Content uploaded by Jennifer S Daks
Author content
All content in this area was uploaded by Jennifer S Daks on Oct 16, 2020
Content may be subject to copyright.
Journal of Contextual Behavioral Science 18 (2020) 214–238
Available online 2 October 2020
2212-1447/© 2020 Association for Contextual Behavioral Science. Published by Elsevier Inc. All rights reserved.
Review Articles
Examining the correlates of psychological exibility in romantic
relationship and family dynamics: A meta-analysis
Jennifer S. Daks, Ronald D. Rogge
*
Psychology Department, University of Rochester, Rochester, NY, USA
ARTICLE INFO
Keywords:
Meta-analysis
Psychological exibility
Family dynamics
Romantic relationships
Child functioning
Adult attachment
Parenting
ABSTRACT
A growing body of research supports the importance of ACT’s therapeutic targets (i.e., dimensions of psycho-
logical exibility and inexibility) for promoting individual wellbeing. The current systematic review and meta-
analysis extended that work by examining how specic dimensions of psychological exibility and inexibility
are linked to family and romantic relationship functioning. Drawing from the ACT, mindfulness, and emotion
regulation literatures, 5006 records were initially identied via PsychInfo, Web of Science, and Google Scholar
searches, resulting in a nal set of 174 papers representing 203 distinct samples (95 romantic relationship
samples, 101 family samples, and 7 samples evaluating both romantic relationship and family functioning),
yielding a combined sample of 43,952 total subjects. Although the review was unable to identify sufcient
numbers of studies to meta-analyze the relationship and family correlates of a subset of exibility dimensions
(contact with values, committed action, experiential avoidance, fusion, lack of contact with values, inaction), the
review identied sufcient studies to meta-analyze the correlates of: acceptance, present moment awareness/
mindfulness, cognitive defusion, self-as-context, global exibility, lack of present moment awareness, self-as-
content, and global inexibility. Correspondingly, a total of 840 effects were extracted from the original
studies, ultimately yielding 137 meta-analytic effects (using random effects models) that show a range of effect
sizes (−0.51 to 0.61). Within families, higher levels of various forms of parental exibility were linked to: (1)
greater use of adaptive parenting strategies, (2) lower use of lax, (3) harsh, and (4) negative parenting strategies,
(5) lower parenting stress/burden, (6) greater corresponding family cohesion, and (7) lower child internalizing
and (8) externalizing symptoms. Within romantic relationships, higher levels of various forms of psychological
inexibility were linked to: (1) lower relationship satisfaction for themselves and (2) their partners, (3) lower
sexual satisfaction and (4) emotional supportiveness, as well as (5) higher negative conict, (6) physical
aggression, (7) attachment anxiety, and (8) attachment avoidance. Taken as a set, these results suggest that
psychological exibility and inexibility may play key roles both in couples and families to shape how in-
dividuals interact with the people closest to them.
The qualities of close interpersonal relationships (e.g., parent-child
relationships, romantic relationships) have shown robust links with
both physical and mental health across the lifespan (e.g., Haley, Levine,
Brown, & Bartolucci, 1987; Steinberg, 2001; Walen & Lachman, 2000).
We conceptualize that these processes begin in early childhood within
the family context, where children witness the dynamics between a
parental romantic relationship (i.e., parental responsiveness) and are
directly involved in parent-child dynamics (i.e., adaptive vs. maladap-
tive parenting behaviors; modeling of emotion regulation skills; and
emotional coaching). Factors such as positive family climate and healthy
parental models of emotion regulation provide children with models for
how to regulate their own emotions (see Morris, Silk, Steinberg, Myers,
& Robinson, 2007 for a review). Such contextual familial factors that
lead children to develop healthy internal coping strategies have also
been linked with positive child development, improved academic
achievement, and fewer internalizing and externalizing problems (e.g.,
Amato & Fowler, 2002; Kaczynski, Lindahl, Malik, & Laurenceau,
2006). Building on the wealth of research surrounding Acceptance and
Commitment Therapy (ACT; Hayes, Strosahl, & Wilson, 1999, 2011),
the current review sought to examine dimensions of psychological
exibility as critical elements of this developmental process. The current
review therefore sought to examine the literature linking psychological
* Corresponding author.
E-mail address: ronald.rogge@rochester.edu (R.D. Rogge).
Contents lists available at ScienceDirect
Journal of Contextual Behavioral Science
journal homepage: www.elsevier.com/locate/jcbs
https://doi.org/10.1016/j.jcbs.2020.09.010
Received 18 March 2020; Received in revised form 23 September 2020; Accepted 25 September 2020
Journal of Contextual Behavioral Science 18 (2020) 214–238
215
exibility to multiple dimensions of romantic relationship, family, and
child functioning.
1. Conceptual frameworks
Psychological Flexibility/Inexibility. Growing out of ACT
(Hayes, Strosahl, & Wilson, 2011; Hayes, Luoma, Bond, Masuda, & Lillis,
2006) and Relational Frame Theory (RFT; Hayes, Barnes-Holmes, &
Roche, 2001), psychological exibility has been dened as a set of skills
that individuals engage when presented with difcult or challenging
thoughts, feelings, emotions, or experiences. Thus, ACT targets six spe-
cic forms of psychological inexibility: (1) experiential avoidance (i.e.,
actively avoiding difcult thoughts, feelings and experiences), (2) lack
of contact with the present moment (i.e., going through daily life in a
distracted and inattentive manner), (3) fusion (i.e., getting stuck in
difcult thoughts and feelings), (4) self as content (i.e., seeing difcult
thoughts and feelings as a reection of oneself and judging/shaming
oneself for having them), (5) lack of contact with values (i.e., losing
track of one’s deeper priorities within the stress and chaos of day to day
life), and (6) inaction (i.e., easily getting derailed by setbacks or difcult
experiences, leaving one unable to take steps toward deeper goals).
Within ACT, these rigid and inexible responses to difcult or chal-
lenging experiences constitute the Hexaex model (Fig. 1A) and are
considered to be dysfunctional, ultimately contributing to and exacer-
bating psychopathology. Thus, these dimensions of inexibility may
serve as a transdiagnostic model for how negative, rigid, and inexible
reactions to both internal and external cues might promote forms of
psychological distress underlying a wide range of diagnoses.
As an intervention, ACT targets the rigid and inexible dimensions
of the Hexaex model with exercises specically designed to help in-
dividuals approach challenging experiences in a more exible manner.
Thus, ACT seeks to promote six corresponding dimensions of psycho-
logical exibility (Fig. 1B): (1) acceptance (i.e., developing the ability
to open oneself to all experiences, both the good and the bad, accepting
them no matter how challenging or difcult they might be), (2) contact
with the present moment (i.e., developing a mindful attentive aware-
ness of the present moment throughout one’s day-to-day life), (3)
cognitive defusion (i.e., developing the ability to gently experience
thoughts and feelings, allowing them to pass through one’s experience
like leaves on a stream without clinging to them), (4) self as context (i.
e., developing the ability to maintain a broader perspective even in the
midst of difcult thoughts and feelings), (5) contact with values (i.e.,
learning to actively maintain contact with one’s deeper values
throughout one’s daily life, no matter how stressful of chaotic each day
becomes), and (6) committed action (i.e., developing the resilient
ability to continue taking steps toward one’s goals, even in the face of
difcult experiences and setbacks).
ACT-based interventions targeting the exible and inexible di-
mensions of the Hexaex model have been shown to improve individual
functioning in over 300 randomized clinical trials spanning a wide range
of clinical diagnoses, populations, and presenting problems (see A-Tjak
et al., 2015; Hayes et al., 2006; Hayes, Pistorello, & Levin, 2012; Ruiz,
2010 for reviews). A growing body of work has also begun to demon-
strate that change experienced on individual components of exibility
during treatment serve as mechanisms to explain the treatment benets
observed in ACT (e.g., Fledderus, Bohlmeijer, Fox, Schreurs, & Spin-
hoven, 2013; Forman, Herbert, Moitra, Yeomans, & Geller, 2007).
Consistent with this, experimental studies deconstructing ACT down to
its component elements of the Hexaex model have demonstrated direct
links between specic dimensions of psychological exibility and indi-
vidual wellbeing (see Levin, Hildebrandt, Lillis, & Hayes, 2012 for a
review). Given this robust body of work, psychological exibility has
Fig. 1. The Hexaex (A & B) and Enduring-Vulnerability Stress-Adaptation (C) models informing the current review.
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
216
been posited as a dynamic process helping to improve individual func-
tioning and well-being (see Kashdan & Rottenberg, 2010 for a review).
Family Systems Theory. To provide an overarching conceptual
framework for our investigation of how psychological exibility might
shape close relationships, we drew upon Family Systems Theory (Bro-
derick, 1993; Minuchin, 1985) and more recent theories building on this
theory (e.g., transactional family dynamics; Schermerhorn & Cum-
mings, 2008). These theories view the family as a complex system with
multiple levels of functioning (e.g., romantic relationship dyads,
co-parent dyads, parent-child dyads, sibling dyads) positing that an in-
dividual’s wellbeing and emotional functioning are intimately tied to
processes and behavioral interactions going on at all of those different
levels within the family system. Thus, to fully understand (and to
effectively improve) the functioning of an individual child or parent
(herein referred to as parent(s), yet included in this construct are
non-parental individuals serving as primary caregivers to children), it is
necessary to understand the dynamics operating at various levels within
that family. For example, a conict within the romantic relationship of
two parents could lead to one of those parents disengaging from
parenting responsibilities (e.g., engaging in more lax parenting) or even
actively undermining the parenting efforts of their co-parent (e.g.,
engaging in higher co-parenting conict and triangulation), potentially
leading to greater family conict and greater amounts of child problem
behavior. Thus, although research on romantic relationships is a robust
eld of study in its own right, the romantic dyad functions as a key
system within the larger family systems framework, as disruptions
within that relationship have been shown to spillover into other areas of
family and child functioning (e.g., Sturge-Apple et al., 2006). From the
family systems perspective, to fully understand how psychological
exibility might impact family functioning, it is necessary to examine
not only individual functioning (e.g., parenting stress), but also
romantic relationship functioning (e.g., relationship satisfaction, sexual
satisfaction, relationship conict/aggression, emotional support,
attachment insecurities), broader family dynamics (e.g., family con-
ict/chaos, family cohesion), parenting dynamics (e.g., lax, over-
reactive, harsh/negative, adaptive), and child functioning (e.g., child
internalizing and externalizing symptoms). Thus, Family Systems The-
ory provided a broad conceptual framework for the current investiga-
tion (1) identifying key systems (i.e., relationships), within families (2)
identifying key dynamics (i.e., specic processes), to assess within each
system, and (3) providing an over-arching framework moving forward
to understand how dynamics within one system or relationship could
spillover into other aspects of family dynamics (e.g., Erel & Burman,
1995; Krishnakumar & Buehler, 2000).
Enduring-Vulnerability, Stress-Adaptation Model. To provide a
more focused framework to model how psychological exibility might
impact the dynamics within specic close relationships, the current
investigation drew up on the Enduring-Vulnerability, Stress-Adaptation
model (EVSA; Karney & Bradbury, 1995) from the romantic relationship
literature. The EVSA model focuses on a single relationship and helps to
model how individual traits and vulnerabilities along with external
stressors might interact with individual and relationship processes to
affect the health and wellbeing of that one specic relationship across
time. This process-focused model posits that individuals bring both
personal strengths and weaknesses (i.e., enduring strengths/vulner-
abilities) into their close relationships (Fig. 1C). Although daily life and
the broader context in which individuals live will present dyads with
stressful events to navigate, the EVSA model posits that the enduring
vulnerabilities they bring with them can also produce stressful events (e.
g., misunderstandings, conict, betrayals). When a stressful or difcult
experience occurs, the two individuals in a close relationship must draw
upon their own internal (e.g., psychological exibility) and interper-
sonal skills (e.g., communication, support, compassion, empathy) to
navigate that stressful event. A failure to navigate such events in a
manner where both people feel understood, nurtured, and valued can
then not only serve to erode the quality of that close relationship, but it
can also serve to produce additional stressful events in the future (e.g.,
more conict, greater avoidance). From this perspective, trait levels of
psychological inexibility could be conceptualized as enduring vulner-
abilities that individuals bring with them into families and romantic
relationships. Those rigid and inexible responses to difcult situations
could serve to intensify negativity and reactivity within various family
dynamics (e.g., processes like parenting behavior, family conict, &
relationship conict), leading to lower levels of relationship quality (e.
g., lower family cohesion and relationship satisfaction), which would in
turn lead to poorer individual functioning (e.g., higher parenting stress,
higher child externalizing & internalizing symptoms). Trait inexibility
could also serve to generate larger numbers of difcult and challenging
events for families to navigate. In contrast, trait level of psychological
exibility could be conceptualized as enduring strengths that in-
dividuals bring into their family dynamics, buffering the impact of
stressful events and promoting kinder and more loving interactions at all
levels within the family system. As psychological exibility is likely to
also uctuate across time and contexts (e.g., in response to ACT; e.g.,
Fledderus et al., 2013), within this modied version of the EVSA model
(Fig. 1C) the dynamic (i.e., state-like) aspects of psychological exibility
and inexibility could be conceptualized as critical intrapersonal (i.e.,
within individual) factors at the center of this model, serving to buffer or
exacerbate (respectively) the adverse effects of stress and conict on
subsequent relationship dynamics. This modied form of the EVSA
model therefore serves as a method of modeling each of the dyadic re-
lationships within a family (e.g., the parents’ romantic relationship,
individual parent-child relationships, relationships among siblings,
etc.), thereby identifying how exibility within specic family members
might inuence their relationships with other members of the family. In
contrast, the Family Systems model provides a larger framework for
understanding how the dynamics of individual relationships can spill-
over into other family systems, organizing those interdependent re-
lationships into a coherent model.
2. Bridging and integrating multiple literatures
Recent measurement work (e.g., Rogge, Daks, Dubler, & Saint, 2019;
Rolffs, Rogge, & Wilson, 2016) has provided the rst comprehensive
measure of the 12 distinct dimensions of psychological exibility (pre-
sent moment awareness, values, committed action, self-as-context,
defusion, and acceptance) and inexibility (lack of present moment
awareness, lack of contact with values, inaction, self-as-content, fusion,
and experiential avoidance) within the Hexaex model: the Multidi-
mensional Psychological Flexibility Inventory (MPFI; Rolffs et al.,
2016). Work with the MPFI has also claried that measures from the
mindfulness and emotion regulation literatures demonstrated strong
convergent validity with subscales of the MPFI (Rogge et al., 2019).
Those results therefore highlighted that despite having different con-
ceptual labels, key scales and subscales from those literatures measure
aspects of psychological exibility within the Hexaex model (e.g., the
Mindful Attention Awareness Scale; MAAS; Brown & Ryan, 2003; the
Five Facet Mindfulness Questionnaire; FFMQ; Baer, Smith, Hopkins,
Krietemeyer, & Toney, 2006, 2008; the Difculties with Emotion
Regulation Scale; DERS; Gratz & Roemer, 2004). This work also claried
that total scores on common measures like the Acceptance and Action
Questionnaire – II (AAQ-II; Bond, Baer, Hayes, & Waltz, 2011) and the
FFMQ primarily represent measures of global inexibility. The current
review is the rst to capitalize on these new insights, allowing us to draw
from all three literatures when identifying studies for inclusion. This
measurement work also provides the rst operational framework for
examining how the 12 distinct dimensions of exibility and inexibility
within the Hexaex model are linked to relationship and family func-
tioning. In fact, the measurement work clarifying convergent validity
and conceptual overlap across these three elds also allowed us to
determine how more focused scales like the Interpersonal Mindfulness
in Parenting Questionnaire (IM-P; Duncan, 2007) could most
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
217
appropriately be included in the meta-analysis.
3. Previous reviews
The comprehensive review conducted for the current study
(described below) failed to identify any published systematic reviews
or meta-analyses with a similar comprehensive focus. However, the
review uncovered a smaller and more focused meta-analysis of the
romantic relationship correlates of mindfulness (McGill, Adler-Baeder,
& Rodriguez, 2016) and 9 of its 10 papers were contained within the
174 records of the current review. Similarly, a number of smaller re-
views focused on the benets of mindfulness-based parenting in-
terventions demonstrated small amounts of overlap with the current
review (4 of the 7 articles in Alexander, 2018 overlapped; 3 of the 25
articles in Burgdorf, Szabo, & Abbott, 2019; 1 of the 17 articles in
Taylor, Cavanaugh, & Strauss, 2016). Finally, a number of reviews of
mindfulness and acceptance-based interventions failed to demonstrate
any overlap with the current review (i.e., no overlap with the 10 ar-
ticles in Cachia, Anderson, & Moore, 2016; the 22 articles of Fried-
mutter, 2015; the 24 articles in Harnett & Dawe, 2012; the 7 articles in
Hwang & Kearney, 2014; the 11 articles in Stephenson, 2017; the 7
articles in Townshend, Jordan, Stephenson, & Tsey, 2016; or the 4
articles in Whittingham, 2014). Thus, the current review offers a
unique contribution to the literature.
4. The current study
Within the broader framework of Family Systems Theory and the
more process-oriented framework of the EVSA model, the current study
sought to examine dimensions of psychological exibility and inexi-
bility as critical processes inuencing family and close-relationship dy-
namics (drawing from the complete family and romantic relationship
literatures). Building on recent measurement work (e.g., Rolffs et al.,
2016) this meta-analysis is the rst to integrate and synthesize ndings
from the ACT, mindfulness, and emotion regulation literatures. Litera-
ture searches conducted in the Fall of 2019 identied 5006 titles/ab-
stracts which resulted in 174 records containing 203 unique (sub)
samples and yielding 840 effects. This sample allowed us to test the
following hypotheses.
Hypothesis 1. Consistent with the EVSA model, we hypothesized that
dimensions of psychological exibility would be associated with greater
use of adaptive processes in families (e.g., adaptive parenting) and
romantic relationships (e.g., emotional support) whereas dimensions of
psychological inexibility would be associated with greater maladaptive
family processes (e.g., family conict, negative/harsh/lax parenting).
Hypothesis 2. We further hypothesized that dimensions of psycho-
logical exibility would be associated with higher quality family re-
lationships (e.g., family cohesion) and romantic relationships (e.g.,
relationship satisfaction, sexual satisfaction) whereas dimensions of
psychological inexibility would be associated with lower quality
relationships.
Hypothesis 3. Finally, we hypothesized that dimensions of parent
psychological exibility would be associated with higher levels of parent
functioning (i.e., lower parenting stress) and child functioning (i.e.,
lower child internalizing and externalizing symptoms).
5. Method
5.1. Selection of studies
In accordance with PRIMSA guidelines (Moher et al., 2009), a sys-
tematic literature search was conducted using PsycInfo (on October 14,
2019). Our search specied that records (i.e., their titles, abstracts, key-
words) should include at least one term referring to psychological
exibility or mindfulness (i.e., exibility, present moment awareness,
mindful*, defusion, cognitive fusion, committed action, decentering,
non-reactivity, nonjudge*, distress endurance, inexibility, experiential
avoidance, inaction, overidentication, multidimensional experiential avoid-
ance questionnaire, Toronto Mindfulness Scale, Five Facet Mindfulness
Questionnaire, Five Factor Mindfulness Questionnaire, Mindful Attention
Awareness Scale, Difculties with Emotion Regulation Scale, Self-compassion
Scale) as well as including at least one term representing the family or
romantic relationship correlates (i.e., family conict, family cohesion,
family chaos, parent*, relationship satisfaction, relationship quality, rela-
tionship conict, breakup, divorce, intimacy, romantic, social support,
emotional support, partner, spous*, dating). As shown in Fig. 2, this yielded
3873 records. In a second round of identication (November to December
2019), we conducted a search within Web of Science for articles whose
titles contained relevant content using identical logic/search terms to
those used in the rst round of identication. At this stage, we also con-
ducted a Google Scholar reverse citation search on a set of highly relevant
(i.e., articles whose titles/abstracts highlighted links between ex-
ibility/mindfulness and relationship or family functioning) and widely
cited articles from the rst stage of identication. This reverse citation
search within Google Scholar helped to ensure the inclusion of recent
work within the set of samples collected. To minimize publication biases,
this second stage of identication also searched unpublished disserta-
tions, master’s theses, and honors theses. Finally, in the second stage of
identication we also screened the introductions of articles identied in
the rst round to ensure complete coverage of all prior work in the
literature. The second stage of identication ultimately yielded another
1474 records for a total of 5006 unique records to be screened. To further
address possible publication bias, we contacted the authors of 83 records
who had assessed one or more of our exibility-interpersonal functioning
pairs of constructs but failed to present the corresponding correlation(s)
within their manuscripts. A total of 16 authors responded to our requests,
yielding effects from another 12 samples.
5.2. Inclusion/exclusion criteria
Studies were considered to meet eligibility criteria if they: (a) were
written in English, (b) were conducted with human subjects, (c)
included a measure of at least one of the 12 dimensions of psychological
exibility (see Supplemental Tables S1 and S2), (d) also included one of
the family or romantic relationship correlates examined, (e) reported an
effect size either in the form of a Pearson’s r correlation coefcient or a
standardized regression coefcient, (f) and represented a novel and
previously unpublished effect (relevant if multiple papers had been
published out of the same data). Fig. 2 details the screening process
which ultimately identied 174 records yielding 203 unique (sub)sam-
ples from which effects could be extracted.
5.3. Data extraction
Procedure. The rst author coded sample characteristics (e.g., cross-
sectional vs. longitudinal), study characteristics (e.g., sample size, de-
mographics), record characteristics (published vs. unpublished, ACT vs.
Mindfulness literature), and effect sizes in the 203 (sub)samples. The
second author then checked each point of data extracted. Discrepancies
were rare (<1%) and were resolved through discussion.
Classication of exibility variables. Our review comprehensively
explored all 12 dimensions of exibility and inexibility posited within
the Hexaex model, including two additional global dimensions (Global
Flexibility and Global Inexibility), yielding a total of 14 possible pre-
dictor variables (PV). This process was facilitated by ndings with the
MPFI (Rolffs et al., 2016) which not only offered distinguishable sub-
scales assessing those separate components, but also highlighted that
scales from the mindfulness literature (e.g., the FFMQ, the MAAS)
assessed components of exibility using a different set of conceptual
labels. For example, the FFMQ non-judgment subscale is effectively a
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
218
measure of self-as-content within the Hexaex framework and the FFMQ
non-reactivity subscale serves as a measure of defusion (see supple-
mental Tables S1 & S2 for complete details on the content coding of
various scales). Despite the conceptual breadth of our search, only 8 of
the 14 dimensions offered sufcient numbers of effects to be
meta-analyzed (i.e., at least 5 distinct effects with at least one of the
family or romantic relationship correlates).
Preserving the tone of item content in the direction of effects. A
large portion of the research in this area has been done using either the
most common mindfulness measure, the MAAS (e.g., Barnes, Brown,
Krusemark, Campbell, & Rogge, 2007), or the most common psycho-
logical inexibility measure, the AAQ-II (e.g., Whittingham, Wee,
Sanders, & Boyd, 2013). Both the MAAS and the AAQ-II are typically
reverse scored. Thus, responses to their predominantly negatively
worded items assessing inattention/distraction (i.e., lack of present
moment awareness) and inexibility, respectively, are reversed to
represent mindfulness and exibility. However, recent measurement
work (Rogge et al., 2019; Rolffs et al., 2016) has challenged this prac-
tice, suggesting that dimensions of psychological exibility, though
correlated with their inexibility counterparts, are conceptually
distinct from them (e.g., acceptance is correlated yet meaningfully
distinct from experiential avoidance; fusion from defusion, etc.). Thus,
the positively worded items of existing scales seem to assess distinct
constructs (i.e., forms of exibility) from negative items (which seem to
assess forms of inexibility). In extracting effects, we were therefore
careful to attend to the direction of the majority of the original items of
each scale, ensuring that all effects were recorded in a manner consistent
with the direction of those items. Thus, the MAAS was considered a
measure of lack of present moment awareness, the AAQ-II was consid-
ered a measure of global inexibility, and the directions of their effects
were adjusted accordingly.
Classication of family and relationship correlates. All in-
struments and questions used to assess family functioning and romantic
relationship functioning across the records were evaluated at the item
level to determine the common constructs they each measured. For
example, scales with item content assessing global romantic relationship
evaluations (e.g., “In general, how satised are you with your relationship?”
“I have a warm and comfortable relationship with my partner?” “How
rewarding is your relationship with your partner?“) were classied as
measures of relationship satisfaction. Similarly, scales with item content
assessing parenting stress/burdens (e.g., “I feel overwhelmed by the re-
sponsibility of being a parent” “It is difcult to balance different re-
sponsibilities because of my children”) were classied as measures of
parenting stress (see supplemental Table S3 for a complete listing of this
coding process).
Effects extracted. A total of 840 effects were extracted from the 203
distinct (sub)samples identied by our screen. A single effect was
extracted for each distinct exibility-correlate pair within a study. If a
study measured the association between psychological exibility and
interpersonal processes across multiple time points, we extracted the
baseline cross-sectional association and the association between base-
line and the longest follow-up time point to separately examine longi-
tudinal effects of that association. Although the vast majority (96.4%) of
our extracted effects between psychological exibility and interpersonal
functioning were presented as Pearson r correlations in their respective
studies, we decided to include the 30 effects (3.6%) which had been
presented as standardized regression coefcients. In doing so we
recognize that this introduces some heterogeneity in the estimation of
our effects, though given the small number of these coefcients, and the
potential biases in excluding studies we decided to proceed with them in
the analyses (Peterson & Brown, 2005).
5.4. Data analysis
Meta-analytic estimates. Our analyses were conducted using
Rstudio v.1.1.453 (R Core Team, 2018) using the metafor package
(Viechtbauer, 2010). Random effects modeling using maximum likeli-
hood estimation was used to estimate our meta-analytic effects in the
current study (Borenstein, Hedges, Higgins, & Rothstein, 2009), as this
approach assumes heterogeneity in effect sizes. We saw this as an
Fig. 2. Flow of information through the different phases of the systematic review.
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
219
appropriate method given the heterogeneity in our sample sizes, mea-
surement instruments, populations, and methods of data collection, all
of which may be factors that impact the results obtained across studies.
I
2
indices were estimated to quantify the level of heterogeneity present
in each meta-analytic effect.
Examining effects. All extracted effects in their forms as correla-
tions were transformed to Fisher’s Z values and weighted by sample size
before analysis. These effects were then meta-analyzed, and the results
were subsequently transformed back into correlations for ease of inter-
pretation (Lipsey & Wilson, 2001). The presence of potential outliers
was assessed using the inuence measures function within the metafor
package in R, which calculates outlier diagnostics (e.g., studentized
residuals, Cook’s distances, covariance ratios), and identies individual
effects that are disproportionally inuential to the overall effect.
Although these analyses identied a small number of outlying effects,
results from analyses with and without the outliers remined largely
unchanged across the 137 meta-analytic effects calculated. Thus, to
ensure maximum representation of the current literature, all reported
results included those potentially outlying effects.
Publication bias. Our search strategy was successful at identifying
61 unique samples from unpublished sources, representing 30% of all of
the samples analyzed and thereby lessening the risk of publication bias
articially inating our meta-analytic effects. To further address the
possibility of publication bias, we examined funnel plots for each effect
and ran Egger’s regression tests to examine possible funnel plot asym-
metry (Egger, Davey Smith, Schneider, & Minder, 1997). When these
screens indicated possible publication bias, we ran trim and ll analyses
to provide an adjusted average effect size (Duval, 2005) to help correct
our funnel plot asymmetry. Finally, to further explore possible publi-
cation bias, we ran selection method analyses for each meta-analytic
effect using R functions developed and validated by McShane, Bock-
enholt, and Hansen (2016). These analyses estimated the relative
probability of a contradictory nding (i.e., non-signicant or in the
opposite direction of the predominant ndings) being included in the
analysis in comparison to the probability of a consistent and signicant
nding being included. On this index, relative probabilities close to a
value of 1.0 would suggest the presence of very little publication bias
within the current sample whereas relative probabilities much lower
than 1.0 would suggest publication bias. For the 108 meta-analytic ef-
fects based on 3 or more effects, a majority of these relative probabilities
(75%) were close to a value of 1.0 or greater than 1.0 (i.e., ≥0.80),
yielding a median relative probability of 86% that contradictory nd-
ings would have been included across those effects.
6. Results
6.1. Description of studies
Sources of samples. As shown in Fig. 2, our screening of 5006 titles
and abstracts yielded 697 possible records (i.e., manuscripts, theses, and
dissertations) for inclusion. Screening the full texts of those 697 records
then identied the nal 174 records, providing 203 distinct (sub)sam-
ples representing 43,952 unique respondents. A total of 840 effects were
extracted from these (sub)samples, ultimately yielding 137 meta-
analytic effects (72 effects based on 5 or more studies). Table 1 pre-
sents details on the 174 manuscripts included in the current sample and
Table 2 provides a comprehensive overview of the resulting 203 sam-
ples. As seen in Tables 1 and 2, although 142 of our samples (70%) came
from peer-reviewed articles, 61 of the samples (30%) came from un-
published sources including dissertations, master’s theses, honors the-
ses, and papers currently under review. A majority of the samples (86%)
were grounded within the larger literature on mindfulness and the vast
majority (98–99%) yielded cross-sectional, self-reported effects.
Although we screened for effects with all 12 dimensions of the Hexaex
model (as well as for global exibility and inexibility), only 8 di-
mensions emerged (lack of present moment awareness, self-as-content,
global inexibility, present moment awareness, cognitive defusion,
acceptance, self-as-context, and global exibility) with effects from 5 or
more studies linking them to at least one dimension of family or rela-
tionship functioning (see Table 2). The systematic and meta-analytic
reviews therefore focused on those eight dimensions.
Demographics of samples. Although a majority (58%) of the
samples included both men and women, the overall sample included
more women than men as 73% of the 43,952 respondents were women.
The samples were fairly diverse with 66% Caucasian respondents across
all studies. The samples also represented the full adult lifespan as the
average ages within each sample ranged from 17.7 to 70.4 years old,
ranging from emerging adults still in college to older adults in their 70’s
and 80’s. The average age across all 43,952 subjects was 34.1 suggesting
that a majority of subjects were in their 20’s, 30’s, and 40’s across these
samples. The samples spanned a range of relationship stages with sample
average relationship lengths spanning 0.7–28.7 years. However, as these
studies focused on romantic relationships and individuals with children,
65% of the subjects across these studies were married or cohabiting and
the average relationship length across all subjects was 7.9 years, rep-
resenting fairly long-term relationships. Within the family studies, the
ages of the target children ranged from newborns to 16 years old with an
average child age of 9.4 years across all of the studies. Taken together,
these results suggest that the 43,952 subjects across these studies
generally represent a sample of young adults in their 20’s, 30’s, and 40’s
starting their lives and their families with romantic partners.
6.2. Links to family functioning
Parenting stress. As shown in Table 3, one of the most widely
studied family correlates of psychological exibility was parenting
stress/burden. Fig. 3 presents forest plots for some of these meta-
analytic effects. Using Cohen’s (1992) guidelines for correlational ef-
fect sizes (small: r =0.1-0.29; moderate: r =0.3-0.49; and large: r =0.5
and above), the three dimensions of inexibility examined with this
correlate all demonstrated moderate links to current levels of parenting
stress. Thus, higher levels of lack of present moment awareness (most
commonly measured with the MAAS or the FFMQ awareness subscale;
see Fig. 3A), self as content (i.e., judging or shaming difcult feeling-
s/experiences, most commonly measured with the FFMQ non-judgment
subscale), and global inexibility (most commonly measured with the
AAQ-II or the FFMQ total scores; Fig. 3C) were predictive of higher
levels of parenting stress. This suggests that, consistent with the Hexa-
ex model and Hypothesis 3, parents’ rigid and inexible responses to
daily challenging or difcult situations within families might serve to
exacerbate the adverse impact of those experiences, leaving parents
feeling more stressed out, overwhelmed and burdened. In contrast,
higher levels of attentive awareness, cognitive defusion, acceptance
(Fig. 3B), self as context, and global exibility (Fig. 3D) were robustly
linked to lower parenting stress. Thus, parents’ abilities to approach
challenging feelings and situations with non-judgmental attentive
awareness, accepting them, and gently experiencing them while main-
taining a broader perspective might help to buffer parents from the
potential draining impact of those events.
Family conict and cohesion. Consistent with Hypothesis 2, lack of
present moment awareness and global inexibility were moderately
predictive of lower family cohesion, suggesting that generally being
inattentive and distracted as a parent and responding to difcult or
challenging experiences in an inexible, rigid and avoidant manner
might serve to weaken family bonds. In contrast, parents’ global exi-
bility was moderately predictive of lower family conict and greater
family cohesion, highlighting that engaging psychologically exible
skills in response to challenging thoughts, feelings, and moments could
not only serve to enhance individual well-being but also to improve the
quality of interactions within families.
Adaptive parenting. Consistent with Hypothesis 1, lack of present
moment awareness and global inexibility were also mildly linked to
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
220
Table 1
Structural Review of Records Included.
Study Authors (year) Sample M
age
Dimensions of
Flexibility/Inexibility
# of effects
ID Code Nmeta Population Fam Rom Long
100 DS Myhr (2015) 104 parents of children ages 3-5 30.0 INFLEX 2
101 PR Beer, Ward, and Moar (2013) 28 parents of children with Autism Spectrum
Disorder (ASD)
43.2 PMA, SCX, ACPT, FLEX 8
102 DS Dodsworth (2018) 152 parents of elementary school children 30.5 PMA, SCX, ACPT 3
103 PR Bluth and Wahler (2011a) 118 mothers of middle school adolescents 45.5 LPMA 3
104 PR Bluth and Wahler (2011b) 50 mothers of preschoolers 0.0 LPMA 1
105 PR Neff and Faso (2015) 51 parents of children with ASD 41.1 FLEX 2
106 DS Donnelly (2012) 15 adolescent-parent dyads 47.1 INFLEX 1
107 PR Brown, Whittingham, and Sofronoff
(2015)
59 parents of children with an acquired traumatic
brain injury age 2-12
0.0 INFLEX 2
108 PR Burke and Moore (2015) 192 parents of children 10-18 44.0 FLEX, INFLEX 10
109 PR Cheron, Ehrenreich, and Pincus
(2009)
267 parents of children being assessed for anxiety
disorders (partially-dyadic)
45.3 INFLEX 6
110 PR Conner and White (2014) 154 mothers of children with and without ASD age
4-17
41.0 LPMA 1
111 PR Coyne and Thompson (2011) 74 mothers at high risk b/c SES/minority status 31.3 INFLEX 2
112 PR de Bruin et al. (2014) 311 mothers of adolescents w/and w/o diabetes 45.5 PMA, SCX, ACPT, FLEX 24
113 PR Moyer et al. (2018) 43 mothers with PTSD symptoms from
relationship or sexual violence
37.7 INFLEX 1
114 DS Workman (2018) 102 parents of children with ASD receiving ABA
services
34.3 FLEX, INFLEX 6
115 PR Feinberg, Kerns, Pincus, and Comer
(2018)
45 mothers of anxious and non-anxious children 38.0 INFLEX 1
116 PR Geurtzen, Scholte, Engels, Tak, and
van Zundert (2015)
901 parents of adolescents 45.2 PMA, SCX, ACPT 6
117 PR Gouveia, Carona, Canavarro, and
Moreira (2016)
333 parents of school-aged children 42.3 FLEX, LPMA 8
118 PR Greco et al. (2005) 66 mothers of preterm children 0.0 INFLEX 1
119 PR Jones, Hastings, Totsika, Keane, and
Rhule (2014)
110 parents of children with ASD 45.5 INFLEX 4
120 PR Lloyd and Hastings (2008) 91 mothers of children with intellectual
disabilities (ID)
41.6 FLEX, LPMA 6
121 PR MacDonald, Hastings, and Fitzsimons
(2010)
99 fathers of children receiving services for an ID 47.1 INFLEX 2
122 PR McCafferey, Rietman, & Black (2017) 203 parents of children 2-16 41.0 PMA, SCX 3
123 PR Westphal, Leahy, Pala, and
Wupperman (2016)
326 adults in outpatient psychotherapy 34.1 FLEX 1
124 DS Kil (2018) 127 mother-child adolescent dyads 44.9 FLEX, INFLEX 2
125 PR Chan and Lam (2017) 424 parents of children with ASD 43.6 FLEX 1
126 PR Gouveia, Canavarro, and Moreira
(2019)
726 parent-child dyads with normal weight,
overweight, and obese children
42.4 FLEX 1
127 PR Mikl´
osi, Szab´
o, and Simon (2017) 144 Hungarian parents 40.4 LPMA 2
128 PR Moreira and Canavarro (2017) 679 Portuguese parents of school aged children 40.4 PMA, SCX, ACPT, FLEX 28
129 PR Kelly and Dupasquier (2016) 153 Canadian female college students
retrospectively reporting on parenting in
childhood
20.2 FLEX 1
130 HT Mihacsi (2016) 87 mothers of 12-week-old infants 27.0 DFN, LPMA, SCN 3
131 DS Murdock (2017) 132 parents of children diagnosed with ADHD 0.0 PMA, SCX, ACPT, FLEX 4
132 PR Parent et al. (2010) 162 parent-child dyads with children at risk for
depression
41.9 LPMA 4
133 PR Parent et al. (2014) 242 parents from low-income black single-mother
families with male cohabitating parents
(dyadic)
39.6 LPMA 6
134 PR Parent, McKee, N.Rough, and
Forehand (2016a)
615 parents of children 3-17 35.9 FLEX, LPMA, INFLEX 8
134b PR Brassell et al. (2016) 615 parents of children 3-17 35.9 INFLEX 15
135 PR Kim, Kr¨
ageloh, Medvedev, Duncan,
and Singh (2019)
554 Korean parents 42.7 PMA, SCX, ACPT, FLEX 8
136 PR Lo et al. (2018) 394 parents of pre-school children from Hong Kong 35.5 PMA, SCX, ACPT, FLEX 12
137 MT Benton (2017) 30 mother-adolescent dyads 0.0 PMA, SCX, ACPT, FLEX 8
138 DS Kambouras (2018) 594 parents of children with ASD age 2-20 42.0 INFLEX 2
139 PR Potter, Yar, Francis, and Schuster
(2014)
211 adults reporting on parenting received in
childhood
30.2 FLEX 1
140 PR Shea and Coyne (2011) 144 low-income mothers of Head Start pre-
schoolers
32.0 INFLEX 4
141 PR Siu, Ma, and Chui (2016) 216 Chinese mothers of kindergarteners 35.5 LPMA 5
142 PR Wang et al. (2018) 168 Chinese mother-child dyads 42.5 FLEX 1
143 DS Jones (2011) 128 parents of children 0-17 33.2 DFN, LPMA, SCN,
INFLEX
12
144 DS Hegg (2013) 21 parents of adolescents receiving special
education services
0.0 DFN, LPMA, SCN, 3
145 DS Kelly (2016) 130 parents of children age 2-5 34.5 INFLEX 2
146 PR 228 parents of children with ASD age 6-21 0.0 INFLEX 1
(continued on next page)
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
221
Table 1 (continued )
Study Authors (year) Sample M
age
Dimensions of
Flexibility/Inexibility
# of effects
ID Code Nmeta Population Fam Rom Long
Weiss, Cappadocia, MacMullin,
Viecili, and Lunsky (2012)
147 PR Whittingham et al. (2013) 94 parents of children with cerebral palsy 0.0 INFLEX 2
148 PR Williams and Wahler (2010) 40 mothers of children with internalizing or
externalizing problems
38.0 LPMA 4
149 DS Duncan (2007) 753 mother-adolescent dyads 40.7 FLEX 2
150 PR Wong, Mak, and Liao (2016) 180 Chinese parents of children with ASD 42.3 FLEX 1
151 PR Corthorn and Milicic (2016) 62 Chilean mothers of preschool children 36.0 PMA, SCX, DFN, ACPT,
FLEX, SCN, INFLEX
14
152 MT Carlson (2017) 244 fathers of children 0.0 FLEX 3
153 PR Fischer et al. (2016) 117 Australian college students retrospectively
reporting on parenting in childhood
20.0 LPMA, INFLEX 2
154 PR Fulton, Kiel, Tull, and Gratz (2014) 380 undergraduate students retrospectively
reporting on parenting in childhood
20.2 INFLEX 1
155 DS Cohen (2010) 184 rst time mothers in 2nd or 3rd trimester 30.3 FLEX, INFLEX 3
156 PR Ahmed and Bhutto (2016) 116 college students (undergraduates and
graduates)
24.0 FLEX 1
158 PR Emerson, Ogielda, and Rowse (2019) 85 parents of children ages 8-12 41.8 INFLEX 2
159 PR G¨
ordesli, Arslan, Çekici, Sünbül, and
Malkoç (2018)
380 Turkish parents 40.0 PMA, SCX, FLEX 2
160 DS Greunke (2019) 330 parents of children 6–12 years old 38.1 LPMA 1
161 PR Han et al. (2019) 2237 Chinese parents of 6–12 year old children 38.5 FLEX, INFLEX 8
162 PR Henrichs, van den Heuvel, Witteveen,
Wilschut, and Van den Bergh (2019)
118 mothers of 4 year olds 36.0 FLEX 2
163 MT Rayburn (2019) 18 expectant and new fathers 32.0 INFLEX 1 1
164 PR Kantrowitz-Gordon (2018) 857 pregnant women 29.5 DFN, LPMA, SCN,
INFLEX
4
165 PR Lengua, Ruberry, McEntire, Klein,
and Jones (2018)
44 primary caregivers of preschoolers 0.0 DFN, LPMA, SCN,
INFLEX
8
166 PR Moreira, Fonseca, Caiado, and
Canavarro (2019)
335 Portuguese working parents 38.9 PMA, SCX, ACPT 3
167 PR O’Leary, Dockray, and Hammond
(2016)
87 pregnant women 31.3 LPMA 1
168 PR Pan, Liang, Zhou, and Wang (2019) 765 Chinese parents of children with and without
ASD
37.2 PMA, SCX, ACPT 9
169 PR Pickard, Townsend, Caputi, and
Grenyer (2017)
36 pregnant mothers 31.6 DFN 0 1
169b PR Pickard, Townsend, Caputi, and
Grenyer (2018)
30 mothers followed through rst 6 months of an
infant’s life
31.7 INFLEX 1 0
170 PR Brown et al. (2018) 28 families involved with child protective
services
31.0 DFN, LPMA, SCN 12
171 DS Ward (2019) 32 mothers of children with and without ASD 0.0 LPMA 1
172 PR Zhang, Zhang, and Gewirtz (2019) 313 wives of deployed service-men 35.7 INFLEX 1
172b PR Zhang, Piehler, Gewirtz, Zamir, and
Snyder (2019)
155 parents in families with a currently or recently
deployed male service member (dyadic)
36.9 DFN, LPMA, SCN 6
173 PR Zhang, Wang, and Ying (2019) 472 parents of preschool children 0.0 FLEX, LPMA 4
174 DS Mobley (2012) 109 undergraduate students 19.4 INFLEX 1
175 PR Maughan and Weiss (2017) 57 parents of children with ASD 43.6 FLEX, 3
176 PR Lo, Chan, Szeto, Chan, and Choi
(2017)
180 parents of preschool children with
developmental disabilities
38.9 PMA, SCX, ACPT, FLEX 12 4
177 BK Hwang and Kearney (2015) 6 mothers of children with ASD 40.5 FLEX 3
178 PR Smit, Martens, Ackland, and Mikami
(2018)
24 parents of children age 3–5 with behavioral
problems
32.5 PMA, SCX, ACPT, FLEX 8 8
179 PR Potharst, Zeegers, and B¨
ogels (2018) 15 mothers of toddlers 37.3 PMA, SCX, DFN, ACPT,
FLEX, LPMA, SCN,
INFLEX
24
180 PR Potharst, Aktar, Rexwinkel, Rigterink,
and B¨
ogels (2017)
39 mothers of infants 33.6 PMA, SCX, DFN, ACPT,
FLEX, LPMA, SCN,
INFLEX
24
181 PR Chaplin et al. (2018) 83 highly stressed mothers of adolescents 47.4 PMA, SCX, ACPT, FLEX 8
182 DS Nerenberg (2014) 587 parents from military families post-
deployment
36.2 PMA, SCX, LPMA, SCN,
INFLEX
16
183 PR Brockman et al. (2016) 184 military families (both parents and a target
child)
37.2 INFLEX 1 2
184 PR McCracken and Gauntlett-Gilbert
(2011)
183 adolescents in tx for chronic pain and their
parents
43.6 FLEX 2
185 PR Psychogiou et al. (2016) 253 parents of young children 37.6 FLEX 4
186 MT Geier (2012) 50 mother-adolescent dyads 40.0 PMA, SCX, ACPT 6
187 PR Turpyn and Chaplin (2016) 157 mother-adolescent dyads 0.0 FLEX 2
188 PR Cheung, Leung, and Mak (2019) 142 Chinese parents of children with ASD under 18
years old
43.1 INFLEX 2
189 PR Hicks, Dayton, Brown, Muzik, and
Raveau (2018)
82 parents from a metropolitan Midwest city 27.2 DFN, LPMA, SCN 3
190 PR Wu et al. (2019) 1057 parents of Chinese middleschool students 44.5 PMA, SCX 3
(continued on next page)
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
222
Table 1 (continued )
Study Authors (year) Sample M
age
Dimensions of
Flexibility/Inexibility
# of effects
ID Code Nmeta Population Fam Rom Long
191 PR Potharst et al. (2019) 67 Dutch mothers of pre-school children 36.3 INFLEX 5
200 PR Pakenham and Samios (2013) 69 Australian individuals with multiple sclerosis
(MS) and their partners
43.2 LPMA, INFLEX 4
201 UR Daks, Rogge, & Fincham (2019) 2109 individuals in romantic relationships 32.0 PMA, SCX, LPMA 8 2
202 PR Reddy, Meis, Erbes, Polusny, and
Compton (2011)
49 male soldies who recently returned from
combat and their spouses
34.5 INFLEX 8
203 DS Saavedra (2011) 1355 men in romantic relationships 31.7 DFN, LPMA, SCN 40
204 PR Saavedra, Chapman, and Rogge
(2010)
1702 individuals involved in a romantic relationship 28.5 LPMA 4
205 PR Shorey, Seavey, Quinn, and Cornelius
(2014)
481 female undergraduate students who have been
in at least one dating relationship
18.6 DFN, LPMA, SCN 6
206 PR Shorey, Elmquist, et al. (2014) 109 male undergraduate students currently in a
dating relationship
18.4 INFLEX 3
207 PR Shorey, Larson, and Cornelius (2014) 379 female undergraduate students that have been
in at least one romantic relationship
18.5 DFN, LPMA, SCN 6
208 PR Wachs and Cordova (2007) 29 married couples 40.0 LPMA 3
209 PR Williams and Cano (2014) 51 individuals with chronic pain and their
partners (dyadic)
58.5 DFN, LPMA, SCN 12
210 DS Chang (2018) 510 college students 27.3 DFN, LPMA, SCN,
INFLEX
8
211 DS Asber (2014) 39 couples in romantic relationships (≥6mo) with
one partner in a MBSR intervention
45.0 LPMA 2
212 DS Clark (2014) 60 undergrads in exclusive relationships (≥1mo) 20.5 DFN, LPMA, SCN,
INFLEX
8
213 DS Giolzetti (2012) 328 individuals in close relationships 30.5 DFN, LPMA, SCN,
INFLEX
8
214 DS Horst (2013) 247 undergraduate students currently in
relationships (dyadic)
20.0 LPMA 1 1
215 DS Imholte (2017) 26 partners in mod-distressed/sexually
dissatised couples (dyadic)
51.4 INFLEX 4
216 DS Kessie (2017) 268 partners in married, working couples 44.0 LPMA 2
217 DS Kimmes (2016) 531 individuals currently in relationships 20.3 DFN, LPMA, SCN 9
218 DS Larson (2012) 97 female undergrads who experienced physical
violence or psychological abuse from a partner
in the last year
21.8 INFLEX 2
219 DS Lute (2015) 196 undergraduate students currently in dating
relationships
19.3 DFN, LPMA, SCN 3
220 DS Michaels (2007) 49 individuals in relationships and completing a
mindfulness group intervention
45.6 INFLEX 1
221 DS Ormiston (2011) 298 individuals in relationships (≥6mo) 38.6 LPMA 4
222 DS Wiggins (2012) 331 individuals in relationships (≥1yr) 35.1 DFN, LPMA, SCN,
INFLEX
6
223 PR Adair, Boulton, and Algoe (2018) 127 couples together ≥1 year (dyadic) 27.1 DFN, LPMA, SCN 6
224 PR Zamir, Gewirtz, Labella, DeGarmo,
and Snyder (2018)
456 partners in military couples in which the male
recently returned from deployment (dyadic)
36.3 INFLEX 6
225 PR Barnes et al. (2007) 146 emerging adults (18–25) in dating
relationships
19.7 LPMA 4 1
226 PR Homan (2018) 126 older adults (aged 60–95) 70.4 FLEX 2
227 PR Gallagher, Hudepohl, and Parrott
(2010)
167 heterosexual males who reported drinking
with in the last month
26.4 LPMA 1
228 PR Hertz, Laurent, and Laurent (2015) 103 couples in relationships (≥2mo; dyadic) 21.3 INFLEX 2
229 PR Johns, Allen, & Coop-Gordon (2015) 94 participants who partners cheated (57% no
longer with those partners)
42.0 DFN, LPMA, SCN 3
230 PR Kappen, Karremans, Burk, and
Buyukcan-Tetik (2018)
448 individuals in relationships (118 in
mindfulness workshops)
39.3 ACPT, INFLEX 8
231 PR Khaddouma and Gordon (2018) 376 emerging adults in monogamous dating
relationships (dyadic)
18.1 DFN, LPMA, SCN,
INFLEX
16
232 PR Khaddouma, Coop-Gordon, & Bolden
(2015a)
104 undergraduate students in dating relationships 18.1 INFLEX 2
233 PR Khaddouma, Coop-Gordon, & Bolden
(2015b)
322 sexually active undergraduate students in
dating relationships
18.8 DFN, LPMA, SCN 6
234 PR Khalian and Barry (2016) 162 spouses from newlywed couples in their rst
marriages (dyadic)
30.1 LPMA 6
235 PR Rolffs et al. (2016) 1460 adults in romantic relationships and/or with
children
34.8 PMA, SCX, DFN, ACPT,
FLEX, LPMA, SCN,
INFLEX
8 5
236 PR Kimmes, Jaurequi, May, Srivastava,
and Fincham (2018)
185 undergraduate students in relationships 19.9 LPMA 4 1
237 DS Farzan-Kashani (2018) 316 undergrads in relationships (study1) and men
in tx for domestic violence (study2)
28.6 LPMA 4
238 PR Lenger, Gordon, and Nguyen (2017) 173 long-term married couples 52.5 DFN, LPMA, SCN,
INFLEX
8
239 PR Baugh, Cox, Young, and Kealy (2019) 231 individuals in romantic relationships for >1
year
38.2 INFLEX 1 1
(continued on next page)
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
223
lower levels of adaptive parenting practices, suggesting that inattentive
and distracted parents who generally respond in rigid and inexible
manners to stressful experiences, might have a more difcult time
responding to their children’s misbehavior in sensitive, compassionate,
and responsive ways. In contrast, global exibility was strongly pre-
dictive of higher adaptive parenting across the 8 studies (n =4404
parents) examining that link, yielding one of the strongest meta-analytic
effects in the study (r =0.61). This nding therefore highlights the
Table 1 (continued )
Study Authors (year) Sample M
age
Dimensions of
Flexibility/Inexibility
# of effects
ID Code Nmeta Population Fam Rom Long
240 PR Brem et al. (2018) 203 female undergrad psychology students in
relationships (≥1mo)
19.2 LPMA 0 1
240b PR Brem et al. (2019) 203 female undergrad psychology students in
relationships (≥1mo)
19.2 DFN, SCN 0 4
241 PR Shaver, Lavy, Saron, and Mikulincer
(2007)
70 individuals with previous meditation
experience
50.0 DFN, LPMA, SCN,
INFLEX
0 8
242 PR Walsh, Balint, Smolira SJ,
Fredericksen, and Madsen (2009)
127 students and staff in a university psychology
department
24.0 LPMA 0 2
243 DS Pepping (2013) 942 Australian undergraduate psychology students 21.1 DFN, LPMA, SCN,
INFLEX
1 6
243b PR Pepping, Davis, & O’Donovan (2013) 942 Australian undergraduate psychology students 21.3 INFLEX 0 2
244 MT Balvaneda (2017) 66 college student romantic couples 24.5 INFLEX 0 1
245 DS Brownlee (2016) 179 individuals in relationships who experienced
at least 1 adverse childhood event
29.7 INFLEX 1 3
246 HT Buchanan (2019) 146 undergraduate psychology students in
exclusive relationships (≥6mo)
20.0 INFLEX 1
247 PR Don (2019) 191 individuals 19–75 years old 37.7 LPMA, INFLEX 5
248 MT Edwards (2014) 162 newly married spouses (dyadic) 30.8 LPMA 4
249 DS Miller (2012) 47 spouse of a partner experiencing chronic pain
for >3 months
51.8 ACPT 1
250 PR Gambrel and Piercy (2015a) 33 couples expecting their rst child 31.7 DFN, LPMA, SCN,
INFLEX
8
251 PR Forste (2017) 63 adults aged 18+in relationships (≥3mo) 0.0 FLEX 1
252 MT Greer (2017) 809 individuals in romantic relationships 34.1 DFN, LPMA, SCN 6
253 PR Harvey, Crowley, and Woszidlo
(2019)
338 partners in relationships (dyadic) 29.8 LPMA 6
254 PR Knowles, Manusov, and Crowley
(2015)
91 individuals in relationships 38.1 LPMA, SCN 4
255 MT Hermes (2018) 434 partners in heterosexual couples (dyadic) 38.2 INFLEX 5 2
256 MT Russell (2017) 34 parents expecting a child (dyadic) 30.0 LPMA 4
257 MT Schaeffer (2019) 262 undergraduate college students 19.9 FLEX 1
258 PR Simou and Moraitou (2018) 92 adults in relationships (≥5yrs) 51.2 LPMA, SCN 2
259 MT Smith (2015) 222 mothers of 6th and 7th grade students 0.0 INFLEX 1
260 PR Janovsky, Clark, and Rock (2019) 117 single and coupled adults in Australia 34.7 LPMA, SCN, INFLEX 3
261 PR Jones, Bodie, and Hughes (2016) 183 college students with little or no meditation
experience
21.0 DFN, LPMA, SCN 3
262 PR Kappen, Karremans, and Burk (2019) 113 adults in relationships (≥1yr) 27.3 ACPT, INFLEX 4 2
263 PR Leavitt, Lefkowitz, and Waterman
(2019)
194 heterosexuals aged 35–60 who were married
(≥2yrs)
45.3 PMA, SCX, INFLEX 6
264 DS Magin (2018) 55 undergraduate friend/partner dyads (33% in
romantic rel)
19.6 LPMA 1
265 PR Manusov, Stoeth, Harvey, and
Crowley (2018)
137 undergraduates in relationships 23.2 DFN, LPMA, SCN,
INFLEX
4
266 DS McGill (2016) 157 women in relationships 36.3 LPMA 2
267 PR Montes-Maroto, Rodríguez-Mu˜
noz,
Antino, and Gil (2018)
60 cohabiting, dual-earning couples 41.5 LPMA 1
268 PR Pepping, Cronin, Lyons, and Caldwell
(2018)
407 individuals in relationships (≥1yr) 38.5 LPMA 1
269 PR Pickard, Caputi, and Grenyer (2016) 151 undergraduate students 21.3 INFLEX 2
270 PR Galhardo, Cunha, and Pinto-Gouveia
(2011)
400 partners with and without infertility problems
(dyadic)
35.5 SCX, INFLEX 8
271 PR Pratscher, Wood, King, and
Bettencourt (2019)
141 undergraduates in relationships (≥3mo) 19.0 PMA, SCX, ACPT, LPMA 12
272 MT Williams (2010) 51 individuals with chronic pain condition and
their partners
58.3 LPMA 2
273 PR Evans, Whittingham, and Boyd
(2012)
127 mothers of preterm children 32.5 INFLEX 1
274 PR Laurent, Hertz, Nelson, and Laurent
(2016)
176 students in heterosexual relationships (dyadic) 21.3 ACPT 2
275 PR Crosswell, Coccia, and Epel (2019) 183 mothers of children 2-16yo 43.0 LPMA 1
276 PR Bell and Higgins (2015) 232 women from Midwestern community with
history of IPV
32.0 INFLEX 1 1
277 PR van Schoors et al. (2019) 123 parents of children with leukemia or non-
Hodgkin lymphoma
38.7 INFLEX 2
NOTE: Study code PR =peer-reviewed article, BC =book chapter, DS =unpublished dissertation, MT =unpublished masters thesis, HT =unpublished honors thesis.
Flexibility/inexibility dimension PMA =present moment awareness, SCX =self as context, DFN =defusion, ACPT =acceptance, FLEX =global exibility, LPMA =
lack of present moment awareness, SCN =self as content, INFLEX =global inexibility. Long =the study included longitudinal effects that could be extracted for
analysis (other studies had longitudinal data but failed to include the necessary effects). (dyadic) =both co-parents or romantic partners provided data.
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
224
potential importance of psychological exibility within families, sug-
gesting the possibility that exibility might have its most immediate
impact on various domains of family functioning by helping parents
respond in more kind and compassionate ways to their children’s
misbehavior.
Maladaptive parenting. Global inexibility was also predictive of
lax or inconsistent parenting, harsh parenting, overreactive parenting,
and other common negative parenting practices. These results suggest
that routinely engaging in more rigid, inexible and avoidant responses
toward difcult thoughts, feelings, and experiences might predispose
parents to engage in more reactive, volatile, and negative forms of
parenting. In contrast, acceptance, attentive awareness, self-as-context,
and global exibility were predictive of lower levels of lax, harsh, and
negative parenting. Although these results are based on cross-sectional
correlations, leaving the directions of causality unclear, these results
begin to suggest the possibility that parents who tended to respond to
difcult or challenging feelings and experiences in a more mindful,
accepting, and exible manner and who were able to maintain a deeper
perspective even in the heat of the moment were potentially more able
to decenter from difcult moments with their children, creating the
emotional space they needed to refrain from reactive parenting in
response to challenges or conict.
Child externalizing and internalizing symptoms. Consistent with
Family Systems Theory and Hypothesis 3, psychological inexibility
within parents not only impacted the parent-child interactions (i.e.,
parenting behaviors) but also children’s individual functioning. Thus,
parents generally being inattentive and distracted, judging and shaming
themselves for difcult feelings or experiences, and generally reacting to
difcult experiences in a rigid and inexible manner was linked to
higher levels of both internalizing and externalizing symptoms for the
children within those families. In contrast, parents abilities to be
attentively aware and accepting of difcult feelings and experiences was
associated with fewer child internalizing and externalizing symptoms,
suggesting that parental psychological exibility might help to buffer
families and children from the adverse effects of challenging or difcult
experiences.
6.3. Links to romantic relationship functioning
Relationship satisfaction. As shown in Table 4, one of the most
widely studied romantic relationship correlates of psychological exi-
bility was relationship satisfaction. Fig. 4 presents forest plots for some
of these meta-analytic effects. Consistent with Hypothesis 3, a lack of
mindful contact with the present moment (Fig. 4A) and generally
responding inexibly to difcult thoughts, feelings and experiences
(Fig. 4B) were predictive of lower levels of current relationship satis-
faction as well as predicting drops in satisfaction over time (across 6
months on average). Those forms of psychological inexibility also
predicted lower levels of relationship satisfaction in romantic partners.
Similarly, self as content (i.e., judging or shaming difcult feelings/ex-
periences) was associated with lower current relationship satisfaction
(Fig. 4D). Taken together, these results highlight that employing
inexible and rigid responses to the daily stressors that occur within all
romantic relationships could likely take a toll on those relationships,
possibly eroding their overall quality. Alternatively, as most of those
effects were cross-sectional, these results could also suggest that rela-
tionship dissatisfaction and discord (i.e., low levels of relationship
satisfaction) might shape how individuals respond to daily stressors and
difculties, drawing them toward more rigid and inexible responses. In
contrast, being able to both gently experience (i.e., defusion; Fig. 4C)
and nd internal compassion for and acceptance of difcult or chal-
lenging thoughts, feelings, and experiences was linked to higher levels of
current relationship satisfaction, highlighting the potential benets of
psychological exibility for global relationship functioning.
Sexual satisfaction. Although global inexibility failed to demon-
strate signicant links to sexual satisfaction across the 6 studies
Table 2
Sample Characteristics.
Classication
category
Specic sub-types
Counts
of
Samples
Demographic
category
Statistic
presented
Descriptives
across
samples
(details)
Publication type Percentage
male
participants
(n =196)
Journal articles 142 Lowest 0% (n =60)
Book chapters 1 Highest 100% (n =21)
Dissertations 43 Weighted
mean
27%
Masters Theses 14 Percentage
Caucasian
participants
(n =152)
Honors theses 2 Lowest 0% (n =12)
Under review 1 Highest 100% (n =5)
Year of publication Weighted
mean
66%
2005 to 2009 11 Average ages (n =187)
2010 to 2014 54 Lowest 17.7
2015 to 2019 138 Highest 70.4
Conceptual
framework**
Weighted
mean
34.1
ACT - Hexaex 40 Average yearly
incomes
(n =79)
Mindfulness 173 Lowest $16,374
Type of
population**
Highest $151,351
Undergraduate
students
32 Weighted
mean
$49,137
Community 130 Percentage
married or
cohabiting
(n =119)
Clinical 48 Lowest 0% (n =2)
Study design** Highest 100% (n =13)
Yielding self-
report effects
201 Weighted
mean
65%
Yielding effects
with
observational
data
7 Average
relationship
lengths
(n =65)
Yielding cross-
sectional effects
201 Shortest 0.7 years
Yielding
longitudinal
effects
11 Longest 28.7 years
Domains of
outcomes
examined
Weighted
mean
7.9 years
Relationship
functioning
95 Average child
age
(n =88)
Family
functioning
101 Youngest 0 years (newborns)
Both rel and
family
functioning
7 Oldest 16 years
Domains of
exibility
examined**
Weighted
mean
9.39 years
Present moment
awareness
31 Percentage
boys
(n =63)
Self as context 25 Lowest 20%
Cognitive
defusion
37 Highest 100%
Acceptance 32 Weighted
mean
54%
Global exibility 49 SUMMARY
LPMA 89 Distinct (sub)
samples
203
Self as content 40 Unique
subjects
43,952
Global
inexibility
85 Distinct effects
extracted
840
NOTE: ** the corresponding subgroups are not mutually exclusive and therefore
the count across those subgroups will not total to 203.
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
225
providing effects, lack of present moment awareness was associated
with lower levels of sexual satisfaction. Thus, approaching life in a
distracted and inattentive manner demonstrated small but consistent
negative links to the dyadic process of sharing physical intimacy within
relationships, suggesting that this form of inexibility could potentially
interfere with the quality of intimate bonding.
Social support. Unexpectedly, the three dimensions of exibility/
inexibility that had been examined in conjunction with perceived
support (across at least 5 studies; Table 4) failed to demonstrate sig-
nicant links to perceived support from a romantic partner. Thus, lack of
present moment awareness, self-as-content, and defusion did not seem
to inuence the degree to which individuals saw their partners as
compassionate, validating, and emotionally supportive of them, poten-
tially suggesting an ability of individuals to distinguish their partners’
supportiveness in a manner separate from their own exible or inexible
responses to difcult situations. In contrast to this nding, lack of pre-
sent moment awareness was associated with individuals reporting lower
levels of being supportive, empathic, and compassionate toward their
partners, highlighting how psychological inexibility might still impact
this dyadic process by attenuating individuals’ own abilities to respond
to their partners’ emotional needs. Of course, this different pattern of
ndings in predicting one’s own and a partner’s support could also
result from individuals judging themselves more harshly than they do
their partners.
Negative conict and physical aggression. Lack of present
moment awareness was also linked to higher levels of negative conict
behavior (e.g., yelling, shouting, calling names, insulting) and common
couple violence (typically referring to more mild forms of common
couple violence involving behaviors like pushing, shoving, slapping,
throwing something at one’s partner). Similarly, self-as-content and
Table 3
Meta-analytic effects linking psychological exibility to family functioning (k ≥5).
Family Correlate k N R SE 95% CI Q I
2
SM
Rel-Prob
FP Asym Trim and Fill adj.est.
Flex/Inex dimension Z r k
Parenting stress
LPMA 14 1575 .357*** .037 [.285, .429] 26.0* 56.1 1.10 0.16
Self-as-content 7 380 .366*** .060 [.249, .484] 10.7 37.4 0.30 −1.26
Global Inexibility 15 1195 .482*** .045 [.395, .569] 92.4* 76.2 0.50 −2.01
Attentive awareness 13 2060 -.342*** .042 [-.424, −.260] 40.6* 73.9 0.93 1.47
Defusion 7 378 -.313*** .075 [-.459, −.167] 13.3* 57.0 0.89 −0.80
Acceptance 13 2060 -.406*** .037 [-.478, −.335] 39.0* 69.1 0.93 −0.30
Self-as-context 12 1666 -.386*** .046 [-.476, −.296] 43.2* 75.3 0.93 0.36
Global Flexibility 19 3114 -.491*** .039 [-.567, −.416] 107.9* 84.3 0.95 0.99
Family conict
Global Flexibility 8 1180 -.336*** .047 [-.427, −.244] 24.5* 65.9 0.90 1.33
Family cohesion
LPMA 7 2657 -.217*** .042 [-.299, −.134] 17.9* 75.9 0.89 0.60
Global Inexibility 8 2237 -.339*** .062 [-.461, −.217] 89.4* 90.0 0.91 −0.06
Global Flexibility 13 3461 .392*** .066 [.263, .520] 444.0* 96.1 0.31 −1.12
Authoritative parenting
Attentive awareness 5 2193 .419*** .035 [.350, .488] 15.2* 67.9 0.14 −1.67
Adaptive parenting
LPMA 6 1747 -.144* .056 [-.254, −.034] 26.9* 78.5 0.88 0.00
Global Inexibility 11 3975 -.281*** .048 [-.376, −.186] 84.0* 85.6 0.92 1.10
Attentive awareness 5 1931 .342*** .086 [.173, .510] 71.8* 92.7 0.59 0.00
Global Flexibility 8 4404 .610*** .052 [.508, .711] 325.0* 96.5 0.10 −1.54
Lax parenting
Global Inexibility 8 1734 .390*** .045 [.303, .477] 35.5* 77.3 0.10 −0.80
Attentive awareness 6 1950 -.230*** .033 [-.294, −.166] 8.1 41.5 0.88 2.80** -.265*** 9
Global Flexibility 7 1272 -.317*** .025 [-.367, −.268] 5.1 0.1 0.89 1.67
Harsh parenting
Global Inexibility 12 2587 .303*** .053 [.200, .407] 109.5* 88.0 1.01 −1.94†.318*** 13
Attentive awareness 10 3397 -.244*** .049 [-.340, −.147] 88.8* 87.6 0.92 0.40
Acceptance 7 1550 -.335*** .065 [-.462, −.209] 55.8* 86.0 0.89 0.25
Self-as-context 7 1550 -.376*** .084 [-.539, −.212] 100.2* 92.8 0.89 0.75
Global Flexibility 9 1838 -.409*** .062 [-.530, −.287] 95.0* 89.3 0.91 0.45
Negative parenting NOS
Global Inexibility 11 3449 .378*** .064 [.252, .504] 323.9* 3.0 0.53 −1.03
Global Flexibility 7 3487 -.421*** .093 [-.603, −.239] 273.9* 96.5 0.89 −1.52
Child Externalizing
LPMA 14 2524 .242*** .052 [.140, .344] 92.5* 84.9 0.50 0.38
Self-as-content 6 984 .251** .084 [.085, .416] 25.5* 80.2 0.01 4.52** .113 9
Global Inexibility 14 4440 .280*** .043 [.196, .364] 85.5* 84.9 0.46 −0.13
Attentive awareness 8 1433 -.162** .051 [-.261, −.062] 21.6* 61.4 0.90 −0.28
Defusion 5 397 -.063 .072 [-.204, .077] 6.2 39.2 0.86 0.65
Acceptance 7 846 -.266*** .032 [-.329, −.203] 2.4 0.0 0.89 −0.69
Self-as-context 5 422 -.109 .100 [-.305, .086] 14.5* 70.4 0.86 0.72
Global Flexibility 11 3955 -.259*** .042 [-.342, −.176] 69.3* 75.9 0.92 −0.38
Child Internalizing
LPMA 8 2242 .270*** .048 [.176, .365] 42.2* 79.5 1.03 −0.25
Global Inexibility 11 3976 .296*** .053 [.191, .400] 87.7* 89.0 0.40 0.11
Global Flexibility 9 4124 -.193*** .048 [-.286, −.100] 83.2* 84.9 0.91 −0.01
NOTE: SM Rel-Prob =the relative probability of nding a contradictory nding (i.e., non-signicant or in the opposite direction to a majority of the other effects for
that pair of variables) compared to nding a consistent nding (estimated using the selection method analyses of McShane et al., 2016). Numbers close to 1 suggest no
publication bias and numbers over 1 suggest a greater relative likelihood of non-signcant or counter-intuitive ndings getting published. FP Asym Z =Eggers test of
funnel plot asymmetry, another test of possible publication bias. Trim and ll methods were used to adjust the estimates of effects showing signicant asymmetry.
LPMA =Lack of present moment awareness. Signicant effects have been bolded for ease of interpretation. ***p <.0001 **p <.01 * p <.05.
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
226
global inexibility were linked to higher levels of negative conict,
suggesting that responding to stressors in rigid, inexible, and distracted
manners might predispose individuals to react to conict within re-
lationships in more reactive and aggressive manners. In contrast, defu-
sion was linked to lower levels of negative conict behaviors, suggesting
that individuals’ abilities to gently experience difcult thoughts, feel-
ings, and emotions might help them decenter from relationship conict,
potentially creating the space to be able to select more kind and
compassionate responses to conict instead of engaging in more reactive
and aggressive behaviors.
Attachment anxiety and avoidance. Lack of present moment
awareness, self-as-content, and global inexibility were all moderately
linked to higher levels of attachment anxiety and demonstrated slightly
weaker links to attachment avoidance (both typically assessed with the
ECR-R). Consistent with the Attachment System Activation model, this
begins to suggest that attachment insecurities might shape the skills
individuals engage in response to difcult or challenging thoughts,
feelings, and experiences, potentially predisposing individuals to
engage in more rigid and inexible skills (conceptualized as hyper-
activating or deactivating strategies within that theory; see bib_-
Shaver_and_Mikulincer_2002Shaver & Mikulincer, 2002). Consistent
with this, attachment anxiety was linked to lower levels of cognitive
defusion.
6.4. Broader summary of effects represented in the current literature
Table 5 presents the current state of the literature examining links
between the 8 dimensions of psychological exibility/inexibility and
the 25 specic correlates of family and relationship functioning exam-
ined in the current review. As can be seen in the table, 118 signicant
effects emerged (including the 58 signicant effects detailed in Tables 3
and 4) linking these 8 dimensions of exibility and inexibility to the
family and relationship processes examined. Across all of the effects
examined, 65 of the combinations of constructs have been examined
across only 2 to 4 studies and should therefore be interpreted as only
approximate estimations of true effects (see Hedges & Vevea, 1998).
Furthermore, 14 of the effects presented in Table 5 were examined
within a single study and therefore represent point estimates rather than
meta-analytic effects. Finally, 46 of the combinations of the variables
included have yet to be examined (and 6 dimensions of exibility and
inexibility had too few studies to support meta-analysis), highlighting
areas for future work.
Consistent with the Hexaex model, dimensions of exibility were
generally predictive of more adaptive interpersonal functioning and
lower levels of conict and maladaptive behaviors whereas dimensions
of inexibility were generally predictive of greater conict and mal-
adaptive processes. Within these broader patterns of results, unique
patterns of results emerged for specic dimensions of exibility and
inexibility. For example, although acceptance was signicantly linked
to lower child externalizing symptoms (r = − 0.27, k =7), cognitive
defusion failed to demonstrate a similar link (r = − 0.06, k =5). Simi-
larly, although lack of present moment awareness was linked to lower
levels of one’s own supportiveness (r = − 0.21, k =6), self-as-content
failed to demonstrate a similar link (r = − 0.10, k =3). These meta-
analytic results extend ndings with the MPFI, suggesting that the in-
dividual dimensions of exibility and inexibility within the Hexaex
model might represent distinct (yet related) processes, potentially of-
fering their own unique insights (i.e., unique predictive variance) within
models of parent, family, child, and romantic relationship functioning.
Table 5 also presents effects from the 203 (sub)samples examining
cross-sectional Actor-Partner Interdependence paths linking one’s own
exibility/inexibility to one’s own relationship satisfaction (i.e., actor
effects) and to one’s partner’s satisfaction (i.e., partner effects) in men
and women separately (see Kenny, Kashy, & Cook, 2006). Although
fewer than a dozen studies provided this level of specicity in their ef-
fects, the results suggest that one’s own lack of present moment
awareness and global inexibility were linked to lower levels of partner
relationship satisfaction.
Table 5 also presents effects on a novel area of research uncovered by
our comprehensive review. Specically, our review identied a small
but growing body of studies examining how retrospective reports of
maladaptive parenting in childhood (e.g., harsh parenting, emotional
abuse and neglect by parents - as reported by college students) were
linked to current levels of psychological exibility and inexibility in
Fig. 3. Selected forest plots of meta-analytic effects linking parenting stress to dimensions of exibility and inexibility.
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
227
those college students. As seen in Table 5, retrospective reports of
maladaptive parenting during childhood were linked to higher levels of
global inexibility and lower levels of global exibility as a young adult.
In addition, a single study (Fischer, Smout, & Delfabbro, 2016) linked
retrospective reports of maladaptive parenting during childhood to
higher current reports of lack of present moment awareness in those
adult children. Despite the biases associated with retrospective report-
ing over such long time-intervals, these results begin to highlight how
family dynamics could impact the development of children’s psycho-
logical exibility/inexibility. Consistent with this, in one of the few
studies to have examined such links prospectively, Williams, Ciarrochi,
and Heaven (2012) demonstrated that adolescents’ perceived drops in
their parents’ use of authoritarian (i.e., high control, low warmth)
parenting behaviors were predictive of greater adolescent exibility in
high school.
7. Discussion
Based upon the exhaustive literature search conducted across three
databases and 5006 records for the current review, the current structural
and meta-analytic review represents the most comprehensive and in-
depth effort to characterize links between psychological exibility and
multiple levels of family and relationship functioning in the current
literature. The study specically used Family Systems Theory (Min-
uchin, 1985) as a broader conceptual framework and the EVSA model
(Karney & Bradbury, 1995, Fig. 1C) as a more focused lens to charac-
terize how the components of the Hexaex model might interact with
various family and romantic relationship dynamics. This conceptual
framework allowed the current review to synthesize and integrate pre-
vious work, clarifying the numerous ways that psychological exibility
might impact interpersonal dynamics in close relationships. With 174
studies examining the various links, the ndings highlighted a solid
foundation of work conducted across three distinct literatures (i.e., ACT,
mindfulness, emotion regulation). The meta-analytic ndings robustly
supported our hypotheses and suggested that psychological exibility
might indeed represent a set of critical processes within families and
within children’s development, providing both parents and children not
only with key intrapersonal skills to help promote their individual
wellbeing but also with key interpersonal skills to promote more
accepting and kind interactions at all levels of family functioning.
7.1. Theoretical implications
Flexibility within Family Systems Theory. The current meta-
analytic results uncovered widespread links between various forms of
psychological exibility and multiple levels of family functioning,
Table 4
Meta-analytic effects linking psychological exibility to relationship functioning (k ≥5).
Relationship Correlate k N r SE 95% CI Q I
2
SM
Rel-Prob
FP Asym Trim and Fill adj.est.
Flex/Inex dimension Z r k
Own relationship satisfaction
LPMA 42 12,944 -.219*** .022 [-.262, −.175] 340.3* 83.2 0.98 0.02
Self-as-content 20 5362 -.185*** .029 [-.243, −.128] 66.4* 74.8 0.95 1.11
Global Inexibility 34 6039 -.246*** .019 [-.283, −.209] 64.2* 51.1 0.97 3.14** -.277*** 43
Defusion 18 5179 .146*** .035 [.079, .214] 87.5* 81.1 2.68 −0.76
Acceptance 8 2389 .339*** .061 [.220, .457] 69.0* 87.2 3.60 −0.35
Own rel satisfaction over time
LPMA 6 1640 -.203** .061 [-.322, −.085] 23.2* 77.6 0.88 −0.22
Global Inexibility 5 737 -.199*** .036 [-.270, −.128] 4.5 4.5 0.86 1.53
Partner’s rel satisfaction
LPMA 9 984 -.158** .045 [-.246, −.070] 16.3* 50.5 0.91 −0.14
Global Inexibility 10 1334 -.159*** .027 [-.212, −.106] 11.2 2.3 0.92 0.49
Own sexual satisfaction
LPMA 6 3367 -.167*** .017 [-.200, −.134] 0.5 0.0 0.88 0.07
Global Inexibility 6 948 -.068 .102 [-.268, .132] 45.3* 87.9 0.25 0.39
Perceived partner support
LPMA 6 2706 -.114 .097 [-.305, .077] 97.6* 91.6 0.27 0.15
Self-as-content 5 423 -.045 .072 [-.186, .096] 8.5 53.7 0.86 −1.52
Defusion 5 423 .109 .093 [-.074, .291] 14.3* 73.6 2.95 1.42
Own supportiveness
LPMA 6 607 -.212*** .039 [-.287, −.136] 8.6 0.0 0.88 0.83
Own negative conict behv
LPMA 20 8093 .245*** .024 [.199, .291] 84.2* 74.6 1.68 −1.18
Self-as-content 7 1465 .262*** .058 [.149, .375] 29.2* 80.1 1.40 1.99* .262*** 7
Global Inexibility 9 1184 .216** .058 [.103, .329] 35.7* 76.3 1.14 −0.32
Defusion 5 1257 -.203*** .036 [-.274, −.133] 7.4 36.2 0.86 1.58
Own physical aggression
LPMA 7 1575 .196** .056 [.087, .305] 21.2* 78.3 0.11 3.72** .196** 7
Global Inexibility 6 767 .110 .095 [-.076, .295] 45.9* 85.3 0.15 −0.14
Own attachment anxiety
LPMA 20 6112 .356*** .012 [.333, .380] 23.1 6.2 0.95 −1.75
Self-as-content 6 1530 .422*** .022 [.379, .465] 4.2 5.0 0.13 −1.18
Global Inexibility 7 1681 .407*** .033 [.342, .471] 14.1.* 53.4 0.11 −0.08
Defusion 6 1530 -.243*** .064 [-.368, −.118] 25.1* 83.2 0.88 −0.39
Own attachment avoidance
LPMA 20 6112 .260*** .023 [.215, .305] 46.0* 62.9 4.04 −2.52* .263*** 21
Self-as-content 6 1530 .290*** .033 [.225, .355] 8.1 39.6 0.13 1.06
Global Inexibility 8 1784 .269** .075 [.122, .416] 56.3* 90.9 0.53 −0.49
Defusion 7 1566 -.093 .083 [-.256, .070] 32.4* 89.5 0.01 1.09
NOTE: SM Rel-Prob =the relative probability of nding a contradictory nding (i.e., non-signicant or in the opposite direction to a majority of the other effects for
that pair of variables) compared to nding a consistent nding (estimated using the selection method analyses of McShane et al., 2016). Numbers close to 1 suggest no
publication bias and numbers over 1 suggest a greater relative likelihood of non-signcant or counter-intuitive ndings getting published. FP Asym Z =Eggers test of
funnel plot asymmetry, another test of possible publication bias. Trim and ll methods were used to adjust the estimates of effects showing signicant asymmetry.
LPMA =Lack of present moment awareness. Signicant effects have been bolded for ease of interpretation. ***p <.0001 **p <.01 * p <.05.
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
228
including: (1) the global family environment (i.e., family cohesion,
family conict/chaos), (2) the dynamics of the co-parenting relationship
(i.e., relationship & sexual satisfaction, emotional support, negative
conict, aggression, attachment anxiety & avoidance), (3) parent-child
interactions (i.e., lax, harsh, negative, authoritative, & adaptive
parenting), (4) parents’ individual functioning (i.e., parenting stress/
burdens), and (5) child individual functioning (i.e., child internalizing &
externalizing behaviors). These results would therefore suggest that the
dimensions of psychological exibility and inexibility identied in the
Hexaex model might have central roles spanning all levels of family
functioning within Family Systems Theory (Minuchin, 1985), potenti-
ating either deeper levels of bonding and intimacy (i.e., dimensions of
exibility) or greater levels of conict and strife (i.e., dimensions of
inexibility). Thus, we see the Hexaex model as identifying funda-
mental processes operating at all levels of functioning with the Family
Systems Theory framework.
Although primarily used to organize classes of correlates in the
current meta-analysis, the Family Systems Framework could also be
used to help extend the current meta-analytic ndings and shape future
studies of parental exibility. Traditional family systems models might,
for example, build mediational models representing a top-down cascade
in which overarching family dynamics (e.g., family cohesion, family
conict) shape the romantic and co-parenting dynamics of the parents,
which then shape the types of parenting behavior used, thereby ulti-
mately shaping child and parent outcomes. Given the robust meta-
analytic links of parental exibility to all of the various family systems
examined, the current results would suggest that psychological exi-
bility and inexibility might best be represented as the fundamental
predictor variables within such a mediational model of family dynamics.
Placing exibility and inexibility to the far left of the resulting model
would allow them to be causally linked to all subsequent stages of the
process model (i.e., potentially exerting inuence on all levels of family
functioning) as one models how disruptions within one system within
the family can have ripple effects throughout the rest of the family
systems. Thus, by grounding the current meta-analysis within the
broader Family Systems Theory framework, we were able to move
beyond simply cataloging a diverse range of correlates, instead offering
readers a way of integrating those ndings into specic process models
that could be tested in future studies.
Flexibility within the EVSA model. The current meta-analytic re-
sults were also consistent with the EVSA model of interpersonal func-
tioning (Karney & Bradbury, 1995). In the context of this model, the
results begin to suggest that trait levels of exibility and inexibility
might represent enduring strengths and vulnerabilities that parents
bring into their roles within families and that state levels of exibility (as
they vary from day to day or context to context) might represent
essential individual adaptive processes with the power to shape the
dynamics of multiple relationships within the family system. Thus, di-
mensions of parental psychological exibility might serve as bonding
agents within families, promoting deeper emotional connections by
enabling parents to decenter (e.g., using defusion, self-as-context, and
acceptance) from stressful events (e.g., a child’s tantrum), creating
emotional space in they need to refrain from immediately engaging in
harsh or reactive responses (e.g., shouting at the child) and to instead
nd kinder and more compassionate responses to that difcult behavior
(e.g., using inductive parenting), thereby promoting greater parent,
child, and family wellbeing over time. As the EVSA was originally
developed in the couples literature, the ndings from the current project
Fig. 4. Selected forest plots of meta-analytic effects linking relationship satisfaction to dimensions of exibility and inexibility.
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
229
supported our modied EVSA model by demonstrating that psycholog-
ical exibility might help to shape the tone of dynamics within romantic
relationships. Thus, future studies could test mediation or even moder-
ated mediation models consistent with the modied EVSA presented in
Fig. 1C, thereby modeling exibility as a critical individual skill for
thriving romantic relationships. However, the current ndings extend
this one step further by also supporting the application of the EVSA
model to parent-child dynamics, and more broadly family dynamics. In
this manner, the current study has used Family Systems Theory to
contextualize the romantic relationship as a key system but just one of
many systems within the larger interpersonal context of the family.
Psychological exibility as critical intra- and inter-personal life-
skills. The current results extend research linking psychological exi-
bility to individual wellbeing in lab studies (e.g., Levin et al., 2012) and
highlighting exibility as a set of critical processes helping to explain
treatment gains in ACT (e.g., Fledderus et al., 2013). The current results
dovetail with that body of work to highlight that dimensions of psy-
chological exibility might also play critical roles in within close re-
lationships, potentially buffering those relationships from daily stressors
and conict that naturally arise and promoting more compassionate,
accepting, kind, and loving interactions within families. In contrast, the
current ndings suggest that engaging in forms of psychological
inexibility in response to stressful or uncomfortable events would
likely serve to exacerbate family conict, intensifying the impact of
daily stress on family wellbeing and potentially eroding the quality of
those interpersonal relationships. From this perspective, the domains of
psychological exibility and inexibility could be viewed as essential
life-skills spanning multiple domains of intrapersonal (i.e., within in-
dividuals) and interpersonal (i.e., between individual) functioning, in
much the same way that ACT is considered a transdiagnostic interven-
tion (e.g., Newby, McKinnon, Kuyken, Gilbody, & Dalgleish, 2015).
Thus, it is likely that dimensions of psychological exibility and
inexibility could meaningfully contribute to our understanding of dy-
namics in other domains of interpersonal functioning (e.g., in
caregiver-aging parent/partner dyads, in friendships, among siblings).
7.2. Psychological exibility in the family context
Learning exibility within the family: A majority of the work
examining dimensions of mindfulness (e.g., Carmody, Reed, Kristeller, &
Merriam, 2008) and psychological exibility (e.g., Forman, 2007) has
been conducted in adults and has focused on individual outcomes. The
current review builds on this work by highlighting that adults might
learn to engage in various psychologically exible or inexible skills at a
very young age. The current meta-analytic effects have linked retro-
spective reports of maladaptive parenting in childhood to higher levels
of inexibility in college students. Consistent with this, analyses in a
sample of 685 mother-adolescent dyads uncovered cross-sectional links
between parental exibility and adolescent exibility (Moreira & Can-
avarro, 2018). Similarly, a recent study of 749 school-aged students
from Australia demonstrated prospective links between parenting styles
and the development of psychological exibility in high school
Table 5
Summary of all individual and meta-analytic effects linking psychological exibility to relationship functioning.
Interpersonal Domain LPMA Self as
Content
Global
Inexibility
Present
Moment
Awareness
Cognitive
Defusion
Acceptance Self as
Context
Global
Flexibility
Specic Correlates est k est k Est k est k est k est k est k est k
Family Functioning
Family cohesion -.217 7 -.163 3 -.339 8 .265 2 .175 3 .231 2 .290 1 .392 13
Family conict .321 2 .283 3 .254 3 -.238 4 -.144 3 -.388 4 -.237 3 -.336 8
Parent Stress .357 14 .366 7 .482 15 -.342 13 -.313 7 -.406 13 -.386 12 -.491 19
Lax parenting .172 2 .390 8 -.265 6
(9)
-.250 4 -.250 4 -.317 7
Harsh parenting .155 4 .075 2 .318 12
(13)
-.244 10 -.335 7 -.376 7 -.409 9
Negative parenting NOS .292 3 .310 1 .378 11 -.306 4 -.090 1 -.399 4 -.511 4 -.421 7
Authoritative parenting -.294 2 .419 5 .392 3 .360 3 .558 4
Adaptive parenting -.144 6 -.075 3 -.281 11 .342 5 .229 2 .382 3 .413 3 .610 8
Child Externalizing .242 14 .113 6
(9)
.280 14 -.162 8 -.063 5 -.266 7 -.109 5 -.259 11
Child Internalizing .270 8 .235 3 .296 11 -.030 3 -.346 2 -.014 2 .210 1 -.193 9
Maladaptive parenting in childhood to
current ex/inex
.300 1 .239 8 -.171 4
Romantic Relationship Functioning
Own relationship satisfaction -.219 42 -.185 20 -.277 34
(43)
.171 4 .146 18 .339 8 .183 4 .307 3
Own rel sat MALES -.173 5 -.085 2 -.271 7 .188 2 .110 1
Own rel sat FEMALES -.155 6 -.180 2 -.249 7 (8) .043 2 .270 1
Own rel sat across time -.203 6 -.110 2 -.199 5 .008 1 -.030 2 .420 1
Partner’s relationship satisfaction -.158 9 -.028 2 -.159 10 .254 2 .440 1
M-ex to F-satisfaction -.236 2 -.224 3
F-ex to M-satisfaction -.095 2 -.162 3
Own sexual satisfaction -.167 6 -.170 3 -.068 6 .010 1 .073 3 .197 2
Perceived partner support -.114 6 -.045 5 -.260 1 .470 1 .109 5
Own supportiveness -.212 6 -.096 3 -.125 3 .042 4
Own physical aggression .196 7 .080 3 .110 6 -.131 3
Own negative conict behavior .245 20 .262 7 .216 9 -.270 1 -.203 5 -.256 2
Own attachment anxiety .356 20 .422 6 .407 7 -.231 2 -.243 6 -.299 2 -.303 2 -.454 2
Own attachment avoidance .263 20
(21)
.290 6 .269 8 -.191 2 -.093 7 -.262 2 -.274 2 -.390 2
NOTE: LPMA =Lack of present moment awareness. k values in parentheses are the adjusted values resulting from trim and ll analyses following signicant Egger’s
tests of funnel plot asymmetry. M-ex to F-satisfaction =cross-partner effects in which male exibility/inexibility predicted satisfaction in their female partners. F-
ex to M-satisfaction =cross-partner effects in which female exibility/inexibility predicted satisfaction in their male partners. Maladaptive parenting in childhood
predicting current ex/inex =Retrospective reports of college students on the quality of parenting during their childhoods predicting their current levels of exi-
bility/inexibility. Signicant effects have been bolded for ease of interpretation.
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
230
(Williams et al., 2012). Taken together, these transgenerational results
begin to suggest that the early childhood environment might serve as a
crucible in which children learn how to exibly or inexibly handle
their own difcult thoughts, feelings, and experiences, learning associ-
ated interpersonal skills and forming behavioral patterns that will likely
serve as the templates for their adult close relationships.
Promoting exibility in children. Given that the developmental
process of learning exible and inexible skills could start in early
childhood, interventions targeting exibility in children could poten-
tially be started as early as 2–4 years of age. For example, a self-
compassion and mindfulness-based intervention was developed and
tested in preschoolers, and demonstrated signicant reductions in
attention problems, aggression, and emotional reactivity (Garrison,
2017). Although this area of work is still at its earliest stages, these re-
sults reveal the utility of teaching and modeling psychological exibility
even before children start kindergarten, highlighting families as a key
point of intervention to be explored in future work.
Promoting exibility in parents. As parental exibility was linked
to family and child functioning across all family systems examined, the
current meta-analytic results would also support the utility of devel-
oping ACT-based parenting interventions. Consistent with this, an ACT-
based online parenting intervention was successful in reducing psy-
chological distress and burnout among parents of children with chronic
conditions by promoting defusion and present moment awareness
(Sairanen et al., 2019, 2020). Similar ACT-based parenting interventions
have shown benets for parents of children with Autism Spectrum
Disorder (Blackledge & Hayes, 2006; Lunsky, Fung, Lake, Steel, & Bryce,
2018), Cerebral Palsy (Whittingham, Sanders, McKinlay, & Boyd, 2016),
and chronic pain (Wallace, Woodford, & Connelly, 2016). The current
ndings highlight that in helping parents to engage more exibility
skills in the context of their families, such interventions would likely
have secondary benets, helping to buffer family conict and to pro-
mote family cohesion and child wellbeing by promoting adaptive
parenting. Thus, by intervening with parents, ACT-based parenting in-
terventions might help to transform the dynamics in families, thereby
instilling healthy psychological exibility skills in children as well.
Although much of the work in this area has focused on families con-
taining children with chronic or serious conditions, the current ndings
would suggest that ACT-based parenting interventions would likely be
benecial to a far broader range of families.
Promoting exibility in romantic relationships. The current re-
sults further highlighted the salience of psychological exibility within
romantic relationships, suggesting that couples might benet from in-
terventions promoting greater levels of exibility. Popular third-wave
couples interventions like Integrative Behavioral Couples Therapy
(IBCT; Jacobson & Christensen, 1996) contain similar foci to the Hex-
aex model (e.g., promoting acceptance and understanding), make use
of techniques that promote defusion and self-as-context (e.g., empathic
joining around the problem), and have been shown to be highly effective
(e.g., Jacobson, Christensen, Prince, Cordova, & Eldridge, 2000). More
recently, a novel self-guided intervention, the Promoting Awareness,
Improving Relationships or PAIR program, has been developed to
encourage couples to watch popular movies portraying relationship
dynamics and then use those movies as a method of easing into dis-
cussions of their own relationships by rst discussing the dynamics of
the couples on screen (Rogge, Fincham, Crasta, & Maniaci, 2016). Thus,
PAIR also seeks to promote awareness and connection through mecha-
nisms like defusion and self-as-context within the context of those
movie-based discussions, and has been shown to improve relationship
satisfaction in dating couples (Rogge et al., 2016) and to cut the rate of
divorce over the rst three years of marriage in newlywed couples
(Rogge, Cobb, Lawrence, Johnson, & Bradbury, 2013). Despite the
promise of interventions like ICBT and PAIR, these interventions are not
fully grounded in the Hexaex model and therefore do not directly
target psychological exibility to the same degree as ACT.
Only very recently have clinicians begun to examine ACT as a
possible method of treating relationship dissatisfaction and discord.
Consistent with the current ndings, an ACT-based group treatment for
couples demonstrated benets to relationship functioning and rela-
tionship satisfaction in a single arm treatment study (Ziapour et al.,
2017). Another ACT-based relationship education group intervention
for couples expecting their rst child demonstrated benets in a small,
mixed-methods randomized clinical trial (Gambrel & Piercy, 2015a;
2015b). ACT-based couples therapies delivered to individual couples
have shown similar benets in recent small-scale randomized clinical
trials (e.g., Omidi & Talighi, 2017; Veshki, Shaabady, Farzad, &
Fatehizade, 2017). Thus, a small growing group of studies have begun
examining forms of ACT as methods of strengthening relationships and
treating relationship discord. The current results would help to clarify
how promoting greater levels of psychological exibility within intimate
relationships could promote adaptive relationship processes and
long-term relationship satisfaction.
7.3. Future directions
Distinguishing exibility from inexibility. The current ndings
helped to clarify important distinctions between dimensions of psy-
chological exibility and inexibility. Although these dimensions show
low to moderate correlations with one another (Rolffs et al., 2016), the
various dimensions of psychological exibility and inexibility also
demonstrated distinct patterns of association with family and relation-
ship functioning in this meta-analysis. These differing patterns can be
traced back to the individual studies within the meta-analysis. For
example, Burke and Moore (2015) showed that parent exibility was
associated with slightly lower negative and inconsistent parenting be-
haviors whereas parent inexibility was more robustly linked to higher
negative parenting in a sample of parents of (pre)adolescents. In another
sample of 91 mothers of children with intellectual disabilities, higher
levels of mothers’ global exibility were linked to lower levels of
parental stress and maladaptive child behaviors whereas mothers’ lack
of present moment awareness was not predictive of those outcomes.
Distinct patterns also emerged for specic dimensions of exibility
within romantic relationships. For example, in a sample of 322 young
adult dating partners, Khaddouma, Gordon, and Bolden (2015b) found
that individuals who were consistently distracted and inattentive (i.e.,
high on lack of present moment awareness) and who tended to judge and
shame themselves for difcult thoughts and feelings (i.e., high
self-as-content) generally reported lower levels of relationship and
sexual satisfaction whereas defusion was not linked to either relation-
ship or sexual satisfaction in this sample. The current meta-analytic ef-
fects echoed these distinctive patterns, highlighting the conceptual
uniqueness (i.e., discriminant validity) of the individual dimensions of
exibility and inexibility. These results are also consistent with recent
ndings suggesting that dimensions of psychological exibility and
inexibility could best be conceptualized as distinct (yet related) con-
structs, as they: (1) show distinct patterns of association with mean-
ingful dimensions of interpersonal functioning, and (2) have the ability
to change independently from one another across time (e.g., Rogge
et al., 2019; Rolffs et al., 2016). Building on this larger body of work,
these ndings suggest that each of the dimensions of exibility and
inexibility within the Hexaex model will likely offer unique predic-
tive validity to models of relationship and family functioning. Thus,
certain forms of parental inexibility could more strongly linked to
specic forms of parenting or specic types of co-parenting or family
dynamics, potentially shaping them more directly. As a result, future
work would benet from distinguishing psychological exibility from
inexibility both at the measurement level (e.g., selecting scales that
separately assess those dimensions) and at the level of the models being
tested. At a clinical level, the results suggest that in working with cou-
ples or families, it would be helpful to know the specic ways in which
parents or partners are excessively rigid or inexible in their responses
to challenging feelings and experiences as well as they ways (and
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
231
contexts) in which they are able to respond more exibly.
Exploring the full Hexaex model. The current review uncovered
an uneven distribution of studies across the 12 possible dimensions of
the Hexaex model, with some dimensions (e.g., global inexibility
and lack of present moment awareness) being robustly represented in
the previous work whereas other dimensions (e.g., committed action,
contact with values, inaction, experiential avoidance, lack of contact
with values, and fusion) have barely been examined in family and
relationship contexts. A closer inspection of the 174 records revealed
that much of the previous work was driven heavily by self-report scale
development. Thus, as the elds of mindfulness and ACT research
adopted and popularized the use of scales like the AAQ-II (assessing
global inexibility), the MAAS (assessing lack of contact with the
present moment), the FFMQ (whose subscales assess defusion, self-as-
content, & lack of contact with the present moment), and the Inter-
personal Mindfulness in Parenting scale (IM-P; Duncan, 2007; with
subscales assessing acceptance, self-as-context, and present moment
awareness), those scales enabled researchers to examine specic di-
mensions of exibility and inexibility more readily than others. Given
the pervasive associations of the eight dimensions of exibility and
inexibility that could be meta-analyzed in this review, the current
ndings would suggest that future work could take advantage of
comprehensive scales like the MPFI (Rolffs et al., 2016) to examine
how all 12 dimensions of the Hexaex model might inform models of
relationship functioning, family functioning, and child development.
Consistent with this, both fusion and committed action have demon-
strated meaningful links with family cohesion, relationship satisfaction
(Rolffs et al., 2016), adaptive parenting (Moyer et al., 2018), children’s
retrospective reports of maladaptive parenting practices (Ahmed &
Bhutto, 2016; Fischer et al., 2016), and child distress (Wallace,
McCracken, Weiss, & Harbeck-Weber, 2015) in the handful of studies
that have examined them. Although based on too few studies to sup-
port meta-analysis, these results are consistent with (1) the current
ndings, (2) the EVSA model, and (3) Family Systems Theory. Thus,
these preliminary ndings suggest that valued living within parents
could not only promote individual wellbeing in those parents, but
could also help to improve the quality of family processes and dy-
namics throughout the family system, helping those parents select
parenting practices and responses to interpersonal conict that are
more in line with their deeper values on a day to day basis. Future work
could therefore extend the current ndings by examining all forms of
psychological exibility and inexibility as distinct mechanisms
within the EVSA and Family Systems models.
Examining exibility in families/relationships over time.
Although a robust body of work supported the current meta-analysis, the
vast majority of effects that could be extracted were cross-sectional in
nature. The smaller subset of studies that collected longitudinal data
tended to be treatment studies, so that the follow-up assessments and
analyses were focused on examining pre-post treatment effects of ACT-
based or mindfulness-based interventions. Given the meta-analytic
links uncovered between dimensions of exibility/inexibility and
both family and romantic relationship processes in the current review,
the current ndings begin to suggest that (beyond just representing
enduring strengths and vulnerabilities) psychological exibility and
inexibility could be considered as a set of interpersonal processes
themselves (i.e., as adaptive and maladaptive processes in the EVSA
model). Thus, when family conict arises or when a stressful external
event impinges on the family dynamics, the abilities of each of the
family members to engage that challenging or difcult experience in a
exible manner could be conceptualized as key processes that would
likely uctuate over time, helping to shape how the family collectively
navigates each of those stressful events. To explore psychological ex-
ibility and inexibility as a set of dynamic family processes, future work
could make use of multi-wave longitudinal designs, assessing exibility
alongside other common family processes (e.g., family conict,
parenting behaviors, co-parent cooperation). In one of the rst studies to
take such a longitudinal approach, Psychogiou et al. (2016) demon-
strated that global exibility in fathers was linked to residual drops in
child internalizing symptoms over 16 months. Given how rapidly
interpersonal dynamics can change within families, we would suggest
that these results could be meaningfully extended by examining longi-
tudinal links on a much shorter time-frame and with multiple waves of
assessment. Thus, future studies could consider collecting family diary
data on a daily, weekly, or monthly basis to fully model how exibility
might shape (and be shaped by) dynamics across various levels of the
family system.
Collecting data from multiple informants. The vast majority of
the studies reviewed here collected data from one individual (commonly
the primary caregiver) from each family or romantic relationship.
Although data from such designs provided support for the wide range of
meta-analytic effects estimated in the current review, restricting the
information collected on each family or in each romantic relationship to
a single individual limits our ability to fully model the dynamics within
those relationships, particularly the interdependent nature of those
interpersonal dynamics. Additionally, collecting information from just
one informant inherently raises issues related to reporting biases (e.g.,
limited awareness of one’s own behavior, social desirability, inuence of
mood and/or general dynamics within a couple or family). Thus, to fully
and more accurately explore how psychological exibility and inexi-
bility in individual family members might inuence both interpersonal
family dynamics and exibility/inexibility in other family members,
future work could collect data from multiple individuals in each family.
Expanding beyond self-report data. As mentioned above, much of
the work in this area seems to have been driven by the availability of key
self-report scales, allowing these constructs to be readily examined
within studies of romantic relationships and families. As a result, the
vast majority of the 840 effects extracted in the current meta-analysis
were based on self-report data. However, recent work has begun to
develop behavioral coding systems to assess mindful/exible parenting
(e.g., Benton, 2017; Geier, 2012), providing future researchers with
tools to assess exibility within families in a manner no longer con-
strained by the limitations of self-reports (e.g., response biases, lack of
insight). In addition, a handful of studies have begun examining how
aspects of exibility/mindfulness might shape physiological reactions to
conict (e.g., Laurent, Laurent, Hertz, Egan-Wright, & Granger, 2013;
Laurent, Laurent, Nelson, Wright, & Sanchez, 2015), tracking how as-
pects of mindfulness and exibility might help to shape cortisol reac-
tivity/recovery. Future work could continue to explore such links not
only in romantic couples but also in co-parenting dyads and parent-child
dyads to better characterize how psychological exibility might pro-
mote effective decentering (i.e., defusion, acceptance, self-as-context) at
a physiological level.
Flexibility & inexibility as moderators. The current ndings
support modeling the direct effects of psychological exibility and
inexibility within families and romantic relationships, potentially
modeling them as dynamic processes that uctuate and show mean-
ingful change over time. However, consistent with Relational Frame
Theory it would also be possible for future researchers to explore how
the various dimensions of psychological exibility and inexibility
might effectively change the underlying meaning or context of specic
family dynamics. Family Systems Theory posits that the different levels
of family functioning are interconnected so that conict in one area (e.
g., within the romantic dyad) could spill over into other areas of family
functioning (e.g., adversely affecting co-parenting efforts and child
behavior). Within that framework, psychological exibility could serve
as a buffer by helping to reduce negative spillover between levels of
family functioning whereas psychological inexibility might promote
greater amounts of negative spillover. Similarly, within the EVSA
framework, psychological exibility could both promote the use of more
adaptive processes and could help to buffer couples and families from
the adverse effects of daily stress whereas dimensions of inexibility
would likely exacerbate the effects of daily stress. Thus, future work
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
232
could explore such hypotheses to fully characterize how psychological
exibility and inexibility might fundamentally shift dynamics within
families.
8. Recommendations
1. Include a correlation matrix. A total of 67 records had collected
relevant data but could not be included in the current meta-analysis
as they did not present the corresponding correlations (nor did the
authors provide those correlations after being contacted). To prevent
that loss of information in the future, we recommend that researchers
consistently present study correlation matrices, particularly for
studies presenting results across multiple variables.
2. Distinguish exibility from inexibility. Results suggested that
dimensions of exibility and inexibility showed unique patterns of
association with various forms of interpersonal functioning. Thus,
whenever possible researchers are recommended to examine these as
separate constructs in their models or in their clinical work.
3. Explore the full Hexaex model. To thoroughly explore the impact
of exibility on family and relationship functioning, researchers are
recommended to use multidimensional scales like the MPFI (Rolffs
et al., 2016) and the Comprehensive Assessment of ACT Processes
(CompACT; Francis, Dawson, & Golijani-Moghaddam, 2016) to
assess and model more specic forms of psychological exibility and
inexibility.
4. Examine exibility in families/relationships over time. The vast
majority (99%) of effects in the current review were cross-sectional,
leaving directions of causality unclear. We would recommend
incorporating multi-wave longitudinal designs into future research
to support techniques like cross-lagged models, thereby allowing
researchers to quantify the varying strengths of the reciprocal links
between exibility and both family and romantic processes.
5. Collect data from multiple informants. As a majority of studies
collected data from one individual (in families, it was often the
mothers) we recommend collecting data from both partners in a
romantic relationship and/or from multiple individuals within each
family.
6. Expand beyond self-report data. We recommend extending this
work by including objective (e.g., behavioral coding of interaction
tasks, physiological) and indirect (e.g., priming tasks, word sorting
tasks) measures in future work to meaningfully deepen the assess-
ment of both interpersonal processes and psychological exibility.
7. Flexibility and inexibility as moderators. Recent work has
begun to highlight the moderating role of psychological exibility in
buffering the impact of anxiety on quality of life (e.g., Leonidou,
Panayiotou, Bati, & Karekla, 2019). To build on this growing body of
work, we recommend that researchers test the dimensions of psy-
chological exibility and inexibility as possible moderators in
models of relationship and family functioning.
8. Select a model to match your focus. The current review supported
both Family Systems Theory and the EVSA model, offering future
researchers a choice of frameworks in which to ground future
studies. Thus, if a researcher is trying to understand and model how
exibility might shape broader family dynamics spanning multiple
relationships, then that work would likely be most directly informed
by Family Systems theory. Such a framework would promote the use
of mediation models in which dynamics within one relationship or
family system might have ripple effects, spilling over into the dy-
namics of the rest of the relationships within the family. If instead, a
researcher is trying to understand a single relationship (be it
romantic, co-parenting, or even parent-child relationships), then that
work (and the resulting models) might be more directly informed by
the EVSA. Such a framework would encourage researchers to build
more moderated mediation models in which enduring vulnerabil-
ities, external stressors, internal processes and dyadic processes
might feed into one another and potentially interact to shape the
quality of that relationship across time.
9. Attachment theory as an alternative organizing framework. The
current ndings linked psychological exibility to attachment in-
securities. Future work could examine how attachment representa-
tions may shape an individual’s use of exible or inexible responses
to interpersonal experiences. Drawing upon the Attachment System
Activation Model (Shaver & Mikulincer, 2002), futures studies could
also examine how various forms of inexibility (e.g., experiential
avoidance, cognitive fusion) might predispose individuals to engage
in deactivating or hyperactivating behaviors in response to an
interpersonal threat activating their attachment systems whereas
exibility might serve to buffer individuals from getting triggered by
difcult or challenging experiences.
8.1. Limitations
Despite the breadth and depth of the current meta-analytic ndings,
the current results are limited by a number of factors. First, although we
dened our constructs of interest in a clear and targeted manner, we
chose to conduct a purposefully widespread and comprehensive litera-
ture search. As a result, our studies used a variety of measures across an
even greater variety of populations, introducing heterogeneity into our
estimates. Thus, although our meta-analytic effects are broadly gener-
alizable, they might not reect accurate point estimates of each effect in
highly specic subpopulations. Second, despite including data from 60
unpublished samples in the current review (30%), we cannot say that we
included every possible unpublished study. Although our statistical
analyses suggested only nominal levels of publication bias for the vast
majority of the meta-analytic effects, the effects presented should still be
interpreted with caution as they could have been slightly inated by
publication bias. Third, although 70 of our 137 meta-analytic effects
were based on data from 5 or more samples, nearly half of our meta-
analytic effects were based on only 2–4 samples and should therefore
be interpreted cautiously. Fourth, the vast majority of the meta-analytic
effects were based on cross-sectional data, and so the direction of cau-
sality remains unclear. For example, a small but growing body of studies
has begun to demonstrate the reciprocal impact that child functioning
and behavior can have on family dynamics across the various family
systems (e.g., Schermerhorn & Cummings, 2008), highlighting that
various processes within families are most likely transactionally linked
(i.e., exerting reciprocal causal inuence on one another across time),
thereby helping to explain the robust cross-sectional links observed
between family functioning and child behavior (e.g., Piotrowska, Stride,
Maughan, & Rowe, 2019). Thus, although the current model hypothe-
sized that parental inexibility would shape child behavior and out-
comes, it is also possible that child behavior could, in turn, inuence
parental exibility. Future work using cross-lagged models within
multi-wave longitudinal data (and possibly lab-based experimental
paradigms) is therefore needed to clarify directions of causality posited
within the current model. Lastly, the samples represented in the current
study were predominantly female, younger, and in heterosexual, dyadic
relationships. Future studies should aim to examine these associations
across diverse populations to further determine the generalizability of
the results. Despite these limitations, the current study offers the rst
truly comprehensive perspective on how psychological exibility might
inuence family and close-relationship dynamics, providing a concep-
tual and empirical foundation for future work.
Contributions
J. Daks and R. Rogge collaborated to develop the focus of the review
and outline the processes of screening and coding the studies. J. Daks
was responsible for the vast majority of the screening of records as well
as the data extraction and coding of the nal records in continuous
J.S. Daks and R.D. Rogge
Journal of Contextual Behavioral Science 18 (2020) 214–238
233
consultation with R. Rogge. R. Rogge double checked the extracted data.
J. Daks and R. Rogge collaborated on running the analyses, preparing
the tables and gures, and writing the manuscript. Both authors
approved the nal version of the manuscript for submission, and are
responsible for its content.
Author declaration
We wish to conrm that there are no known conicts of interest
associated with this publication and there has been no signicant
nancial support for this work that could have inuenced its outcome.
We conrm that the manuscript has been read and approved by all
named authors and that there are no other persons who satised the
criteria for authorship but are not listed.
We further conrm that the order of authors listed in the manuscript
has been approved by all of us.
We conrm that we have given due consideration to the protection of
intellectual property associated with this work and that there are no
impediments to publication, including the timing of publication, with
respect to intellectual property. In so doing we conrm that we have
followed the regulations of our institutions concerning intellectual
property.
We further conrm that any aspect of the work covered in this
manuscript that has involved human subjects has been conducted with
the ethical approval of all relevant bodies and that such approvals are
acknowledged within the manuscript.
We understand that the Corresponding Author is the sole contact for
the Editorial process (including Editorial Manager and direct commu-
nications with the ofce). He/she is responsible for communicating with
the other authors about progress, submissions of revisions and nal
approval of proofs.
We conrm that we have provided a current, correct email address
which is accessible by the Corresponding Author and which has been
congured to accept email from (rogge@psych.rochester.edu).
Signed by all authors as follows:
1st author: Jennifer S. Daks, 3/18/2019 2 nd & corresponding
author: Ronald D. Rogge, 3/18/2019.
Declaration of competing interest
None.
Acknowledgements
We would like to acknowledge the contributions of the researchers
who conducted studies contained in the 174 records included in this
review.
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.jcbs.2020.09.010.
References
*References marked with an asterisk indicate studies included in the meta-analysis.
A-Tjak, J. G. L., Davis, M. L., Morina, N., Powers, M. B., Smits, J. A. J., &
Emmelkamp, P. M. G. (2015). A meta-analysis of the efcacy of Acceptance and
Commitment Therapy for clinically relevant mental and physical health problems.
Psychotherapy and Psychosomatics, 84, 30–36. https://doi.org/10.1159/000365764
*Adair, K. C., Boulton, A. J., & Algoe, S. B. (2018). The effect of mindfulness on
relationship satisfaction via perceived responsiveness: Findings from a dyadic study
of heterosexual romantic partners. Mindfulness, 9, 597–609. https://doi.org/
10.1007/s12671-017-0801-3
*Ahmed, N., & Bhutto, Z. H. (2016). Relationship between parenting styles and self
compassion in young adults. Pakistan Journal of Psychological Research, 31, 441–451.
Alexander, K. (2018). Integrative review of the relationship between mindfulness-based
parenting interventions and depression symptoms in parents. Journal of Obstetric,
Gynecologic, and Neonatal Nursing, 47, 184–190. https://doi.org/10.1016/j.
jogn.2017.11.013
Amato, P. R., & Fowler, F. (2002). Parenting practices, child adjustment, and family
diversity. Journal of Marriage and Family, 64, 703–716. https://doi.org/10.1111/
j.1741-3737.2002.00703.x
*Asber, C. E. (2014). Qualities of mindfulness and relationship satisfaction: How mindfulness
training in one partner affects a relationship (Doctoral dissertation. San Diego,
California): Alliant International University. Retrieved from interlibrary loan
(Proquest UMI 3632190).
Baer, R. A., Smith, G. T., Hopkins, J., Krietemeyer, J., & Toney, L. (2006). Using self-
report assessment methods to explore facets of mindfulness. Assessment, 13, 27–45.
https://doi.org/10.1177/1073191105283504
Baer, R. A., Smith, G. T., Lykins, E., Button, D., Krietemeyer, J., Sauer, S., et al. (2008).
Construct validity of the ve facet mindfulness questionnaire in meditating and
nonmeditating samples. Assessment, 15, 329–342. https://doi.org/10.1177/
1073191107313003
*Balvaneda. (2017). The interpersonal effects of mindfulness and social anxiety in romantic
relationships. Retrieved from. Boston, Massachusetts): Masters thesis, University of
Massachusetts Boston https://search.proquest.com/docview/1993162858.
*Barnes, S., Brown, K. W., Krusemark, E., Campbell, W. K., & Rogge, R. D. (2007). The
role of mindfulness in romantic relationship satisfaction and responses to
relationship stress. Journal of Marital and Family Therapy, 33, 482–500. https://doi.
org/10.1111/j.1752-0606.2007.00033.x
*Baugh, L. M., Cox, D. W., Young, R. A., & Kealy, D. (2019). Partner trust and childhood
emotional maltreatment: The mediating and moderating roles of maladaptive
schemas and psychological exibility. Journal of Contextual Behavioral Science, 12,
66–73. https://doi.org/10.1016/j.jcbs.2019.02.001
*Beer, M., Ward, L., & Moar, K. (2013). The relationship between mindful parenting and
distress in parents of children with an autism spectrum disorder. Mindfulness, 4,
102–112. https://doi.org/10.1007/s12671-012-0192-4
*Bell, K. M., & Higgins, L. (2015). The impact of childhood emotional abuse and
experiential avoidance on maladaptive problem solving and intimate partner
violence. Behavioral Sciences, 5, 154–175. https://doi.org/10.3390/bs5020154
*Benton, J. (2017). Examining the association between emotional availability and mindful
parenting. Retrieved from. Fort Collins, Colorado): Masters thesis, Colorado State
University https://search.proquest.com/docview/1916010618.
Blackledge, J. T., & Hayes, S. C. (2006). Using acceptance and commitment training in
the support of parents of children diagnosed with autism. Child & Family Behavior
Therapy, 28, 1–18. https://doi.org/10.1300/J019v28n01_01
*Bluth, K., & Wahler, R. G. (2011a). Does effort matter in mindful parenting?
Mindfulness, 2, 175–178. https://doi.org/10.1007/s12671-011-0056-3
*Bluth, K., & Wahler, R. G. (2011b). Parenting preschoolers: Can mindfulness help?
Mindfulness, 2, 282–285. https://doi.org/10.1007/s12671-011-0071-4
Bond, F. W., Baer, R. A., Hayes, S. C., & Waltz, T. (2011). Preliminary psychometric
properties of the acceptance and action questionnaire – II: A revised measure of
psychological inexibility and experiential avoidance. Behavior Therapy, 42,
676–688. https://doi.org/10.1016/j.beth.2011.03.007
Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2009). Introduction to
meta-analysis. United Kingdon: John Wiley & Sons Ltd.
*Brassell, A. A., Rosenberg, E., Parent, J., Rough, J. N., Fondacaro, K., & Seehuus, M.
(2016). Parent’s psychological exibility: Associations with parenting and child
psychosocial well-being. Journal of Contextual Behavioral Science, 5, 111–120.
https://doi.org/10.1016/j.jcbs.2016.03.001 n
*Brem, M. J., Khaddouma, A., Elmquist, J., Florimbio, A. R., Shorey, R. C., & Stuart, G. L.
(2019). Relationships among dispositional mindfulness, distress tolerance, and
women’s dating violence perpetration: A path analysis. Journal of Interpersonal
Violence, 34, 2996–3016. https://doi.org/10.1177/0886260516664317
*Brem, M. J., Wolford-Clevenger, C., Zapor, H., Elmquist, J., Shorey, R. C., & Stuart, G. L.
(2018). Dispositional mindfulness as a moderator of the relationship between
perceived partner indelity and women’s dating violence perpetration. Journal of
Interpersonal Violence, 33, 250–267. https://doi.org/10.1177/0886260515604415
*Brockman, C., Snyder, J., Gewirtz, A., Gird, S. R., Quattlebaum, J., Schmidt, N., et al.
(2016). Relationship of service members’ deployment trauma, PTSD symptoms, and
experiential avoidance to postdeployment family reengagement. Journal of Family
Psychology, 30, 52–62. https://doi.org/10.1037/fam0000152
Broderick, C. B. (1993). Understanding family process: Basics of family systems theory.
Thousand Oaks, CA: Sage.
*Brown, S. M., Bender, K. A., Bellamy, J. L., Garland, E. L., Dmitrieva, J., & Jenson, J. M.
(2018). A pilot randomized trial of a mindfulness-informed intervention for child
welfare-involved families. online rst Mindfulness, 1–16. https://doi.org/10.1007/
s12671-018-1025-x.
*Brownlee, A. V. (2016). Depression, childhood adverse experiences, and mentalizing as
predictors of attachment anxiety and avoidance (Doctoral dissertation. Retrieved from.
San Diego, California: Alliant International University https://search.proquest.com/
docview/1865307615.
Brown, K. W., & Ryan, R. M. (2003). The benets of being present: Mindfulness and its
role in psychological well-being. Journal of Personality and Social Psychology, 84,
822–848. https://doi.org/10.1037/0022-3514.84.4.822
*Brown, F. L., Whittingham, K., & Sofronoff, K. (2015). Parental experiential avoidance
as a potential mechanism of change in a parenting intervention for parents of
children with pediatric acquired brain injury. Journal of Pediatric Psychology, 40,
464–474. https://doi.org/10.1093/jpepsy/jsu109
*de Bruin, E. I., Zijlstra, B. J. H., Geurtzen, N., van Zundert, R. M. P., van de Weijer-
Bergsma, E., Hartman, E. E., et al. (2014). Mindful parenting assessed further:
Psychometric properties of the Dutch version of the interpersonal mindfulness in
J.S. Daks and R.D. Rogge