Altering the Trajectory of Affect and Affect Regulation:
the Impact of Compassion Training
&Ihno A. Lee
&James R. Doty
James J. Gross
&Philippe R. Goldin
#Springer Science+Business Media, LLC 2017
Abstract A growing literature has begun to document the
effects of compassion training on a variety of important inter-
personal behaviors (e.g., helping behavior). What is not yet
well understood, however, is what impact compassion training
has on affect and affect regulation. To examine this issue, we
implemented a 9-week compassion training program in which
51 adults provided twice-daily ratings of four affective states
(anxiety, calm, fatigue, alertness) as well as their desire and
capability to regulate these affective states. In addition, partic-
ipants provided weekly responses regarding five specific reg-
ulatory strategies. Analysis of day-to-day trajectories of affec-
tive experience showed a decrease in anxiety and increase in
calmness. Day-to-day trajectories of affect regulation demon-
strated that participants were more likely to choose to accept
and thus not influence or modulate affective experience (as
opposed to dampen, enhance, or hold on to or maintain the
affective state). At the same time, participants also reported
being more capable in meeting their respective regulatory
goals. Finally, analysis of week-to-week trajectories of specif-
ic regulatory strategies over the course of the compassion
training program demonstrated that participants shifted to
lesser use of expressive suppression and greater acceptance
when experiencing stress/anxiety. These results suggest that
interventions such as compassion training may help modulate
specific affective states and modify the use of and self-efficacy
for specific regulatory strategies.
Keywords Affect .Affect regulation .Affective trajectories .
Affect dynamics .Self-efficacy .Compassion .Diary study .
Experience sampling .Ecological momentary assessment .
One commonly cited definition of compassion is a Bfeeling
that arises in witnessing another’s suffering and that motivates
a subsequent desire to help^(Goetz et al. 2010b,p.352).To
date, much of the research on compassion has focused on the
latter part of this definition—trait compassion has been shown
to predict things such as generosity (Saslow et al. 2013), vol-
unteerism (Omoto et al. 2009), and altruistic behavior (Batson
et al. 1999). Structured compassion training programs have
been shown to be effective in improving altruism and caring
behaviors (e.g., Jazaieri et al. 2016a;Wengetal.2013). Much
less is known about the association between compassion train-
ing and people’s affective responses, their attempts at regulat-
ing these affective responses, and their self-efficacy beliefs
(i.e., their beliefs about their ability to regulate these affective
Affect is an overarching term that includes states such as
stress, mood, and emotion (Gross and Thompson 2007).
Affective states unfold over time and vary in type, quality,
intensity, duration, and frequency (Gross and Thompson
2007). Such variations in affective states are consequential
for health and well-being and may be influenced both by an
individual’s attempts at regulating these affective states and by
his or her affect regulation self-efficacy.
Department of Psychology, Institute of Personality and Social
Research, University of California, Berkeley, 4152 Tolman Hall,
Berkeley, CA 94720-1650, USA
Center for Compassion and Altruism Research and Education, Palo
Alto, CA, USA
Department of Psychology, Stanford University, Stanford, CA, USA
School of Medicine, Stanford University, Stanford, CA, USA
Betty Irene Moore School of Nursing, University of California,
Davis, Sacramento, CA, USA
One method of examining affect trajectories has been daily
experience sampling. While daily experience sampling has
been utilized to examine affect in non-clinical populations
(e.g., Brans et al. 2013; Brose et al. 2015; Brown and Ryan
2003; Csikszentmihalyi and Hunter 2003; Goetz et al. 2010a;
Killingsworth and Gilbert 2010), most of these studies have
utilized relatively short-term daily assessments (e.g., ranging
from 2 to 21 days). Due to the relative ease of assessment,
most empirical studies with non-clinical populations have uti-
lized weekly assessments to examine affective trajectories
(e.g., Srivastava et al. 2009; Tamir et al. 2007).
Affect regulation refers to attempts made to influence what,
when, and how a person experiences various affective states
(e.g., Jazaieri et al. 2013b). One crucial determinant of affect
regulation is the goals people have at any given time (Gross
and Jazaieri 2014). For example, regulatory goals may include
increasing (enhancing), decreasing (dampening), keeping
around (maintaining), or accepting (embracing while not
modulating at all) one’s current affective state. A second im-
portant determinant of affect regulation is the specific strate-
gies that one chooses in order to achieve one’s regulatory
goal(s) (Gross and Jazaieri 2014).
Two of the most well-researched regulatory strategies are
cognitive reappraisal and expressive suppression (Gross
2002). Cognitive reappraisal, largely considered to be an
Badaptive^strategy (e.g., Gross 2002; John and Gross
2004), refers to efforts made to alter one’s affect by modifying
the subjective meaning of the situation. When employed ap-
propriately, cognitive reappraisal can modify one’sreactions
to affect-provoking situations and enhance psychological flex-
ibility and emotional well-being (Gross and Thompson 2007).
On the other hand, expressive suppression is mostly consid-
ered to be a Bmaladaptive^regulatory strategy (e.g., Gross
2002; John and Gross 2004) and refers to efforts made to alter
one’s own physiological, experiential, or behavioral responses
in a given situation. Expressive suppression is used once one
has experienced the onset of the affective experience. Across
affect, social functioning, and well-being, converging empiri-
cal research suggests that individuals who engage in cognitive
reappraisal seem to fare much better in life than those who rely
primarily on suppression (e.g., Gross 2002; John and Gross
The regulatory strategy of situation modification refers to
attempts to alter external (rather than internal) features of one’s
environment in an effort to influence one’s affect. Situation
modification can be used in anticipation of problematic affect
and before the full onset of the affective state. Situation mod-
ification is considered to be problem focused coping (Lazarus
and Folkman 1984) whereby affective states are modulated
via stimulus control. Another regulatory strategy is attentional
deployment, which refers to efforts to direct (or redirect) at-
tention in such a way as to alter one’s affective response. Two
of the most well-researched forms of attentional deployment
are distraction and rumination—both largely considered to be
maladaptive strategies (Gross 2014). Finally, acceptance has
recently been considered to be a skillful method for regulating
one’s affect (e.g., Gratz and Roemer 2004; Hayes et al. 1994,
1996;Linehan1993,2015). Rather than trying to change or
control one’s affect in some way, acceptance refers to allowing
one’s experience and affect to arise and change effortlessly of
their own accord. Empirical research has found that paradox-
ically, acceptance is quite effective in reducing negative affec-
tive states (Aldao et al. 2010; Hayes et al. 2012;Kengetal.
2016;Levittetal.2004). While some researchers conceptual-
ize acceptance as a form of cognitive reappraisal, we have
followed the conceptualization put forth by others (see
Aldao et al. 2010) who study mindfulness-based programs
and examine acceptance as a separate regulatory strategy, dis-
tinct from cognitive reappraisal.
Commonly, researchers focus on the frequency of use of
specific regulatory strategies such as those listed above
(Gross and John 2003). However, more recently, researchers
have become interested in examining regulatory self-
efficacy or capability, the belief one has in one’s abilities to
implement specific regulatory strategies (e.g., Goldin et al.
2009; Goldin et al. 2012a). Regulatory self-efficacy or capa-
bility has become an increasingly important construct to
measure and has been shown to be predictive of longer term
treatment outcomes (e.g., Goldin et al. 2012a). Self-efficacy
beliefs have also been linked to enhanced affect regulation
and psychosocial functioning more generally (e.g., Bandura
et al. 2003).
There are likely many ways of influencing affect trajecto-
ries, affect regulation, and affect regulation self-efficacy. The
most common approach to date has been to use formal clinical
treatment interventions with clinical samples (e.g., Aldao et al.
2014;Goldinetal.2014; Jazaieri et al. 2016b). However, such
treatments are often expensive and are generally limited to
individuals who meet criteria for specific psychopathology.
It is possible that sub-clinical or Bhealthy^individuals who
also experience challenging affective states might benefit from
enhancing skills to alter affective trajectories and enhance reg-
ulation and self-efficacy.
This has motivated researchers to explore novel interven-
tion programs, such as compassion training. Compassion is a
multidimensional construct that includes an affective compo-
nent (e.g., Goetz et al. 2010a;Jinpa2015; Jinpa and Weiss
2013). Some scholars have even conceptualized compassion
training as a form of cognitive reappraisal (e.g., Dahl et al.
2015; Engen and Singer 2015); however, only recently has
compassion as a state, trait, and intervention been researched
in Western science (e.g., Hutcherson et al. 2008; Klimecki
et al. 2012;Wengetal.2013). The foundation of compassion
training is mindfulness, or the ability to notice and pay atten-
tion (in the case of compassion, paying attention to suffering).
The empirical literature has established that mindfulness
trainings can influence affect and regulation in clinical (e.g.,
Goldin and Gross 2010;Goldinetal.2012b;Jazaierietal.
2016b) and non-clinical populations (e.g., Chambers et al.
2008; Jha et al. 2010); however, far less is known about affect
and affect regulation within the context of compassion train-
ing more generally.
To date, the majority of theoretical and empirical work on
compassion’s relationship to affect has focused on one specif-
ic component of compassion, for example, self-compassion
(e.g., Raes 2010; Van Dam et al. 2011). When considering
positive psychology interventions aimed at increasing positive
emotions, loving-kindness meditation (one specific compo-
nent or form of compassion practice where one cultivates the
wish of happiness for others) has been shown to be beneficial
for one’s affective state (Cohn and Fredrickson 2010;
Fredrickson et al. 2008). What has not yet been addressed,
however, is the impact of broader compassion training inter-
ventions on affective outcomes.
Our goal in the present study was to investigate affective
outcomes (trajectories, regulation, and self-efficacy) in CCT
within a community sample of adults. With regard to day-to-
day affective experience, we predicted reductions in anxiety
and fatigue and increases in calmness and alertness following
compassion training. We also predicted that compassion train-
ing would induce a greater desire to downregulate or dampen
negative affective states (anxiety and fatigue) and upregulate
or enhance positive affective states (calmness and alertness).
Relatedly, we predicted that participants would endorse great-
er capabilities in meeting all of their regulatory goals. With
regard to week-to-week trajectories of regulatory strategy use,
we predicted that participants would report decreases in situ-
ation modification and attentional deployment when
interacting with others and decreases in expressive suppres-
sion when experiencing stress/anxiety. Lastly, we predicted
that participants would report increases in cognitive reapprais-
al when interacting with others and increases in acceptance
when experiencing stress/anxiety.
Participants in this study were a subset (only those participants
randomized to immediate CCT as the participants in the
waitlist condition did not provide daily or weekly responses)
of a larger randomized controlled trial (Jazaieri et al. 2013a).
In this paper, we examine daily and weekly responses regard-
ing affective trajectories, regulation, and self-efficacy from the
51 participants who received the CCT intervention and had at
least a 50% response rate across all daily and weekly assess-
ment points over the 10-week period (9 out of the 60 partici-
pants randomized to immediate CCT did not meet this
criteria). A full CONSORT diagram is available in Jazaieri
et al. (2013a)(Fig.1). Participants in this study were primarily
middle aged (M(years) = 44.36, SD = 12.14), women (70.6%;
n= 36), and Caucasian (76.5%; n=39).
Potential participants were recruited through web-based on-
line community listings throughout the San Francisco Bay
Area, email listservs, and advertisements on community bul-
letin boards. Potential participants had to pass an initial online
screening procedure which excluded individuals who self-
endorsed bipolar disorder, major depressive disorder, psycho-
sis, or active suicidal ideation. Participants provided informed
consent in accordance with Stanford University Human
Subjects Committee rules and were not paid for their
Fig. 1 Screenshot example of iPhone question assessing anxiety, desire
to regulate anxiety, and capability of regulating anxiety
Compassion Cultivation Training
Compassion Cultivation Training (CCT) is a comprehensive
compassion training program with a dialectical focus on (a)
how one relates to suffering (that of oneself and others) and (b)
how one intentionally generates positive feelings for oneself
and others (loving-kindness). On the one hand, CCT develops
compassion, a mental capacity to regulate distress (e.g., stress
and anxiety) and maintain calm in response to suffering (for
oneself and others). On the other hand, CCT generates loving-
kindness, or the mental capacity to generate positive states
such as care, connectedness, and appreciation for oneself
and others. These two dialectics, compassion and loving-kind-
ness, are cultivated in an effort to more willingly and effec-
tively engage with any suffering that is present. Training of
present moment attention and the willingness to hold both
suffering and loving-kindness is intended to help people relate
to emotion and difficult experiences in a new way (Jazaieri
et al. 2014).
CCT is designed to be a structured, secular, and compre-
hensive, self- and other-focused compassion meditation train-
ing program. CCT consists of a 2-h introductory orientation,
eight once weekly 2-h classes, and daily compassion-focused
meditation practices. Specific content of the program includes
progressing through six sequential steps starting with the es-
sential and foundational skill of settling and focusing the mind
(attention training). Steps 2–5 pertain to the actual compassion
cultivation including loving-kindness and compassion for a
loved one (step 2), loving-kindness and compassion for one-
self (step 3), establishing the basis for compassion towards
others (step 4), cultivating compassion for others (step 5),
and lastly, active compassion practice (step 6). For a more
detailed description of the six steps in CCT, see Jinpa (2015)
and Jinpa and Weiss (2013). Each step in CCT builds upon the
prior, always beginning with settling and focusing the mind
(step 1). To integrate these practices into daily life, participants
are expected and encouraged to engage in daily home medi-
tation practice for at least 15 min (building up to approximate-
ly 30 min) using recorded guided meditations. The feasibility
of enhancing compassion through CCT has been established
(see Jazaieri et al. 2013a). When utilizing pre-to-post assess-
ments, CCT has been shown to produce significant increases
in trait mindfulness and happiness and decreases in worry,
stress, and suppression of emotional expression (Jazaieri
et al. 2014).
Daily Experience Sampling Participants were contacted at
fixed times twice daily throughout the duration of the course,
once in the morning and once in the evening (for additional
details, see Jazaieri et al. 2016a). To assess the affect trajecto-
ries of anxiety, calm, fatigue, and alert, participants responded
to the question: BHow [affective state] are you feeling right
now?^on a 7-point Likert scale ranging from 1 (not at all)to7
(very much). To assess affect regulation, participants
responded to the question: BWith regards to this feeling, do
you currently want to:^Bdampen it,^Bmaintain it,^Benhance
it,^or Bnot influence it at all.^To assess affect regulation self-
efficacy, participants were asked, with regard to the selected
strategy (dampen, maintain, enhance, or not influence), BDo
you feel capable of achieving this?^again rated on a 7-point
scale ranging from 1 (not at all) to 7 (very much). See Fig. 1
for an example of how these three questions appeared within
the context of anxiety.
Weekly Diaries To investigate weekly changes in affect reg-
ulation during CCT, we administered an inventory that our
laboratory has developed and have utilized in other studies
(e.g., Aldao et al. 2014;Goldinetal.2014). Participants were
trained on the use of the weekly inventory prior to the start of
the CCT program. During CCT, participants were emailed the
weekly questionnaire each week, on the day prior to their CCT
class. Instructions were to Bselect the percentage that indicates
how often you used a particular strategy toreduce any stress or
anxiety. 0% indicates that you used a particular strategy 0% of
the time and 100% indicates that you used a particularly strat-
egy 100% of the time^during the past week.
Specifically, to assess the regulatory strategy of situation
modification we asked, BHow often did you try to modify
your interactions with others (change the situation itself)?^
To assess attentional deployment we asked, BHow often did
you try to distract yourself during interactions with others?^
To assess cognitive reappraisal we asked, BHow often did you
try to change the way you were thinking or interpreting while
interacting with others?^To assess expressive suppression we
asked, BHow often did you try to hideany visible signs of your
stress or anxiety?^Finally, to assess acceptance we asked,
BHow often did you try to just accept your stress or anxiety
and not change it in any way?^In the weekly assessment
situation modification, attention deployment, and cognitive
reappraisal were pertaining to interactions with others while
expressive suppression and acceptance were pertaining spe-
cifically to states of stress or anxiety.
The daily data have a two-level structure, (diary ratings (level
1) nested within participant (level 2)), which is best analyzed
with multilevel modeling procedures. Univariate analyses ex-
amined daily changes in affect trajectories (feeling anxious,
calm, fatigued, alert), affect regulation (dampen, maintain,
enhance, or not influence), and affect regulation self-efficacy
(for anxiety, calm, fatigue, alertness). Separately, multivariate
analysis examined whether the changes in affect regulation
self-efficacy during CCT were related to changes in affect
trajectories as well as to changes in affect regulation for each
of the four affective states.
The weekly data have a similar two-level structure, with
weekly measures (level 1) nested within participant (level 2).
Univariate analysis examined the weekly changes in affect
regulation (situation modification, attentional deployment, re-
appraisal, suppression, acceptance).
Multilevel analyses were conducted using the following
approach: (a) To account for serial dependency in repeated
measures, we specified an autoregressive residual covariance
matrix. This specification ensures that each dependent vari-
able represents a change in relation to previous scores. (b) For
the daily analyses, the diary number (e.g., diary #3, diary #4)
was includedas a level-1 predictor to test for variation in affect
trajectories and affect regulation during CCT. For the weekly
analyses, week was included as a level-1 predictor. (c) All
intercepts and slopes were specified as random effects, with
covariances allowed among all random components. Our pri-
mary interest was in the covariance across time in (1) affect
regulation self-efficacy and affect trajectories and (2) affect
regulation self-efficacy and the desire to regulate affect.
The daily analyses were based on 112 ratings per person
collected over the CCT course. The weekly analyses were
based on ratings (assessed once per week) collected over the
CCT course. Missing values (daily, 23.4%; weekly, 2.9%)
were handled using full-information maximum likelihood
(FIML) estimation procedures, which generate unbiased pa-
rameter estimates and standard errors using all available ob-
servations (Enders 2001). On average, the percentage of miss-
ing data increased slightly from week 1 (17%) to week 8
(31%); however, no common or consistent patterns of day-
to-day Bmissingness^were found throughout CCT.
Univariate analyses were conducted using SAS PROC
MIXED (for continuous outcomes) and PROC GLIMMIX
(for binary outcomes); multivariate analyses were conducted
in Mplus v.6.1 (Muthén and Muthén 1998–2010). All predic-
tors were group-mean centered which standardizes within-
person responses and is typical for analyses targeting
All trajectories are displayed in Figs. 2,3,and4; estimates are
reported in Table 1. With regard to affect trajectories, during
CCT there were significant decreases in anxiety (γ=−.02,
p= .01) and increases in calmness (γ= .04, p< .01) with
significant variance across participants (p’s < .01). Fatigue
and alertness did not change significantly over time, but did
vary across individuals (p’s < .01).
With each successive rating, participants were less likely to
want to reduce their anxiety (OR=.989, 95% CI = (.983, .994))
and fatigue (OR = .987, 95% CI = (.981, .994)), as well as less
likely to want to enhance their feelings of calmness
(OR = .989, 95% CI = (.983, .995)) and alertness
(OR = .993, 95% CI = (.988, .998) (Fig. 3). Participants also
reported feeling more capable of regulating their affective
states of anxiety (γ=.03,p= .04), calmness (γ=.03,
p= .03), fatigue (γ=.03,p= .05), and alertness (γ=.03,
p= .04), with significant individual variability in these trajec-
tories (p’s < .01). While feelings of self-efficacy or capability
in regulating one’s affective states increased throughout CCT,
the desire to actually regulate affect (i.e., downregulate nega-
tive affect, upregulate positive affect) decreased (Fig. 4)—
hinting again at the paradoxical effects of acceptance within
the context of this comprehensive compassion training
During CCT, we found an inverse relationship between
reductions in anxiety and increases in regulation self-
efficacy of anxiety (r=−.68, p< .01). Similarly, increases in
calmness were related to increases in regulation self-efficacy
of calmness (r=.67,p< .01). These correlations did not differ
in size between anxiety and calmness (χ
(1) = .71, p>.05)
(significance was evaluated using the difference in −2log-
likelihoodvaluesobtained from a focal model and comparison
(nested) model, which is chi-square distributed). A similar
pattern of correlational results for fatigue and alertness and
their respective regulation self-efficacy were found.
However, given the non-significant trajectories in the fatigue
and alertness trajectories, these correlations are not as mean-
ingful and interpretable.
In addition, increases in regulation self-efficacy were relat-
ed to decreases in the desire to regulate affect. Specifically,
there were inverse associations between regulation self-
efficacy and desire to regulate for anxiety (r=−.51,
p< .01), calmness (r=−.71, p< .01), fatigue (r=−.55,
p< .01), and alertness (r=−.38, p<.05).
All affect regulation trajectories are displayed in Fig. 5;
estimates are reported in Table 1. Throughout the course of
the compassion intervention the use of suppression decreased
(γ=−.02, p= .002) and the use of acceptance increased
(γ= .02, p= .002) when experiencing stress/anxiety. The
Intensity (1 7)
Fig. 2 Day-to-day affect trajectories during CCT (reported as means)
frequency with which participants engaged in situation mod-
ification, attentional deployment, and cognitive reappraisal
during interactions with others did not change over time; how-
ever, these trajectories varied across individuals (p’s < .01).
The goal of this study was to investigate affect trajectories,
affect regulation, and affect regulation self-efficacy in a com-
munity sample of adults enrolled in a 9-week compassion
training program. We sought to extend the literature in four
important ways. First, rather than focusing on one specific
aspect of compassion (e.g., self-compassion or loving-kind-
ness), we utilized a 9-week comprehensive compassion med-
itation training program (CCT) that has been shown to
enhance multiple forms of compassion (Jazaieri et al.
2013a). Second, we sought to broaden the examination of
affect regulation beyond cognitive reappraisal and expressive
suppression and also consider the regulatory strategies of
situation modification, attentional deployment, and accep-
tance. Third, we sought to utilize more nuanced methodolog-
ical approaches (daily and weekly reports) to examine the
stability and change of affect and regulation. Finally, we
sought to add to the literature by examining these affective
outcome trajectories over an extended period of time (9-
In partial support of our hypothesis with regards to day-to-
day experiences of affect trajectories, we found reductions in
anxiety and increase in calmness trajectories over the course
of the intervention. There were no changes in the states of
fatigue or alertness over the course of CCT. When considering
day-to-day intention to regulate affect, contrary to our hypoth-
eses, participants were more likely to choose to not influence
their affective states (as opposed to down-regulate negative
states of anxiety and fatigue or upregulate positive states of
calmness and alertness). Participants also reported being more
capable in meeting their respective regulatory goals.
Interestingly, although we did not predict this, while partici-
pants reported being more capable at regulating affective
states, they simultaneously reported less desire to regulate.
Finally, when considering the week-to-week trajectories of
affect regulation over the course of the program, participants
reported decreases in expressive suppression and increases in
acceptance of affective states (stress/anxiety). Contrary to our
hypotheses, there were no changes in situation modification,
attentional deployment, or cognitive reappraisal during inter-
actions with others.
While the absence of a comparison condition prohibits
making conclusive interpretations of these results, preliminary
evidence from this study suggests that this compassion train-
ing program may reduce anxiety and increase feelings of
calm, which may be related to increases in the capability to
regulate both of these states. Prior research has shown that
Freqeuency (%) of Wanting to Regulate
Fig. 3 Day-to-day desire to
regulate affect during CCT
(reported as frequency
Capability (1 7)
Affect Regulation Self Efficacy
Fig. 4 Day-to-day affect regulation self-efficacy during CCT (reported
compassion training can increase positive affective experi-
ences, even in response to witnessing others in distress
(Klimeckietal.2012). Furthermore, experimental studies
with expert practitioners of compassion meditation have
shown that compassion meditation primarily increased posi-
tive affect and that cognitive reappraisal primarily decreased
negative affect when viewing video clips of people in distress
(Engen and Singer 2015). While not the goal of compassion
Table 1 Fixed and random
effects from the multilevel
analyses of affect trajectories,
affect regulation, and affect
Fixed effects Random effects
) Slope (γ
) Intercept (τ
Anxiety 3.096** −.022* .403** .0024**
Calmness 4.532** .039** .567** .0036**
Fatigue 3.759** .011 .620** .0041**
Alertness 4.486** −.002 .479** .0030**
Reduce anxiety −.380 −.011** 1.774** .0003
Enhance calmness −.524
−.011** 3.216** .0003**
Reduce fatigue −.598* −.013** 2.355** .0004
Enhance alertness −1.324** −.007** 2.941** .0002
Affect regulation self-efficacy
Anxiety 4.973** .025* 1.091** .0060**
Calmness 4.950** .029* 1.082** .0075**
Fatigue 4.898** .028* 1.344** .0091**
Alertness 5.051** .027* 1.051** .0072**
Wee kl y
Situation modification .299** −.006 .036** .0010**
Attentional deployment .383** −.008 .047** .0008**
Cognitive reappraisal .259** −.001 .024** .0006**
Expressive suppression .408** −.017** .040** .0004
Acceptance .343** .017** .018** .0007*
†=p<.06,*=p<.05,**=p< .01. Unstandardized estimates are reported
Slope parameters were generated from rescaled outcome variables (multiplied by 10) to facilitate reporting of
estimates. Rescaling does not affect statistical significance
Outcomes are binary, so the fixed effect estimates reflect log-odds ratios. Traditional odds ratios (e
in the text
Frequency of Regulation (%)
Fig. 5 Week-to-week affect
regulation during CCT (reported
as frequency percentages)
practice, the literature suggests that compassion meditation
(and specific subtypes such as loving-kindness meditation)
is associated with increases in positive affect and decreases
in negative affect (for reviews, see Hofmann et al. 2011; Zeng
et al. 2015). The current study provides some preliminary
evidence that calm, a low arousal positive state that is consid-
ered to be a valued state (Koopmann-Holm et al. 2013), may
be able to be increased through compassion training. A ran-
domized controlled trial with a comparison condition will be
able to more definitively address the role of compassion train-
ing on the affective state of calm. Although not tested here, it
is possible that these increases in calm may be related to de-
creases in personal distress, which in turn may be linked to
increased regulatory self-efficacy. This is an important area for
Preliminary findings from this study suggest that CCT may
result in increased abilities to regulate affective experience
while also shifting towards choosing to not influence affective
states in any way—perhaps suggesting that participants in this
compassion training may be more willing to accept one’saf-
fective states (both positive and negative) without choosing to
influence them in any way. This increase in self-efficacy or
perceived capability to regulate affective states may have led
to a greater willingness to experience and/or acceptance of
different affective states, with concomitant decreases in the
need to implement any form of regulation. These intriguing
findings highlight the potential role of affect regulation self-
efficacy in promoting acceptance of affect experience. Given
that CCT is a comprehensive compassion training program
with a dialectical focus on how one relates to suffering (i.e.,
recognizing and approaching, rather than avoiding, suffering)
and also loving-kindness (a state of generating positive emo-
tions for oneself and others), it is reasonable to expect an
increase in acceptance of both positive and negative states.
While this increase in one’s self-efficacy in regulating affect
would likely yield fewer failed regulatory experiences, it is
also possible that people are more efficacious in their regula-
tory attempts which may lead to fewer repeated attempts of
regulation. In the present study, it is difficult to assess the
directionality of the actual regulatory efficacy and desire to
regulate affective states—future research employing a true
experimental design will be able to address the causal nature
of this intriguing relationship.
These daily findings converge nicely with our findings
from the weekly assessments of specific emotion regulatory
strategies, which demonstrated decreases in suppression and
increases in acceptance of affective states (stress/anxiety).
These findings also parallel prior research of pre-post CCT
individual difference findings of emotion regulation which
utilized the Emotion Regulation Questionnaire (ERQ; Gross
and John 2003) and demonstrated a reduction in expressive
suppression, but no changes in general use of cognitive reap-
praisal following CCT when compared to the waitlist control
condition (Jazaieri et al. 2014). Taken together, these prelim-
inary daily and weekly data paired with the prior trait level
data from the randomized controlled trial suggests that on a
daily, weekly, and trait level, this type of compassion training
program may reduce expressive suppression of affect and in-
crease acceptance of positive and negative affective states.
Limitations and Future Research
The present investigation should be interpreted in the context
of several important limitations. This study examined a new
and comprehensive compassion training program. Without
utilizing a comparison condition in addition to CCT, it is dif-
ficult to understand to what extent compassion training influ-
enced daily and weekly affect and affect regulation. Future
research examining affect trajectories, affect regulation, and
affect regulation self-efficacy in compassion training would
benefit from utilizing an active control condition to fully un-
derstand the effects. In the daily experience sampling, we
examined four specific affective states—anxious, calm, fa-
tigued, and alert—that represent different combinations of va-
lence and arousal. We did not find effects on the states of alert
(a high arousal, positive state) or fatigue (a low arousal, neg-
ative state). Future research may choose to examine additional
combinations of high and low arousal and positive and nega-
tive affective states such as joy, awe, and boredom. Relatedly,
in our weekly diary, we were interested in expanding the focus
beyond cognitive reappraisal and expressive suppression to
examine additional forms of affect regulation including situa-
tion modification, attentional deployment, and acceptance. It
is only through detailed empirical investigation into additional
strategies that we can more fully understand the effectiveness
(or potential ineffectiveness) of different regulatory strategies.
Given that dozens of regulatory strategies exist (for a review,
see Webb et al. 2012), future research may choose to examine
additional strategies (e.g., situation selection; Gross 1998), or
subtypes of strategies (e.g., mindfulness as a form of
attentional deployment; Neacsiu et al. 2014). It will also be
important to use similar context frames across all strategies
(e.g., all related to stress/anxiety contexts or all items related to
social interactions with others) in order to address the gener-
alizability of improvements in regulatory strategy use. Finally,
this study relied on daily and weekly self-reports of affect and
regulation. Future studies of compassion training must move
beyond self-report to include implicit or behavioral measures,
observer reports (e.g., romantic partners, co-workers, peers),
and physiological markers.
Collectively, the findings from this study suggest that in-
terventions such as this comprehensive compassion training
program may help modulate specific affective states (anxiety
and calm), and modify one’s self-efficacy and use of various
affect regulation strategies. This study was novel in that it
utilized nuanced methodological approaches ranging from
daily experience sampling to weekly diaries, which allowed us
to look at trajectories of change, while also sampling a variety
of regulation strategies beyond cognitive reappraisal and ex-
pressive suppression. Taken together, this study provides valu-
able preliminary information about the stability and change
related to affective processes during a comprehensive compas-
sion training program.
Acknowledgements The authors would like to thank the handling ed-
itor as well as the two anonymous reviewers for their thoughtful com-
ments that strengthened this paper. We would also like to thank senior
CCT teachers Margaret Cullen and Erika Rosenberg for their contribu-
tions to the development of the CCT protocol and teaching of CCT.
Finally, we would like to thank Benjamin Olmsted for developing the
app used in this research.
Author Contributions HJ co-designed the study, executed the study,
assisted with data analyses, and wrote the manuscript. KM taught CCT
and assisted with editing of the manuscript. IL conducted data analyses
and assisted with writing and editing of the manuscript. TJ created CCT
and assisted with the editing of the manuscript. JD assisted with editing of
the manuscript. JG collaborated in the writing and editing of the manu-
script. PG co-designed the study and collaborated in the writing and
editing of the manuscript.
Compliance with Ethical Standards The research study was ap-
proved by the ethics committee at Stanford University, prior to participant
Conflict of Interest The authors declare that they have no conflicts of
Funding This research was supported by a Fetzer grant as well as
funding from Stanford’s Center for Compassion and Altruism Research
and Education (CCARE).
Human and Animal Rights All procedures performed in studies in-
volving human participants were in accordance with the ethical standards
of the institutional research committee and with the 1964 Helsinki decla-
ration and its later amendments.
Informed Consent Informed consent was obtained from all individual
participants prior to their inclusion in the study.
Aldao, A., Nolen-Hoeksema, S., & Schweizer, S. (2010). Emotion-
regulation strategies across psychopathology: a meta-analytic re-
view. Clinical Psychology Review, 30,217–237. doi:10.1016/j.cpr.
Aldao, A., Jazaieri, H., Goldin, P. R., & Gross, J. J. (2014). Adaptive and
maladaptive emotion regulation strategies: interactive effects during
CBT for social anxiety disorder. Journal of Anxiety Disorders, 28,
Bandura, A., Caprara, G. V., Barbaranelli, C., Gerbino, M., & Pastorelli,
C. (2003). Role of affective self-regulatory efficacy in diverse
spheres of psychosocial functioning. Child Development, 74,769–
Batson, C. D., Floyd, R. B., Meyer, J. M., & Winner, A. L. (1999). BAnd
who is my neighbor?:^intrinsic religion as a source of universal
compassion. Journal for the Scientific Study of Religion, 38,445–
Brans, K., Koval, P., Verduyn, P., Lim, Y. L., & Kuppens, P. (2013). The
regulation of negative and positive affect in daily life. Emotion, 13,
Brose, A., de Roover, K., Ceulemans, E., & Kuppens, P. (2015). Older
adults’affective experiences across 100 days are less variable and
less complex than younger adults'. Psychology and Aging, 30,194–
Brown, K. W., & Ryan, R. M. (2003). The benefits of being present:
mindfulness and its role in psychological well-being. Journal of
Personality and Social Psychology, 84,822–848. doi:10.1037/
Chambers, R., Lo, B., & Allen, N. B. (2008). The impact of intensive
mindfulness training on attentional control, cognitive style, and af-
fect. Cognitive Therapy and Research, 32,303–322. doi:10.1007/
Cohn, M. A., & Fredrickson, B. L. (2010). In search of durable positive
psychology interventions: predictors and consequences of long-term
positive behavior change. The Journal of Positive Psychology, 5,
Csikszentmihalyi, M., & Hunter, J. (2003). Happiness in everyday life:
the uses of experience sampling. Journal of Happiness Studies, 4,
Dahl, C. J., Lutz, A., & Davidson, R. J. (2015). Reconstructing and
deconstructing the self: cognitive mechanisms in meditation prac-
tice. Trends in Cognitive Scienences, 19,515–523. doi:10.1016/j.
Enders, C. K. (2001). The impact of nonnormality on full information
maximum-likelihood estimation for structural equation models with
missing data. Psychological Methods, 6, 352–370. doi:10.1037/
Engen, H. G., & Singer, T. (2015). Compassion-based emotion regulation
up-regulates experienced positive affect and associated neural net-
works. Social Cognitive and Affective Neuroscience, 10, 1291–
Fredrickson, B. L., Cohn, M., Coffey, K. A., Pek, J., & Finkel, S. A.
(2008). Open hearts build lives: positive emotions, induced through
loving-kindness meditation, build consequential personal resources.
Journal of Personality and Social Psychology, 95, 1045–1062. doi:
Goetz, T., Frenzel, A. C., Stoeger, H., & Hall, N. C. (2010a). Antecedents
of everyday positive emotions: an experience sampling analysis.
Motivation and Emotion, 34,49–62. doi:10.1007/s11031-009-
Goetz, J., Keltner, D., & Simon-Thomas, E. (2010b). Compassion: an
evolutionary analysis and empirical review. Psychological
Bulletin, 136,351–374. doi:10.1037/a0018807.
Goldin, P. R., & Gross, J. J. (2010). Effects of mindfulness-based stress
reduction (MBSR) on emotion regulation in social anxiety disorder.
Emotion, 10,83–91. doi:10.1037/a0018441.
Goldin, P. R., Manber, T., Hakimi, S., Canli, T., & Gross, J. J. (2009).
Neural bases of social anxiety disorder: emotional reactivity and
cognitive regulation during social and physical threat. Archives of
General Psychiatry, 66,170–180. doi:10.1001/archgenpsychiatry.
Goldin, P. R., Ziv, M., Jazaieri, H., Werner, K., Kraemer, H., Heimberg,
R. G., & Gross, J. J. (2012a). Cognitive reappraisal self-efficacy
mediates the effects of individual cognitive-behavioral therapy for
social anxiety disorder. Journal of Consulting and Clinical
Psychology, 80,1034–1040. doi:10.1037/a0028555.
Goldin, P. R., Ziv, M., Jazaieri, H., Hahn, K., & Gross, J. J. (2012b).
MBSR vs. aerobic exercise in social anxiety disorder: fMRI of emo-
tion regulation of negative self-beliefs. Social Cognitive and
Affective Neuroscience, 8,65–72. doi:10.1093/scan/nss054.
Goldin, P. R., Lee, I., Ziv, M., Jazaieri, H., Heimberg, R. G., & Gross, J. J.
(2014). Trajectories of change in emotion regulation and social anx-
iety during cognitive-behavioral therapy for social anxiety disorder.
Behaviour Research and Therapy, 56,7–15. doi:10.1016/j.brat.
Gratz, K. L., & Roemer, L. (2004). Multidimensional assessment ofemo-
tion regulation and dysregulation: development, factor structure, and
initial validation of the difficulties in emotion regulation scale.
Journal of Psychopathology and Behavioral Assessment, 26,41–
Gross, J. J. (1998). Antecedent-and response-focused emotion regulation:
divergent consequences for experience, expression, and physiology.
Journal of Personality and Social Psychology, 74, 224. doi:10.
Gross, J. J. (2002). Emotion regulation: affective, cognitive, and social
consequences. Psychophysiology, 39,281–291. doi:10.1017/
Gross, J. J. (2014). Emotion regulation: conceptual and empirical foun-
dations. In J. J. Gross (Ed.), Handbook of emotion regulation (2nd
ed.). New York: Guildford Press.
Gross, J. J., & Jazaieri, H. (2014). Emotion, emotion regulation, and
psychopathology: an affective science perspective. Clinical
Psychological Science, 2, 387–401. doi:10.1177/
Gross, J. J., & John, O. P. (2003). Individual differences in two emotion
regulation processes: Implications for affect, relationships, and well
being. Journal of Personality and Social Psychology, 85,248–362.
Gross, J. J., & Thompson, R. A. (2007). Emotion regulation: conceptual
foundations. In J. J. Gross (Ed.), Handbook of emotion regulation
(pp. 3–24). New York: Guilford Press.
Hayes, S. C., Jacobson, N. S., Follette, V. M., & Dougher, M. J. (1994).
Acceptance and change: content and context in psychotherapy.
Reno: Context Press.
Hayes, S. C., Wilson, K. G., Gifford, E. V., Follette, V. M., & Strosahl, K.
(1996). Experiential avoidance and behavioral disorders: a function-
al dimensional approach to diagnosis and treatment. Journal of
Consulting and Clinical Psychology, 64, 1152–1168.
Hayes, S. C., Bissett, R. T., Korn, Z., Zettle, R. D., Rosenfarb, I. S.,
Cooper, L. D., & Grundt, A. M. (2012). The impact of acceptance
versus control rationales on pain tolerance. The Psychological
Hofmann, S. G., Grossman, P., & Hinton, D. E. (2011). Loving-kindness
and compassion meditation: potential for psychological interven-
tions. Clinical Psychological Review, 31, 1126–1132. doi:10.1016/
Hutcherson, C. A., Seppala, E. M., & Gross, J. J. (2008). Loving-
kindness meditation increases social connectedness. Emotion, 8,
Jazaieri, H., Jinpa, T. L., McGonigal, K., Rosenberg, E., Finkelstein, J.,
Simon-Thomas, E., et al. (2013a). Enhancing compassion: a ran-
domized controlled trial of a compassion cultivation training pro-
gram. Journal of Happiness Studies, 14, 1113–1126. doi:10.1007/
Jazaieri, H., Urry, H. L., & Gross,J. J. (2013b). Affective disturbance and
psychopathology: an emotion regulation perspective. Journal of
Experimental Psychopathology, 4,584–599. doi:10.5127/jep.
Jazaieri,H.,McGonigal,K.M.,Jinpa,T.L., Doty, J. R., Gross, J. J., & Goldin,
P. R. (2014). A randomized controlled trial of compassion cultivation
training: effects on mindfulness, affect, and emotion regulation.
Motivation and Emotion, 38,23–35. doi:10.1007/s11031-013-9368-z.
Jazaieri, H., Lee, I. A., McGonigal, K., Jinpa, T., Doty, J. R., Gross, J. J.,
& Goldin, P. R. (2016a). A wandering mind is a less caring mind:
daily experience sampling during compassion meditation training.
The Journal of Positive Psychology, 11, 37–50. doi:10.1080/
Jazaieri,H., Lee, I. A., Goldin, P. R., & Gross, J. J. (2016b). Pre-treatment
social anxiety severity moderates the impact of mindfulness-based
stress reduction and aerobic exercise. Psychology and
Psychotherapy: Theory, Research and Practice, doi:10.1111/papt.
Jha, A. P., Stanley, E. A., Kiyonaga, A., Wong, L., & Gelfand, L. (2010).
Examining the protective effects of mindfulness training on working
memory capacity and affective experience in a military cohort.
Emotion, 10,54–64. doi:10.1037/a0018438.
Jinpa, T. L. (2015). A fearless heart: how the courage to be compassion-
ate can transform our lives. New York: Hudson Street Press.
Jinpa, T. L., & Weiss, L. (2013). Compassion cultivation training (CCT).
In T. Singer & M. Boltz (Eds.), Compassion: bridging practice and
science. Leipzig: Max Planck Institute for Human Cognitive and
John, O. P., & Gross, J. J. (2004). Healthy and unhealthy emotion regu-
lation: personality processes, individual differences, and life span
development. Journal of Personality, 72,1301–1334. doi:10.1111/
Keng, S. L., Smoski, M. J., & Robins, C. J. (2016). Effects of mindful
acceptance and reappraisal training on maladaptive beliefs about
rumination. Mindfulness, 7,493–503. doi:10.1007/s12671-015-
Killingsworth, M. A., & Gilbert, D. T. (2010). A wandering mind is an
unhappy mind. Science, 330, 932. doi:10.1126/science.1192439.
Klimecki, O. M., Leiberg, S., Lamm, C., & Singer, T. (2012). Functional
neural plasticity and associated changes in positive affect after com-
passion training. Cerebral Cortex, 23, 1552–1561. doi:10.1093/
Koopmann-Holm, B., Sze, J., Ochs, C., & Tsai, J. L. (2013). Buddhist-
inspired meditation increases the value of calm. Emotion, 13,497–
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping.New
Levitt, J. T., Brown, T. A., Orsillo, S. M., & Barlow, D. H. (2004). The
effects of acceptance versus suppression of emotion on subjective
and psychophysiological response to carbon dioxide challenge in
patients with panic disorder. Behavioral Therapy, 35, 747–766.
Linehan, M. M. (1993). Cognitive-behavioral treatment of borderline
personality disorder. New York: Guilford.
Linehan, M. M. (2015). DBT s kills tra ining m anual (2nd ed.). New York:
Muthén, L. K., & Muthén, B. O. (1998–2010). Mplus user’s guide. Los
Angeles: Muthén & Muthén.
Neacsiu, A. D., Bohus, M., & Linehan, M. M. (2014). Dialectical behav-
ior therapy skills: an intervention for emotion dysregulation. In J. J.
Gross (Ed.), Handbook of emotion regulation (2nd ed., pp. 491–
508). New York: Guilford.
Omoto, A. M., Malsch, A. M., & Barraza, J. A. (2009). Compassionate
acts: motivations for and correlates of volunteerism among older
adults. In B. Fehr, S. Sprecher, & L. G. Underwood (Eds.), The
science of compassionate love: theory, research, and applications
(pp. 257–282). Malden: Wiley–Blackwell.
Raes, F. (2010). Rumination and worry as mediators of the relationship
between self-compassion and depression and anxiety. Personality and
Individual Differences, 48, 757–761. doi:10.1016/j.paid.2010.01.023.
Saslow, L. R., Willer, R., Feinberg, M., Piff, P. K., Clark, K., Keltner, D.,
& Saturn, S. (2013). My brother’s keeper? Compassion predicts
generosity more among less religious individuals. Social
Psychological and Personality Science, 4,31–38. doi:10.1177/
Srivastava, S., Tamir, M., McGonigal, K. M., John, O. P., & Gross, J. J.
(2009). The social costs of emotional suppression: a prospective
study of the transition to college. Journal of Personality and Social
Psychology, 96,883–897. doi:10.1037/a0014755.
Tamir, M., John, O. P., Srivastava, S., & Gross, J. J. (2007). Implicit
theories of emotion: affective and social outcomes across a major
life transition. Journal of Personality and Social Psychology, 92,
Van Dam, N. T., Sheppard, S. C., Forsyth, J. P., & Earleywine, M. (2011).
Self-compassion is a better predictor than mindfulness of symptom
severity and quality of life in mixed anxiety and depression. Journal of
Anxiety Disorders, 25, 123–130. doi:10.1016/j.janxdis.2010.08.011.
Webb, T. L., Miles, E., & Sheeran, P. (2012). Dealing with feeling: a
meta-analysis of the effectiveness of strategies derived from the
process model of emotion regulation. Psychological Bulletin, 138,
Weng, H. Y., Fox, A. S., Shackman, A. J., Stodola, D. E., Caldwell, J. Z.
K., Olson, M. C., et al. (2013). Compassion training alters altruism
and neural responses to suffering. Psychological Science, 24, 1171–
Zeng, X., Chiu, C. P., Wang, R., Oei, T. P., & Leung, F. Y. (2015). The
effect of loving-kindness meditation on positive emotions: a meta-
analytic review. Frontiers in Psychology, 6,1–14. doi:10.3389/