoughts, saying thoughts in a funny voice (Larsson et al., self). An inert control group was not included, due to concerns about
2016)). being able to recruit a sufficiently large sample to have a three-group
In summary – RFT has provided an account of the basic relations and study. The within-group condition had four levels: baseline, control
transformations of function that are likely to be involved in perspective- procedure, a ‘giving’ condition and a ‘receiving’ alternative perspective
taking and that initial experimental, cross sectional and longitudinal condition. The control procedure included exposure to and discussion of
evidence has supported this account. From this body of work we know the SRNT and generation of mental imagery without explicit deictic
that cognitive defusion appears to involve adding relations of distinction manipulation, in an attempt to control for experimenter effects and
between self and psychological content and that the addition of hier repeated exposure to the SRNT.
archical framing appears to support the establishment of self as context. During the pilot testing of the experimental protocol it was found
What we do not yet know is how this RFT analysis of perspective- that full randomisation of the within-participant conditions would result
taking applies to interventions commonly used in therapeutic work. in reduced ecological validity of the study (it was not possible for the
For example, when a client describes negative self-referential content, it receiving condition to be self-generated and also come before the giving
is common for a therapist to ask the client to imagine what they might condition), and could potentially increase distress for participants
say to a trusted friend who was describing similar thoughts (giving a (ending the experiment on the control condition of having the SRNT
perspective to another). Similarly, the therapist might ask the client to without manipulating perspective was found to be upsetting and led to
imagine what a trusted friend might say to them if they were to express feelings of isolation). The decision was therefore made to complete the
this content to them (receiving an alternative perspective from another). conditions in a set order: baseline, control, giving, and then receiving
In other intervention examples, the therapist might ask the client how perspective condition.
old they were when they first began thinking of themselves in this way. Statistical analyses were conducted using SPSS version 24 (IBM,
The therapist might ask them to imagine they could offer something to 2016). The planned analyses included descriptive statistics to describe
that child version of themselves and to imagine receiving a kindly or the sample (gender, age, ethnicity, and student status), as well as
wise response from an older version of themselves. These kinds of descriptive statistics of the three dependent variables of discomfort,
therapeutic exercises deliberately involve the evoking of perspective- fusion and compassion at each time point. A mixed within – between
taking, and the giving and receiving of alternative perspectives group repeated measures analysis of variance was used to test study
through gesture, imagery and other language-based cues. hypotheses.
The current study used an RFT framework to experimentally inves Participants were fully informed of the study’s methods and pro
tigate the effects of these kinds of perspective-taking exercises on a vided written informed consent to participate. Participants were free to
participant-generated self-relevant negative thought (SRNT). Based withdraw from the study at any time and a safety protocol was employed
upon McHugh (2015), the experiment manipulated the basic deictic prior to beginning the study to screen for suicidal ideation and current
29
L. Boland et al. Journal of Contextual Behavioral Science 20 (2021) 27–34
severe emotional or mental distress. No participants were excluded perspective-taking condition, based on the order of their recruitment to
based on these criteria. A safety protocol for distress during the exper the study. The study took approximately 30 min with each participant.
iment and the provision of support after the experiment by the research The experimental conditions were created specifically for this study by a
team was devised, as well as signposting to local support services. The peer reviewed trainer and expert in ACT (David Gillanders). For pur
study was reviewed by the University of Edinburgh Clinical Psychology poses of ecological validity, the protocol was not closely scripted, but
Ethics and Integrity Committee and approved. instead a prompt guide and detailed training was used to ensure con
sistency of delivery. Each investigator (all masters level students in
2.4. Measures psychology) was trained and observed in their conduct of the experi
mental sessions by the entire research team (including the peer reviewed
Besides the demographic form that included questions regarding age, trainer), until they could safely and accurately deliver the perspective-
gender, ethnicity, occupation and current mental health status, the study taking exercises. The experiment consisted of seven stages, always
also used the following measures, which were completed in a random conducted in the same order, as outlined below.
order after each stage of the experiment. Stage 1: Demographic Screening and Consent. Participants were
informed, provided consent and were screened according to exclusion
2.4.1. Emotional Discomfort Visual Analogue Scale (ED) criteria. Demographic details were recorded.
Self-reports of current emotional discomfort elicited by the SRNT Stage 2: Thought selection and assessment. Participants generated a
were measured by asking participants to respond to the item “Please rate SRNT that was over 50 on the ED and frequent. They shared it with the
how uncomfortable or distressing you find this thought” on a 100 mm researcher and wrote it down verbatim. Examples of thoughts given
visual analogue scale, anchored on the left with 0 (not at all uncom were ‘I’m not thin enough’, ‘I don’t feel I have achieved enough for my
fortable) and on the right with 100 (extremely uncomfortable). age’, ‘I worry people find me irritating’. Participants reflected silently on
Although emotional discomfort is not usually a primary target in ACT, it the effects this SRNT has on their life and the situations in which it
is still a relevant variable in clinical work and including it here allows for arises. When ready, baseline ratings of the SRNT on the emotional
comparison of this study with other similar experimental studies (e.g. discomfort scale were provided, along with completion of the SCFQ and
Luciano et al., 2011, Foody et al., 2013; 2015). SSCQ.
Stage 3: Generation of self at a younger age or trusted ‘other’. In the
2.4.2. State Cognitive Fusion Questionnaire (SCFQ) temporal group, participants were asked to think about how old they
The SCFQ (Bolderston et al., 2019) is a 7-item self-report state were when their SRNT first appeared and to generate an image of their
measure of cognitive fusion. The instructions ask participants to rate younger self at that age. In the interpersonal group participants were
how true each statement is for them at this moment. Sample items asked to think of a person whom they trust and with whom they would
include, “I am so caught up in my thoughts that I don’t know what to do” feel comfortable sharing the SRNT and to generate an image of that
and “I am very entangled in my thoughts”. Preliminary validation person. In both conditions, participants received cues to imagine details
studies showed adequate test-retest reliability (r = .69) and excellent of the locations they were visualising.
internal reliability (α = 0.95). Likewise, the data from the present study Stage 4: Control Condition. In the temporal group, the control condi
also indicated good internal reliability (α = 0.77). Furthermore, the tion involved participants imagining that they were their younger self,
SCFQ is sensitive enough to detect changes in cognitive fusion following and that their current self was able to travel back in time and be with
brief exercises in lab-based studies (Bolderston et al., 2019). them. They were then asked to imagine that the SRNT came to mind
while they were spending time with their older self. In the interpersonal
2.4.3. State Self-Compassion Scale (SSCS) group, participants were asked to imagine that they were with their
The SSCS is a seven item, self-report measure based on the 26 item trusted person and the SRNT came to mind from time to time. In both
Self-Compassion Scale (SCS; Neff, 2003). The SCS contains six subscales: conditions, participants were told that they were having their SRNT, but
Self-Kindness, Self-Judgement, Common Humanity, Isolation, Mindful that they were not sharing it with their trusted person or older self. This
ness and Over-Identification. The original SCS demonstrated good condition was to control for non-specific effects of the imagery exercise
discriminant validity, test-retest reliability (r = 0.93) and internal reli and for repeated exposure to the SRNT, but without cues to give or
ability (Cronbach’s α = 0.92). The state version was adapted specifically receive any particular response. After this stage, participants were asked
for this study. One item was selected from each subscale and an addi to return their attention to their SRNT and to complete the rating
tional item from the Self-Kindness subscale, based on strength of item measures for emotional discomfort, cognitive fusion and self-
loadings in the original validation studies. The wording was also compassion, according to how the thought now appeared to them.
modified to measure state rather than trait self-compassion for a specific Stage 5: Giving Perspective. In the temporal group, this condition
thought. Example items include, “As I think this thought right now, I am involved the experience of giving perspective to the younger self. Par
being disapproving and judgmental towards myself,” and “As I have this ticipants were asked to mentally travel back in time to be with their
thought right now, I am able to be understanding and patient towards younger self. This time they were to remain in the perspective of their
those aspects of myself that I don’t like”. The resulting scale for this current self and imagine that their younger self disclosed the negative
study was the State Self-Compassion Questionnaire, consisting of 7 items thought to them. Participants were asked to reflect on how it felt hearing
measured on a 5-point Likert scale, with possible answers ranging from the SRNT from their younger self and whether they wanted to offer their
‘almost never’ to ‘almost always’. In the present study, the SSCS younger self anything in return. In the interpersonal group, participants
demonstrated good internal reliability (α = 0.73), suggestive of a unitary were instructed to imagine that they were with their trusted person, and
scale. At baseline, the SSCS was moderately correlated with the State their trusted person shared the same (or very similar) SRNT with them,
CFQ (r = -0.38, p < .01), showing initial support for concurrent validity. as if they were struggling with the same (or very similar) problem.
Participants were then asked to describe how it would feel to hear the
2.5. Procedure SRNT from the trusted person, and what they would like to offer as a
response. It is of note that no instructions or suggestions as to the type of
Participants were convenience sampled using on campus adverts, response to be given were made, and yet all participants spontaneously
social media posts and through association with the experimenters. To generated a response that was supportive and compassionate. Partici
avoid confounds and to protect participant privacy, no experimenter pants in both conditions then rehearsed giving this response to either
met with a participant that they knew personally. Participants were their imagined younger self or their trusted other. After this stage,
assigned to either the temporal perspective-taking or the interpersonal participants were again instructed to rate the three aspects of emotional
30
L. Boland et al. Journal of Contextual Behavioral Science 20 (2021) 27–34
Table 1
Descriptive statistics and findings.
Variable Condition Baseline Control Giving Receiving Within η2p Between η2p Interact η2p
F(3,177) F(1,59) F(3,177)
Mean SD Mean SD Mean SD Mean SD
F†(2.5,145.2) F†(2.5,145.2)
a b c c
Emotional Temporal 64.45 8.3 52.31 14.35 40.48 21.25 36.17 19.39 51.16†** .46 .68, ns .01 .34†, ns .01
Discomfort Interpersonal 65.53a 9.81 58.00a 20.63 42.56b 17.29 37.72c 21.36
State Cognitive Temporal 29.93a 6.38 24.55b 7.69 21.38c 9.6 20.66c 7.4 45.11** .43 .59, ns .01 3.83* .06
Fusion Interpersonal 30.16a 7.05 29.84a 11.25 22.03b 8.13 20.00c 8.31
State Self Temporal 20.59a 3.76 25.72b 3.96 27.86c 4.12 28.48c 3.76 54.19** .48 3.84‡, ns .06 2.73* .04
Compassion Interpersonal 20.37a 4.97 21.94a 6.78 26.69b 3.74 26.88b 4.46
Means with different superscripts are significantly different for within participant comparisons at the level of p < .05. Interact = Within by between interaction,
Significance of F test *p < .05 **p < .001, †Greenhouse-Geisser correction applied to degrees of freedom due to violation of sphericity assumption. ‡The p value for this
test was 0.06. Effect size for partial eta squared can be described as following: 0.01 = Small, 0.06 = Medium, 0.14 = Large.
31
L. Boland et al. Journal of Contextual Behavioral Science 20 (2021) 27–34
32
L. Boland et al. Journal of Contextual Behavioral Science 20 (2021) 27–34
contextual cues and learning processes may have been contributing to Future studies could also explore the balance between the naturalistic
the effects. language and procedure used in this study, versus a more tightly
A further important limitation was the decision not to fully ran controlled use of relational language cues to attempt to isolate the in
domise the order of conditions. During pilot testing of the experimental dependent variable at the basic theory level.
procedure, putting the control condition as the final condition in the
experiment was found to enhance feelings of isolation and was moder 4.3. Implications for practice
ately upsetting. Secondly, in pilot testing the giving and receiving con
ditions it was felt to be more authentic if the participant generated the This study has implications for the use of perspective-taking exer
response themselves, compared to the experimenter creating a generic cises in ACT as it experimentally supports that these exercises enhanced
response for them. For that reason, giving a response had to occur before self-compassion and cognitive defusion and led to reductions in
receiving that response. This meant that it was not possible to fully emotional discomfort from self-related, negative thoughts, with large
randomise the order of the within-participant conditions. Although this effects. Similar exercises to the ones used in this study are used in several
is a limitation as it fails to fully separate the effects of each condition, it therapeutic modalities (e.g. Gilbert, 2014; Villatte, 2016). ACT is highly
does enhance the ecological validity of the study, as receiving perspec personalised and responsive to clients’ presenting issues, but this can
tive was judged to follow more naturally from giving perspective and make manualisation of ACT challenging (Thompson, Luoma, & LeJeune,
thus more closely mirrored how these exercises are used in clinical 2013). Clarifying the specific effects of such therapeutic techniques may
practice. However, the lack of randomisation limits the conclusions that help the development of more cost-effective, modularised ACT protocols
can be drawn about the differential effects of giving versus receiving which make use of more targeted exercises (Villatte et al., 2016). It is of
perspective, and future research in which conditions are randomised note that the temporal perspective-taking led to experimental effects
may provide fuller insight into such effects. Nevertheless, the large effect more rapidly than the interpersonal perspective-taking. This does sug
sizes observed on all dependent variables suggests that the cumulative gest that in therapeutic work, temporal perspective-taking may be the
effects of perspective-taking in sequence has the potential to have more powerful route to transformation. However, some clients may also
beneficial psychological effects. find this experience too challenging. Given that the interpersonal
Another limitation was the design of the temporal control condition, perspective-taking achieved the same magnitude of result across the
which may have inadvertently brought about perspective-taking in entire experiment, that approach could be an equally effective, but more
participants and thus results from the giving and receiving perspective gentle way to work.
conditions are more difficult to interpret. Future research on temporal
deictic framing should seek to design a control condition which avoids 4.4. Conclusion
cueing participants to take the perspective of their younger self before
beginning the active experimental conditions. This study was important in being the first to distinguish between the
A further limitation of this study was the exclusive reliance on self- effects of interpersonal versus temporal perspective-taking, and giving
report measures of dependent variables. The repeated use of these versus receiving perspective. Results indicated that both interpersonal
three measures may have made the dependent variables being measured and temporal perspective-taking decrease emotional discomfort and
obvious to participants and thus resulted in demand characteristics state cognitive fusion and increase state self-compassion in relation to a
(Nichols & Maner, 2008). To avoid this, participants were regularly negative thought, though more tightly controlled studies are needed to
reminded that there were no expected answers, but future research may clarify the effects of giving versus receiving and to be confident that the
benefit from behavioural or physiological measures, in addition to results seen can be interpreted as reflecting deictic relations.
self-reports.
In addition, the adaptation of the Self Compassion Scale Short Form
Funding
as a state measure has limited data on its validity, beyond this study.
Whilst we know from this study that the measure was internally
This research did not receive any specific grant from funding
consistent, correlated as theoretically expected with the standardised
agencies in the public, commercial, or not-for-profit sectors.
measure of state cognitive fusion and was able to detect the effects of the
experimental manipulation, it is acknowledged that suggesting this as
Declaration of competing interest
evidence of the measure’s validity and of the experiment’s validity is
tautological. It is of note that Neff, Tóth-Király, Knox, Kuchar, and
None.
Davidson (2020) have recently published a state version of the Short
Form Self-Compassion scale. It is very similar to the version that was
created for this experiment, with two of the seven items the same. Neff Acknowledgements
and colleagues report the unitary factor structure and psychometric
properties of their State Self Compassion Scale (Short Form) as excellent. The authors would like to thank Dr Matthieu Villatte for conversa
Given that the Neff et al. scale has had more extensive development than tions about giving and receiving perspectives that led to the ideas for
the scale used in this study, future studies should use it. That scale was part of this study, and to Ms Rosina Pendrous and Dr Lee Hulbert Wil
not available at the time that the current study was conducted. liams for suggestions about the state adaptation of the Self Compassion
The study is also limited by being conducted on a predominantly Scale.
student sample, recruited through convenience sampling, without major
mental health difficulty. As such the sample is self-selected, which could References
introduce sampling bias. Participants were also not asked if they had
Atkins, P., & Styles, R. (2016). Measuring self and rules in what people say: Exploring
previously undertaken psychological therapy. The sample may therefore whether self-discrimination predicts long-term wellbeing. Journal of Contextual
have contained people who have experienced perspective taking exer Behavioral Science, 5(2), 71–79.
cises or self-compassion exercises prior to the experiment, which may Beck, A. T., Rush, A. J., Shaw, B. F., & Emery, G. (1979). Cognitive therapy of depression.
New York, NY: Guilford Press.
have been a confound. Bernstein, A., Hadash, Y., Lichtash, Y., Tanay, G., Shepherd, K., & Fresco, D. M. (2015).
Future studies could replicate this experiment with people who Decentering and related constructs: A critical review and metacognitive processes
report more significant distress related to their self-relevant negative model. Perspectives on Psychological Science, 10(5), 599–617.
Bolderston, H., Gillanders, D., Turner, G., et al. (2019). The initial validation of a state
thinking, or who score highly on fears of self-compassion, and could version of the Cognitive Fusion Questionnaire. Journal of Contextual Behavioral
more tightly control previous exposure to psychological interventions. Science, 12, 207–215.
33
L. Boland et al. Journal of Contextual Behavioral Science 20 (2021) 27–34
Foody, M., Barnes-Holmes, Y., Barnes-Holmes, D., & Luciano, C. (2013). An empirical Masuda, A., Feinstein, A., Wendell, J., et al. (2010a). Cognitive defusion versus thought
investigation of hierarchical versus distinction relations in a self-based ACT exercise. distraction: A clinical rationale, training, and experiential exercise in altering
International Journal of Psychology and Psychological Therapy, 13(3), 373–388. psychological impacts of negative self-referential thoughts. Behaviour Modification,
Foody, M., Barnes-Holmes, Y., Barnes-Holmes, D., Rai, L., & Luciano, C. (2015). An 34(6), 520–538.
empirical investigation of the role of self, hierarchy and distinction in a common Masuda, A., Twohig, M., Stormo, A., et al. (2010b). The effects of cognitive defusion and
ACT exercise. Psychological Record, 65(2), 231–243. thought distraction on emotional discomfort and believability of negative self-
Gilbert, P. (2014). The origins and nature of compassion focused therapy. British Journal referential thoughts. Journal of Behavior Therapy and Experimental Psychiatry, 41(1),
of Clinical Psychology, 53, 6–41. 11–17.
Gillanders, D., Bolderston, H., Bond, F., et al. (2014). The development and initial McHugh, L. (2015). A Contextual Behavioural Science approach to the self and
validation of the Cognitive Fusion Questionnaire. The Behavior Therapist, 45(1), perspective-taking. Current Opinion in Psychology, 2, 6–10.
83–101. Moran, O., Almada, P., & McHugh, L. (2018). An investigation into the relationship
Hayes, S., Barnes-Holmes, D., & Wilson, K. (2012b). Contextual behavioural science: between the three selves (self-as-content, self-as-process and self-as-context) and
Creating a science more adequate to the challenge of the human condition. Journal of mental health in adolescents. Journal of Contextual Behavioral Science, 7, 55–62.
Contextual Behavioural Science, 1(1–2), 1–16. Moran, O., & McHugh, L. (2019). Patterns of relational responding and a healthy self in
Hayes, S. C., Barnes-Holmes, D., & Roche, B. (2001). Relational frame theory: A post- older adolescents. Journal of Contextual Behavioral Science, 12, 74–80.
skinnerian account of human language and cognition. New York: Kluwer Academic/ Moran, O., & McHugh, L. (2020). Measuring occurrences of self and other
Plenum Publishers. discriminations in relation to mental health in adolescent textual responses. Journal
Hayes, S., Luoma, J., Bond, F., Masuda, A., & Lillis, J. (2006). Acceptance and of Contextual Behavioral Science, 15, 253–263.
commitment therapy: Model, processes and outcomes. Behaviour Research and Neff, K. (2003). The development and validation of a scale to measure self-compassion.
Therapy, 44(1), 1–25. Self and Identity, 2(3), 223–250.
Hayes, S. C., Strosahl, K. D., & Wilson, K. G. (2012a). Acceptance and Commitment Neff, K. D., Tóth-Király, I., Knox, M. C., Kuchar, A., & Davidson, O. (2020). The
Therapy: The process and practice of mindful change (2nd ed.). New York: The Guilford development and validation of the state self-compassion scale (long- and Short form).
Press. Mindfulness, 2016.
IBM Corp. (2016). IBM SPSS statistics for macintosh, version 24.0. Armonk, NY: IBM Corp. Nichols, A. L., & Maner, J. K. (2008). The good-subject effect: Investigating participant
Ingram, R. E., & Hollon, S. D. (1986). Cognitive therapy for depression from an demand characteristics. The Journal of General Psychology, 135(2), 151–165.
information processing perspective. In R. E. Ingram (Ed.), Information processing Novak, G. (2012). Precursors to perspective-taking. Approaches to the self. In L. McHugh,
approaches to clinical psychology (xi ed., pp. 259–281). San Diego, CA: Academic & I. Stewart (Eds.), The Self and perspective-taking: Contributions and Applications from
Press. modern behavioural science (rev. Edn (pp. 73–89). Oakland, Canada: Context Press.
Kolts, R. L. (2016). CFT made simple: A clinician’s guide to practicing compassion-focussed O’Connor, M., Farrell, L., Munnelly, A., & McHugh, L. (2017). Citation analysis of
therapy. Oakland: New Harbinger. relational frame theory: 2009–2016. Journal of Contextual Behavioral Science, 6(2),
Larsson, A., Hooper, N., Osborne, L., Bennet, P., & McHugh, L. (2016). Using brief 152–158.
cognitive restructuring and cognitive defusion techniques to cope with negative Stockton, D., Kellett, S., Berrios, R., Sirois, F., Wilkinson, N., & Miles, G. (2019).
thoughts. Behaviour Modification, 40(3), 452–482. Identifying the underlying mechanisms of change during acceptance and
Levin, M., Hildebrandt, M., Lillis, J., & Hayes, S. (2012). The impact of treatment commitment therapy (ACT): A systematic review of contemporary mediation studies.
components suggested by the psychological flexibility model: A meta-analysis of Behavioural and Cognitive Psychotherapy, 47(3), 332–362.
laboratory-based component studies. The Behavior Therapist, 43(4), 741–756. Thompson, B. L., Luoma, J. B., & LeJeune, J. T. (2013). Using Acceptance and
Linehan, M. M. (1990). Cognitive-behavioral treatment of borderline personality disorder. Commitment Therapy to guide exposure-based interventions for posttraumatic stress
New York: Guilford Press. disorder. Journal of Contemporary Psychotherapy, 43(3), 133–140.
Linehan, M. M. (1993). DBT skills training manual (2nd ed.). New York: Guilford Press. Villatte, M. (2016). A manifesto for clinical RFT, keynote address presented at the 10th
Luciano, C., Ruiz, F. J., Torres, V., et al. (2011). A Relational Frame analysis of defusion conference of the association for contextual behavioural science, Australia and New
interactions in Acceptance and Commitment Therapy. A preliminary and quasi- Zealand chapter (ANZACBS), melbourne, november 2016. In L. McHugh, I. Stewart,
experimental study with at-risk adolescents. International Journal of Psychology and & P. Almada (Eds.), 2019) A contextual behavioral guide to the self: Theory & Practice.
Psychological Therapy, 11(2), 165–182. Oakland: New Harbinger.
Mandavia, A., Masuda, A., Moore, M., et al. (2015). The application of a cognitive Villatte, J. L., Vilardaga, R., Villatte, M., Plumb Vilardaga, J. C., et al. (2016). Acceptance
defusion technique to negative body image thoughts: A preliminary analogue and Commitment Therapy modules: Differential impact on treatment processes and
investigation. Journal of Contextual Behavioral Science, 4(2), 86–95. outcomes. Behaviour Research and Therapy, 77, 52–61.
Marvin, R. S., Greenberg, M. T., & Mossler, D. G. (2017). The early development of Yu, L., Norton, S., & McCracken, L. M. (2017). Change in “self-as-context” (“Perspective-
conceptual perspective Taking : Distinguishing among multiple perspectives. Child Taking”) occurs in acceptance and commitment therapy for people with chronic pain
Development, 47(2), 511–514. and is associated with improved functioning. The Journal of Pain, 18(6), 664–672.
34