Developmental Psychology

© 2020 American Psychological Association 2020, Vol. 56, No. 2, 350 –363
ISSN: 0012-1649 http://dx.doi.org/10.1037/dev0000877

The Mindful Adolescent: Developmental Changes in Nonreactivity to Inner

Experiences and Its Association With Emotional Well-Being

Brian M. Galla Eli Tsukayama

University of Pittsburgh University of Hawai’i, West O’ahu

Daeun Park Alisa Yu

Chungbuk National University Stanford University
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Angela L. Duckworth
University of Pennsylvania

Little is known about the naturalistic development of mindfulness in adolescence and how it relates to
changes in emotional well-being. The current longitudinal study examined the development of one
dimension of mindfulness, nonreactivity to difficult inner experience (or in more colloquial terms, being
able to notice, but “take a step back” from distressing thoughts), in a racially and socioeconomically
diverse sample (N ! 1,657) during the transition from middle school to high school. Students participated
in up to four assessment waves, from fall of 8th grade through spring of 9th grade, in which they
completed self-report measures assessing nonreactivity, perceived stress, and positive and negative
affect. Latent growth curve models indicated that levels of nonreactivity increased during the 2-year
study period. Developmental change in nonreactivity varied minimally by gender, socioeconomic status
(SES), and race/ethnicity. Parallel process latent growth curve models showed that changes in nonreac-
tivity were associated with concomitant reductions in perceived stress and negative affect, and increases
in positive affect. Random intercept cross-lagged panel models showed that within-person nonreactivity
prospectively predicted changes in perceived stress and positive affect, but not negative affect. This study
is among the first to track the naturalistic development of mindfulness during adolescence. Results
suggest that the nonreactivity dimension of mindfulness is associated with aspects of emotional well-
being during the transition from middle school to high school.

Keywords: adolescence, mindfulness, nonreactivity, perceived stress, emotional well-being

Supplemental materials: http://dx.doi.org/10.1037/dev0000877.supp

Mindfulness training is a large and growing industry (Wieczner, standing how mindfulness develops naturally across the life span
2016), making its way beyond hospitals and businesses and into or how such changes may influence positive development. In fact,
the school system potentially reaching millions of young people we are unaware of a single published study that has explicitly
(www.mindfulschools.org). In a special section in Developmental tracked the naturalistic development of mindfulness. Therefore, we
Psychology, Roeser and Eccles (2015) made an appeal to study know almost nothing about how its expression may change across
mindfulness from an explicitly developmental perspective. Now, adolescence or whether it differs across demographic subgroups.
four years later, the field has yet to make progress toward under- Moreover, foundational assumptions about the nature of mindful-
ness—that its growth prospectively predicts increases in emotional
well-being— have never been empirically tested.
Brian M. Galla, School of Education and Learning Research and De- In the current study, we applied a developmental approach to
velopment Center, University of Pittsburgh; Eli Tsukayama, Business track naturalistic changes in mindfulness during adolescence and
Administration Division, University of Hawai’i, West O’ahu; Daeun Park, to examine the concomitant and prospective effects of such
Department of Child Welfare, Chungbuk National University; Alisa Yu, changes on emotional well-being (operationalized here as per-
Graduate School of Business, Stanford University; Angela L. Duckworth, ceived stress and the experience of positive and negative affect).
Department of Psychology, University of Pennsylvania. We focused on one core dimension of mindfulness, nonreactivity
The current research and manuscript preparation were supported by
to difficult inner experiences. All analyses are based on a novel
funding from the John Templeton Foundation (45636, 54929).
Correspondence concerning this article should be addressed to Brian M.
longitudinal dataset involving a large socioeconomically and ra-
Galla, School of Education and Learning Research and Development cially diverse sample of adolescents during the transition from
Center, University of Pittsburgh, 5940 Wesley W. Posvar Hall, Pittsburgh, middle school to high school. Because we assessed nonreactivity
PA 15260. E-mail: [email protected] and the key outcomes of perceived stress and affect on multiple

350
 THE MINDFUL ADOLESCENT 351

occasions, this dataset provided a unique opportunity to track and pursuit of long-term goals (Shulman et al., 2016; Steinberg et
developmental change across time and apply a rigorous test of core al., 2008, 2009). Dimensions of mindfulness have been shown to
theoretical assumptions about the nature of mindfulness. be positively correlated with executive functioning in adolescence
(Oberle, Schonert-Reichl, Lawlor, & Thomson, 2012; Riggs,
Black, & Ritt-Olson, 2015; Shin, Black, Shonkoff, Riggs, & Pentz,
What Is Mindfulness?
2016).
Mindfulness is a term with roots in Buddhism (Analayo, 2003; Cognitive control development also coincides with changes in
Bodhi, 2011). It is classically understood as a mental faculty for how emotions are regulated. For example, activation in cognitive
becoming consciously aware of ongoing subjective experience control networks during reappraisal of negative stimuli increases
(Bodhi, 2011; Van Dam et al., 2018). Western psychological linearly from childhood to adulthood (McRae et al., 2012). An-
science commonly operationalizes mindfulness as involving two other study found that using reappraisal to reduce negative emo-
interrelated dimensions: self-regulation of attention and a balanced tions increases in effectiveness from childhood to adolescence
mental attitude (Bishop et al., 2004). Attention refers to the con- (Silvers et al., 2012). More recently, a naturalistic study found that
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

scious monitoring of subjective experience, without forgetfulness the effectiveness of self-distancing for down-regulating negative
This document is copyrighted by the American Psychological Association or one of its allied publishers.

or distraction, while balanced mental attitude refers to taking a affect improved linearly across adolescence (White, Kross, &
nonreactive and accepting orientation toward the experiences that Duckworth, 2015). The empirical links between nonreactivity and
arise in conscious awareness. Mindfulness is seen both as a set of various emotion regulation strategies, such as rumination, have
mental faculties whose expression varies naturally across individ- been demonstrated in several prior correlational studies (Abujara-
uals (Baer, Smith, Hopkins, Krietemeyer, & Toney, 2006; Brown deh, Colaianne, Roeser, Tsukayama, & Galla, 2019; Ciesla, Reilly,
& Ryan, 2003), and as an ability that can be cultivated through Dickson, Emanuel, & Updegraff, 2012; Galla, 2016). If develop-
meditation practice. In the current study, we examined the bal- ment of the cognitive control system supports taking a nonreactive
anced mental attitude dimension of mindfulness (specifically, the stance toward difficult experiences (Tang, Holzel, & Posner,
nonreactive orientation taken toward difficult subjective experi- 2015), then nonreactivity might also be expected to increase across
ence) because it has been proposed as the key ingredient for stress adolescence.
relief and flourishing in mindfulness training (Lindsay & Creswell, Another question is whether the pattern of change in nonreac-
2015, 2017). tivity varies across demographic subgroups. Prior research sug-
What is mindful nonreactivity, exactly? Nonreactivity is about gests differences in the ways emotions are experienced, socialized,
becoming consciously aware of distressing thoughts, emotions, expressed, and regulated across gender (Chaplin, Cole, & Zahn-
and mental images without automatically responding to them Waxler, 2005; Silk, Steinberg, & Morris, 2003), culture (Markus &
(Baer et al., 2006; Chadwick et al., 2008; Lindsay & Creswell, Kitayama, 1991; Qu & Telzer, 2017), and socioeconomic strata
2017). It means being able to “take a step back from” difficult (Blair & Raver, 2012). For example, adolescent females show a
mental experiences and view them from a wider, more dispassion- tendency to cope with stress by seeking social support, using
ate and objective perspective (Bernstein et al., 2015). It is not problem solving, and ruminating, whereas males try to avoid
about avoiding or rejecting difficult experiences, nor does it in- thinking about it (Eschenbeck, Kohlmann, & Lohaus, 2007; Silk et
herently involve attempts to alter what arises in conscious aware- al., 2003). It is possible that these factors influence affordances for
ness. Furthermore, though it may enable more adaptive responses the expression of nonreactivity, shaping its development across
to difficult experiences, it is fundamentally about taking note of time. In the current study, we explore gender, race/ethnicity, and
what is happening in subjective experience while maintaining SES as potential moderators of nonreactivity trajectories.
mental composure (Lindsay & Creswell, 2017). The conceptual
opposite of nonreactivity would entail the automatic and uncon-
Associations With Emotional Well-Being
scious filtering of thoughts, emotions, and mental images through
conditioned judgments, distortions, and evaluations. Nonreactivity is considered a dynamic and complex emotion
regulation skill (Lindsay & Creswell, 2017). Consciously “opening
Developmental Changes in Nonreactivity up to” a difficult experience in a dispassionate, receptive manner
can free additional cognitive resources (Garland, Farb, Goldin, &
Across Adolescence
Fredrickson, 2015). Unburdened by conditioned judgments and
Does the capacity to remain nonreactive in the face of difficult emotional distortions, these resources might open the door to a
experiences change across adolescence? If so, does it increase or more expansive, “big picture” view of the experience and the many
decrease with age? Adolescence is a period of life characterized by potential pathways for managing it. Difficult experiences may
the gradual maturation of the cognitive control neural system seem less permanent, overwhelming, or threatening. By contrast,
(Casey, Getz, & Galvan, 2008; Steinberg, 2008), which links to reacting automatically to difficult experiences might prematurely
improvements in other mental faculties that appear conceptually close the door to a broader, more balanced perspective, exagger-
related to nonreactivity. For example, growth in abstract, logical ating the perceived threat posed by the experience.
reasoning enables adolescents to transcend immediate situational Numerous studies show that individual differences in mindful-
details and solve problems from a less egocentric perspective ness, measured using self-report questionnaires (Baer et al., 2006;
(Dumontheil, 2014). Increases in metacognition facilitate more Brown, West, Loverich, & Biegel, 2011; Greco, Baer, & Smith,
effective introspection of one’s own mental states (Schneider, 2011), are correlated with different aspects of emotional well-
2008; Weil et al., 2013). Additionally, improvements in executive being. Cross-sectional analyses show that greater nonreactivity
function permit more effective inhibition of maladaptive impulses (and other “attitudinal” dimensions) is associated with less inter-
 352 GALLA, TSUKAYAMA, PARK, YU, AND DUCKWORTH

nalization and externalization (Greco et al., 2011; Royuela- opportunity for cultivating this mental faculty through interven-
Colomer & Calvete, 2016), lower perceived stress (Bluth & Blan- tions.
ton, 2014), and negative affect (Ciarrochi, Kashdan, Leeson,
Heaven, & Jordan, 2011), and higher quality of life (Bluth &
Blanton, 2014; Greco et al., 2011) and more positive affect (Ci- The Current Study
arrochi et al., 2011). Longitudinal analyses have also found that The current investigation is among the first to examine how a
individual differences in attitudinal dimensions of mindfulness core dimension of mindfulness, nonreactivity to difficult inner
prospectively predict changes in emotional well-being over days experiences, develops naturalistically during adolescence. It is also
(Ciesla et al., 2012), months (Abujaradeh et al., 2019; Royuela- among the first to examine how naturalistic changes in nonreac-
Colomer & Calvete, 2016), and even up to a year later (Ciarrochi tivity may relate to, and prospectively influence, perceived stress
et al., 2011). At the same time, other studies have found no such and affect. We report data from a racially and socioeconomically
reliable longitudinal associations (e.g., Cortazar & Calvete, 2019). diverse longitudinal study of adolescents transitioning from middle
Because mindfulness was only modeled at a single time point in school to high school. First, we used latent growth curve models to
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

these longitudinal analyses, its association with changes in out- examine developmental change in nonreactivity. We tested whether
This document is copyrighted by the American Psychological Association or one of its allied publishers.

comes remained vulnerable to other between-person confounds nonreactivity matured across time, and further, whether it did so in
that may be correlated with both mindfulness and outcomes. a linear or nonlinear fashion. We generally expected increases in
Several intervention studies have shown that changes in mind- nonreactivity from 8th through 9th grade, but anticipated variabil-
fulness following mindfulness-based training programs are corre- ity in trajectories across individuals. To potentially account for this
lated with changes in emotional well-being in adolescents (e.g., variability, we also explored whether the overall growth trajectory
Bluth & Eisenlohr-Moul, 2017; Brown et al., 2011; Galla, 2016). varied as a function of demographic characteristics (i.e., gender,
These studies suggest that the development of mindfulness tracks SES, and race/ethnicity), but had no a priori hypotheses about
concomitant reductions in perceived stress and improvements in which characteristics might matter. Following the same proce-
affect, but few have assessed prospective effects of mindfulness dures, we also examined developmental change in our outcomes:
(cf., Bögels, Hoogstad, van Dun, de Schutter, & Restifo, 2008). perceived stress, and positive and negative affect.
Consequently, the critical hypothesis that changes in mindfulness Next, we used parallel process latent growth curve models to
prospectively predict well-being has not yet been systematically look at multivariate change in nonreactivity and both perceived
tested. To assess directionality and help rule against the possibility stress and affect. Parallel process latent growth curve models chart
of third-variable, between-person confounds, additional research is trajectories of change in two or more variables simultaneously,
required to examine the prospective effect of within-person mind- making it possible to test whether initial levels (intercepts) and
fulness on within-person changes in outcomes. change (growth terms) in one variable are associated with initial
levels and change in another variable. Based on theory and existing
A Focus on the Transition to High School literature on nonreactivity, we expected initial levels of nonreac-
tivity to be negatively correlated with initial levels of perceived
We examined developmental changes in nonreactivity in the stress and negative affect, and positively correlated with initial
context of students making the transition from middle school to levels of positive affect. We also anticipated that adolescents who
high school. Though individuals’ experiences vary, the transition showed greater increases in nonreactivity would also show steeper
to high school often presents common academic, social, and pro- reductions in perceived stress and negative affect, and greater
cedural stressors. Academically, high school students often take increases in positive affect (Galla, 2016).
more rigorous coursework with larger workloads, are held to Finally, we used random intercept cross-lagged panel models to
higher standards by teachers, and are expected to learn and study examine reciprocal, prospective relationships between within-
independently. Socially, students must form and maintain relation- person nonreactivity and within-person perceived stress and affect
ships with many new teachers, make new friends and find their (Hamaker, Kuiper, & Grasman, 2015). By partialing out stable,
place in a new community, and may also have to contend with between-person relationships, random intercept cross-lagged panel
bullying. Procedurally, students must learn to navigate a new models permit a test of the hypothesis that an adolescent’s own
building and set of rules for behavior, and adjust to new and nonreactivity score would prospectively predict changes in that
complex class schedules. Perhaps unsurprisingly, engagement and same adolescent’s perceived stress and positive and negative af-
performance in school typically decline during the high school fect, controlling for prior levels of both constructs and within-wave
transition (Benner, 2011). Ninth grade is a pivotal year, because covariances.
students who do not make a successful transition to high school are
at greater risk of not earning a diploma (Allensworth & Easton,
2005). Method
Understanding nonreactivity during the high school transition is
important for both theoretical and practical reasons. Theoretically, Participants
greater nonreactivity (and mindfulness more generally) is hypothe-
sized to be more strongly associated with enhanced health and The analytic sample included N ! 1,657 8th and 9th grade
well-being in higher stress contexts (Creswell & Lindsay, 2014). students attending schools in California, Idaho, Pennsylvania,
Practically, if nonreactivity is found to be associated with lower and Texas. The analytic sample was drawn from a total of 1,667
perceived stress and negative affect and greater positive affect students who completed a battery of self-report measures as part
during this important life transition, it could suggest a window of of a larger observational longitudinal study on character devel-
 THE MINDFUL ADOLESCENT 353

opment during adolescence.1 Students were initially sampled (25.2%) provided data during two waves, and 174 (10.5%) pro-
from two comprehensive public middle schools and six charter vided data during one wave. Survey completion rates did not differ
or private middle schools. Following 8th grade, students en- significantly (p " .05) by gender (females vs. males), #2(3) !
tered into one of seven high schools: students at the two public 4.48, p ! .214. They did differ significantly by students’ race/
middle schools funneled into a single comprehensive public ethnicity, #2(12) ! 103.83, p " .001, eligibility for free or
high school, and students at the six charter or private schools reduced-price lunch, #2(3) ! 16.74, p ! .001, and school attended,
remained at each for high school. To maximize available data #2(21) ! 110.53, p " .001. Because of the significant associations
(and avoid arbitrary, post hoc exclusions), students were in- between demographic characteristics and survey completion rates
cluded in the analysis as long as they provided any data for at (and between demographic characteristics and the study variables
least one key measure (i.e., nonreactivity, perceived stress, themselves; see online supplemental materials), demographic char-
negative affect, and positive affect) during at least one of the acteristics were included in all models as either missing data
four assessment waves (see Procedure). correlates (i.e., auxiliary variable) or as covariates (see Analytic
According to demographic information obtained from school Plan for more details).
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

records, 47.6% of students in the analytic sample were Black,

This document is copyrighted by the American Psychological Association or one of its allied publishers.

23.4% were White, 14.4% were Latino, 10.4% were Asian, and
1.9% were of mixed or other races and ethnicities; 48.0% were Measures
female. Over half (63.7%) were eligible for free or reduced-price Nonreactivity. Students completed three items taken from the
lunch. Demographic information was missing for between 2.0 and Nonreactivity to Inner Experiences subscale of the Five Facet
3.7% of participants. Mindfulness Questionnaire (Baer et al., 2006, 2008).2 The Five
Though this study relied on a convenience sample of students, Facet Mindfulness Questionnaire is among the most widely used
our analytic sample was generally representative of the school scales for measuring individual differences in mindfulness with
populations from which it was drawn (see online supplemental adults, and several published studies have documented evidence of
materials for details). Participation rates, based on six of the eight its reliability and validity in adolescent samples (Abujaradeh et al.,
schools for which data were available, were 89.9% in fall 8th 2019; Calvete, Gámez-Guadix, & Cortazar, 2017; Ciesla et al.,
grade. 2012; Galla, 2016; Royuela-Colomer & Calvete, 2016). The non-
reactivity scale measures the tendency to be aware of difficult
Procedure thoughts and feelings without getting overtaken by them. Because
the nonreactivity scale was embedded in a larger battery of survey
This study was approved by the University of Pennsylvania’s instruments, time constraints limited us to a maximum of three
Institutional Review Board (820576, “Character Development in items. We chose items with high factor loadings in the initial
Adolescence”). Data collection began during the fall of 8th grade validation study (Baer et al., 2006) that also appeared to capture
(October 2014) and continued through the spring of 9th grade the full content of the subscale. Similar short-form (three- and
(June 2016). In fall of 8th grade, schools sent an informational four-item) versions of the nonreactivity scale in adults have dem-
letter about the study that also contained an opt-out permission onstrated evidence of adequate internal reliability, convergent va-
form. Students also completed assent forms before data collection. lidity with the full scale, and predictive validity for theoretically
There was no stopping rule for data collection; the sample size related constructs (Baer, Carmody, & Hunsinger, 2012; Gu et al.,
reflects the maximum number of students we were able to recruit 2016; Tran, Glück, & Nader, 2013). Other research has also found
within the allotted time provided by the schools. Students who evidence for the reliability and validity of a four-item nonreactivity
were not available during the first assessment were still allowed to scale in an adolescent sample that incorporates two of the three
participate (and given opportunities to provide assent) in the re- items used in this study (Abujaradeh et al., 2019). The current
maining assessment waves. As a result, some students did not items (“When I have upsetting thoughts or images, I ‘step back’
provide study data until later in middle school or until they entered and am aware of the thought or image without getting taken over
high school. by it;” “I can be aware of my feelings and emotions without having
Students participated in up to four assessment waves, each to react to them;” and “I watch my feelings without getting lost in
spaced approximately 6 months apart, during which they com-
pleted self-report measures assessing nonreactivity, perceived
1
stress, and positive and negative affect (in addition to other mea- The research questions, analyses, and conclusions reached in this
article do not overlap with publications using other portions of this dataset
sures not analyzed here). The four assessment waves are hence- (Liauw, Baelen, Borah, Yu, & Colby, 2018; Park, Yu, Baelen, Tsukayama,
forth referred to as T1, T2, T3, and T4, respectively. Two assess- & Duckworth, 2018; Park, Yu, Metz, et al., 2018).
ments (T1, T2) were administered during 8th grade, and two 2
A complete list of items used are provided in the online supplemental
assessments (T3, T4) were administered during 9th grade. All materials. This article analyzes all measures of mindfulness, perceived
measures were completed on school computers during regular stress, and positive and negative affect that were collected as part of the
larger study of character development. We note that while many other
school hours. Demographic information (gender, eligibility for measures of psychological traits and psychological functioning were col-
free or reduced-price lunch, and race/ethnicity) was collected from lected as part of a larger study (e.g., self-control, grit, expressing gratitude,
school records. actively open-minded thinking, empathy, compassion), they do not capture
Survey completion rates were as follows: 878 (53.0%) students mindfulness, perceived stress, or positive and negative affect, and so were
not included in this article. The one exception may be a single-item
in the analytic sample provided data on nonreactivity, perceived measure of life satisfaction (“During the past month, how satisfied or
stress, negative affect, and/or positive affect during all four assess- unsatisfied were you with your life?”). We decided not to include this
ment waves, 188 (11.3%) provided data during three waves, 417 measure because it was only assessed from T2 to T4.
 354 GALLA, TSUKAYAMA, PARK, YU, AND DUCKWORTH

them”) were lightly edited from the original scale to increase Table 1
comprehension, and were endorsed using a 6-point scale, from Descriptive Statistics
1 ! strongly disagree to 6 ! strongly agree. The three items
were averaged to calculate a scale score at each assessment Variable M SD Available n
wave. The observed alphas were .79 (T1), .78 (T2), .80 (T3), 1. Nonreactivity (T1) 3.91 1.10 1,249
and .82 (T4). 2. Nonreactivity (T2) 3.96 1.03 1,314
Perceived stress. Students reported on the degree to which 3. Nonreactivity (T3) 4.06 1.01 1,143
they have recently worried and felt their life was stressful, unpre- 4. Nonreactivity (T4) 4.01 1.04 1,153
5. Perceived stress (T1) 2.83 .69 1,294
dictable, and uncontrollable. The current study adhered closely to 6. Perceived stress (T2) 2.92 .73 1,322
the operationalization of perceived stress used by Park, Yu, Metz, 7. Perceived stress (T3) 2.97 .70 1,199
et al. (2018). Specifically, we used the four-item Perceived Stress 8. Perceived stress (T4) 3.00 .71 1,186
Scale (e.g., “How often have you felt that you were unable to 9. Positive affect (T1) 3.72 .73 1,259
10. Positive affect (T2) 3.66 .78 1,316
control the important things in your life?”; Cohen, Kamarck, &
11. Positive affect (T3) 3.61 .79 1,194
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Mermelstein, 1983), and two items assessing how often students 12. Positive affect (T4) 3.55 .78 1,181
This document is copyrighted by the American Psychological Association or one of its allied publishers.

felt “worried” and “stressed.” These latter two items were included 13. Negative affect (T1) 2.55 .79 1,259
because they were related to another item in the Perceived Stress 14. Negative affect (T2) 2.61 .77 1,316
Scale, “How often have you felt nervous and stressed?” All items 15. Negative affect (T3) 2.61 .80 1,193
16. Negative affect (T4) 2.66 .82 1,180
were endorsed using a 5-point scale, from 1 ! never to 5 !
always. The one deviation from Park, Yu, Metz, et al. (2018) is Note. Total N ! 1,657. T1 to T4 ! assessment Waves 1 to 4. Demo-
that we also used one additional item from the Perceived Stress graphic variables (gender, socioeconomic status, race/ethnicity, and school
attended) were included as missing data correlates.
Scale that was added to the T2–T4 assessment batteries (“How
often have you been upset because of something that happened
ing data, such as listwise deletion (Collins, Schafer, & Kam, 2001;
unexpectedly?”). Thus, our measure of perceived stress has six
Enders & Bandalos, 2001; Schafer & Graham, 2002). The sample
items during T1 and seven items during T2–T4. The items were
for all analyses reported is 1,657. FIML was facilitated by treating
averaged to calculate a scale score at each assessment wave.
demographic characteristics (gender, eligibility for free or
The observed alphas were .69 (T1), .77 (T2), .75 (T3), and .76
reduced-price lunch, race/ethnicity, and school attended) as miss-
(T4).3
ing data correlates (i.e., auxiliary variables). All analyses used the
Positive and negative affect. Students completed four items
robust maximum likelihood (MLR) estimator. Models in which
assessing positive affect (good, happy, joyful, and satisfied) and
demographic characteristics were relevant to hypotheses being
four items assessing negative affect (bad, sad, afraid, and angry)
tested (either as predictor variables or covariates) included them as
taken from the Scale of Positive and Negative Experience (Diener
independent variables rather than auxiliary variables.
et al., 2010). Items were endorsed using a 5-point scale, from 1 !
Model fit was assessed using common indices and their corre-
never to 5 ! always. Items were averaged to calculate a scale score
sponding cutoffs, according to standard conventions. Values of .90
for positive and negative affect at each assessment wave. The
observed alphas for positive affect were .77 (T1), .82 (T2), .84
(T3), and .85 (T4), and for negative affect were .71 (T1), .69 (T2), 3
As a robustness check, we reran all major analyses (i.e., latent growth
.73 (T3), and .76 (T4).4 curve analyses, parallel process latent growth curve analyses, and random
Demographic characteristics. Information about students’ intercept cross-lagged panel models) with slightly different calculations of
perceived stress: (a) using only the four items from the Perceived Stress
gender, eligibility for free or reduced-price lunch, and race/ethnic- Scale that were assessed at each wave; and (b) using the six-item measure
ity was obtained from school records. In analyses reported below, of perceived stress from Park, Yu, Metz, et al. (2018). The results of the
demographic characteristics were scored using a series of dummy reanalysis closely mirrored those of the operationalization of perceived
codes, where male, paid lunch, and Black (including n ! 31 stress used in the current article. Within-wave correlations across the
mixed/other race students) served as reference categories. We also different operationalizations of perceived stress were substantial, ranging
from r ! .89 to .98 (average r ! .93; see online supplemental materials for
included a series of seven dummy codes to account for the fact full correlation matrix).
that students were sampled from different schools. Overall 4
To promote transparency, we note that results of preliminary analyses
sample sizes were n ! 446 and 577 at the two comprehensive, (some final versions of which are reported in this article) were presented at
public schools, and ranged from n ! 41 to 169 at the six charter the 2018 Society for Research on Adolescence (SRA) conference and the
2018 International Symposium of Contemplative Research. The first and
or private schools. The school with the largest sample size (n ! second authors’ submission to SRA (in summer 2017) used a six-item
577) served as the reference category. negative affect scale (including “worried” and “stressed”), but the current
article uses only the four items taken directly from the Scale of Positive and
Negative Experiences (SPANE; Diener et al., 2010) to assess negative
Overview of Analytic Framework affect (worried and stressed are not part of the SPANE). The worried and
stressed items are now incorporated into our measure of perceived stress
All analyses were conducted using Mplus Version 8.3 (Muthén (to better adhere to Park, Yu, Metz, et al., 2018; also see Footnote 3). We
& Muthén, 2017). Data for nonreactivity, perceived stress, and reran major analyses (i.e., latent growth curve analyses, parallel process
affect were available for between 69.0 and 79.8% of students latent growth curve analyses, and random intercept cross-lagged panel
across assessment waves (see Table 1). Students with missing data models) using a six-item measure of negative affect, and the pattern of
results mirrored those of the four-item measure of negative affect used in
were included in all models by using full information maximum the current article. Thus, our final models use a four-item measure of
likelihood (FIML), which produces less biased and more efficient negative affect and a six-item or (for T2, T3, and T4) a seven-item measure
results compared with conventional methods of dealing with miss- of perceived stress.
 THE MINDFUL ADOLESCENT 355

or higher for the comparative fit index (CFI) indicate acceptable fit to Because of the large number of comparisons in these conditional
the data, and values of .95 or higher indicate excellent fit (Bentler & models, we focus our main reporting to statistically significant
Bonett, 1980; Hu & Bentler, 1999; Schumacker & Lomax, 2010). (p " .05) associations of greatest theoretical interest (gender,
Root mean square error of approximation (RMSEA) values of .08 or race/ethnicity, and SES). The online supplemental materials in-
less indicate acceptable fit, and values of .05 or less indicate clude full results of all unconditional and conditional latent growth
excellent fit (Browne & Cudeck, 1993; Schumacker & Lomax, models (see Tables S8 –S11).
2010). In general, we used chi-square difference tests with MLR Nonreactivity. We retained the linear change model, #2(8) !
correction to compare nested models (Satorra & Bentler, 2001), 15.01, RMSEA ! .023, CFI ! .986, BIC ! 19,996.71, AIC !
but did not rely solely on this test because of its sensitivity to 19,217.27. The linear model showed better fit than the no-growth
sample size (Bentler & Bonett, 1980). In latent growth curve model, %#2(3) ! 52.81, p " .001. Though the linear change model
models, we also used a combination of Bayesian Information fit significantly worse than the quadratic model, %#2(4) ! 10.36,
Criteria (BIC) and Akaike Information Criteria (AIC) values to p ! .035, the quadratic fixed effect was not statistically significant,
compare models, where lower values indicate better fit. Moreover, b ! $.02, SE ! .01, p ! .055.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

we considered how well the models estimated the growth trajec- Nonreactivity increased from 8th grade through 9th grade, b !
This document is copyrighted by the American Psychological Association or one of its allied publishers.

tory relative to the observed data by examining the growth param- .04, SE ! .01, p " .001 (see Figure 1). The intercept and slope
eter fixed effects, with a preference toward retaining the most variances were statistically significant, suggesting that initial lev-
parsimonious model. And in the random intercept cross-lagged els and changes in nonreactivity varied across students. The inter-
panel models, we followed prior recommendations where a change cept and slope were also significantly negatively correlated,
in model fit of !.010 in CFI and a change of !.015 in RMSEA r ! $.51, p " .001, indicating that higher initial levels of nonre-
would indicate that the more restrictive model does not fit worse activity were associated with a slower rate of increase across time.
than the unrestricted model (Chen, 2007). We next fit a conditional growth model to examine the moder-
ating effect of gender, race/ethnicity, and eligibility for free or
Results reduced-price lunch on the nonreactivity intercept and linear tra-
jectory. Compared with males, females reported lower initial non-
reactivity (fall of 8th grade), b ! $.31, SE ! .05, p " .001.
Descriptive Statistics
Additionally, compared with African American youth, Latino
Descriptive statistics are provided in Table 1. The mean values youth, b ! $.45, SE ! .11, p " .001, and White youth, b ! $.21,
for nonreactivity were above the midpoint (3.5) at each assessment SE ! .09, p ! .020, reported lower initial nonreactivity.
wave (M ! 3.91 to 4.06), although they varied substantially across Perceived stress. We retained the linear change model be-
participants (range ! 1 to 6). Scores on positive affect were also cause it showed excellent fit to the data, #2(8) ! 39.01, RMSEA !
higher than the midpoint (M ! 3.55 to 3.72), and about at the .048, CFI ! .979, BIC ! 15,418.02, AIC ! 14,638.58. The linear
midpoint for perceived stress (M ! 2.83 to 3.00), but below the model showed better fit than the no-growth model, %#2(3) !
midpoint for negative affect (M ! 2.55 to 2.66). 144.79, p " .001, but worse fit to the quadratic model, %#2(4) !
Bivariate correlations among variables were in the expected 28.82, p " .001. However, the linear model had a nearly identical
direction: At each assessment wave, nonreactivity was negatively BIC value to the quadratic growth model, 15,415.79 (although the
correlated with perceived stress, r ! $.30 to $.37, ps " .001, and AIC value for the quadratic model was lower, 14,614.70). As
negative affect, r ! $.28 to $.35, ps " .001, and positively shown in Figure 1, the linear trajectory appeared to adequately
correlated with positive affect, r ! .30 to .37, ps " .001. The capture observed changes in perceived stress across time. Accord-
magnitude of the correlations within waves suggests medium-sized ingly, we selected the linear model over the quadratic model as the
effects. See online supplemental materials for the full correlation basis for our final model for interpretability and parsimony.
matrix. Perceived stress increased from 8th grade through 9th grade,
b ! .06, SE ! .01, p " .001. The intercept and slope variances
Univariate Change in Nonreactivity, Perceived Stress, were statistically significant, suggesting that initial levels and
changes in perceived stress varied across students. The intercept
and Affect
and slope were also significantly negatively correlated, r ! $.33,
We examined developmental change in nonreactivity, perceived p " .001, indicating that higher initial levels of perceived stress
stress, and affect from the fall of 8th grade to the spring of 9th were associated with slower rate of increase across time.
grade through a series of latent growth curve models. In these The conditional model showed that compared with males, fe-
models, time was centered so that the growth curve intercept males had higher initial perceived stress, b ! .31, SE ! .04, p "
represented initial levels of the outcome in fall of 8th grade (0 ! .001, and a steeper increase across time, b ! .06, SE ! .01, p "
fall of 8th grade, 1 ! spring of 8th grade, 2 ! fall of 9th grade, and .001. Moreover, compared with Black youth, Asian youth, b ! .14,
3 ! spring of 9th grade). Residual variances for each construct SE ! .06, p ! .030, and White youth, b ! .18, SE ! .06, p ! .003,
were constrained to be equal across assessment waves. We com- had higher initial perceived stress.
pared three models: (a) a no-growth model, (b) a linear change Negative affect. We retained the linear change model because
model, and (c) a quadratic (nonlinear) change model. After ade- it showed excellent fit to the data, #2(8) ! 8.78, RMSEA ! .008,
quately modeling the shape of the growth trajectories, we pro- CFI ! .999, BIC ! 16,805.99, AIC ! 16,026.55. The linear
ceeded to test conditional effects of demographic characteristics model showed better fit than the no-growth model, %#2(3) !
(gender, race/ethnicity, eligibility for free or reduced-price lunch, 86.05, p " .001, and equivalent fit to the quadratic model,
and school attended) on the intercept and growth trajectories. %#2(4) ! 4.88, p ! .299. Moreover, the quadratic effect for
 356 GALLA, TSUKAYAMA, PARK, YU, AND DUCKWORTH
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Figure 1. Solid lines depict the average estimated linear trajectory of nonreactivity, perceived stress, positive
affect, and negative affect based on unconditional latent growth curve models. The dotted lines depict the
observed scores for manifest variables at each assessment wave.

negative affect was not statistically significant, b ! .00, SE ! Positive affect decreased from 8th grade through 9th grade,
.008, p ! .994. Figure 1 shows that the linear trajectory appeared b ! $.05, SE ! .01, p " .001. The intercept and slope variances
to adequately capture observed changes in negative affect across were statistically significant, suggesting that initial levels and
time. changes in positive affect varied across students. The intercept and
Negative affect increased from 8th grade through 9th grade, b ! slope were not significantly correlated, r ! .01, p ! .935, indi-
.04, SE ! .01, p " .001. The intercept and slope variances were cating that initial levels of positive affect did not relate to changes
statistically significant, suggesting that initial levels and changes in across time.
negative affect varied across students. The intercept and slope The conditional model showed that compared with males, fe-
were also significantly negatively correlated, r ! $.30, p " .001, males had lower initial positive affect, b ! $.15, SE ! .04, p "
indicating that higher initial levels of negative affect were associ- .001. Moreover, compared with Black youth, White youth had
ated with slower rate of increase across time. lower initial positive affect, b ! $.21, SE ! .06, p ! .001, but
The conditional model showed that compared with males, fe- steeper increases across time, b ! .05, SE ! .02, p ! .030.
males experienced higher initial negative affect, b ! .37, SE ! .04,
p " .001, and a greater rate of increase across time, b ! .04, SE ! Codevelopment of Nonreactivity With Perceived Stress
.02, p ! .030. Compared with Black youth, White youth had
and Affect
higher initial negative affect, b ! .17, SE ! .06, p ! .008.
Positive affect. As with the other variables, we retained the After identifying linear developmental trajectories in nonreac-
linear change model because it showed excellent fit to the data, tivity, perceived stress, and affect, we fit parallel process latent
#2(8) ! 19.91, RMSEA ! .030, CFI ! .985, BIC ! 16,880.46, growth curve models to assess the association (i.e., codevelop-
AIC ! 16,101.02. Moreover, the quadratic fixed effect growth ment) between latent slopes. Three separate models examined the
term was not statistically significant, b ! .001, SE ! .01, p ! .934. covariances between the nonreactivity and perceived stress growth
As shown in Figure 1, the linear trajectory appeared to adequately slopes; nonreactivity and positive affect growth slopes; and non-
capture observed changes in positive affect across time. reactivity and negative affect growth slopes. The primary goal for
 THE MINDFUL ADOLESCENT 357

this analysis was to examine the correlation between the linear results show that adolescents who in fall of 8th grade reported
slopes, or the degree to which change in one variable tracked higher nonreactivity also reported lower negative affect. More-
change in another. In all models, we permitted the two intercepts over, increases in nonreactivity from fall of 8th grade through
to covary, indicating the degree to which initial levels of variables spring of 9th grade were associated with steeper decreases in
were correlated. We also constrained the residual variances in each negative affect.
construct to equality across the assessment waves. These models The nonreactivity and positive affect model fit the data well,
also included several other paths: first, each growth slope was #2(72) ! 97.31, RMSEA ! .015, CFI ! .990. As shown in Table
regressed on its corresponding intercept (e.g., nonreactivity slope 2, the nonreactivity and positive affect intercepts were positively
regressed on nonreactivity intercept); second, the growth slope for correlated, r ! .52, p " .001, and the slopes were also significantly
one variable was regressed on the intercept from the other variable positively correlated, r ! .97, p ! .038. These results show that
(e.g., nonreactivity slope regressed on perceived stress intercept); adolescents who in fall of 8th grade reported higher nonreactivity
third, residuals for variables at each assessment wave were allowed also reported higher positive affect. Moreover, increases in non-
to covary (e.g., T1 nonreactivity was correlated with T1 perceived reactivity from fall of 8th grade through spring of 9th grade were
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

stress); and finally, the effect of gender, race/ethnicity, SES, and associated with steeper increases in positive affect.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

school attended were included as covariates of the initial levels

(fall of 8th grade) and trajectories of change by regressing the Cross-Lagged Associations Between Nonreactivity and
latent intercept and growth slope on each.
Perceived Stress and Affect
The nonreactivity and perceived stress model fit the data well,
#2(72) ! 120.05, RMSEA ! .020, CFI ! .985. As shown in Table Finally, we examined reciprocal, prospective associations be-
2, the nonreactivity and perceived stress intercepts were negatively tween nonreactivity and perceived stress and affect using random
correlated, r ! $.48, p " .001, and the slopes were also signifi- intercept cross-lagged panel models. In three separate models, we
cantly negatively correlated, r ! $.52, p ! .001. In other words, specified cross-lagged paths between nonreactivity and perceived
adolescents who in fall of 8th grade reported higher nonreactivity stress, between nonreactivity and positive affect, and between
also reported lower perceived stress. Moreover, increases in non- nonreactivity and negative affect, which estimated prospective
reactivity from fall of 8th grade through spring of 9th grade were effects of one variable on another. We included wave-to-wave
associated with steeper declines in perceived stress. stability (autoregressive) paths and within-wave covariances among
The nonreactivity and negative affect model also fit the data variables. We also specified random intercepts for each construct by
well, #2(72) ! 86.77, RMSEA ! .011, CFI ! .994. As shown in creating latent factors with loadings across all four waves constrained
Table 2, the nonreactivity and negative affect intercepts were to 1. The random intercept partials out stable, between-person vari-
negatively correlated, r ! $.37, p " .001, and the slopes were also ance in the variables, so that the cross-lagged paths only reflect
significantly negatively correlated, r ! $.34, p ! .023. These within-person relationships.

Table 2
Parallel Process Latent Curve Models

Nonreactivity and perceived Nonreactivity and positive Nonreactivity and negative

stress affect affect
Model parameters b SE p b SE p b SE p

Model intercepts
Nonreactivity intercept 4.14 .08 ".001 4.12 .08 ".001 4.15 .08 ".001
Nonreactivity slope .83 .20 ".001 .64 .13 ".001 .92 .16 ".001
Outcome intercept 2.66 .05 ".001 3.84 .06 ".001 2.31 .05 ".001
Outcome slope .09 .13 .485 .01 .11 .916 .39 .13 .002
Growth factor correlations and standardized
regression coefficients
Intercept ↔ intercept $.48 .05 ".001 .52 .06 ".001 $.37 .06 ".001
Slope ↔ slope $.52 .15 .001 .97 .47 .038 $.34 .15 .023
Nonreactivity intercept ¡ nonreactivity $.57 .07 ".001 $.53 .09 ".001 $.59 .06 ".001
slope
Outcome intercept ¡ outcome slope $.31 .08 ".001 .26 .26 .327 $.43 .07 ".001
Nonreactivity intercept ¡ outcome slope .18 .11 .102 $.47 .27 .086 $.13 .09 .172
Outcome intercept ¡ nonreactivity slope $.09 .09 .323 .03 .11 .789 $.18 .08 .027
model fit
#2 120.05 97.31 86.77
df 72 72 72
MLR scaling 1.04 1.05 1.08
RMSEA .020 .015 .011
CFI .985 .990 .994
Notes. N ! 1,657. #2 ! chi-square; CFI ! comparative fit index; MLR ! robust maximum likelihood; RMSEA ! root mean square error of
approximation. Demographic covariates (gender, socioeconomic status, race/ethnicity, and school attended) were included as predictors of the reported
intercepts and covariances, but their effects are not displayed to ease presentation. “↔” indicate correlations, and “¡” indicate regressions.
 358 GALLA, TSUKAYAMA, PARK, YU, AND DUCKWORTH

To enhance parsimony and interpretability, we tested a series of .022, were also statistically significant. Neither the cross-lagged
models where cross-lagged paths (M2) and autoregressive paths paths from nonreactivity to negative affect, b ! $.02, SE ! .03,
(M3), and their combination (M4), were constrained to be time- p ! .487, nor from negative affect to nonreactivity, b ! $.06,
invariant. These models were compared with the unconstrained SE ! .05, p ! .228, were statistically significant. Results of the
baseline model (M1). The online supplemental materials shows models for perceived stress and positive affect are depicted in
model fit and model comparisons for each of these models (see Figures 2 and 3, respectively (the model for negative affect is
Table S13). For all models, the constrained models fit the data as shown in Figure S1).
well as the unconstrained model. Therefore, we report results from
the most parsimonious model (M4) in which cross-lagged and
autoregressive paths were time-invariant (within-wave covariances Discussion
were allowed to freely estimate).
The between-person part of the models showed that time- This study is among the first to explicitly track naturalistic
invariant individual differences (i.e., random intercept) in nonre- development of a dimension of mindfulness during adolescence
and test how developmental change is associated with emotional
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

activity were negatively correlated with perceived stress,

well-being. Latent growth curve models indicated that, from 8th
This document is copyrighted by the American Psychological Association or one of its allied publishers.

r ! $.48, p " .001, and negative affect, r ! $.55, p " .001, and
positively correlated with positive affect, r ! .43, p " .001. grade through 9th grade, adolescents matured in their capacity to
The within-person part of the models showed that the cross- become aware of yet remain nonreactive to difficult experiences.
lagged paths from nonreactivity to perceived stress, b ! $.05, This finding is consistent with a large body of research showing
SE ! .02, p ! .016, and positive affect, b ! .10, SE ! .02, p " that the cognitive control system, which subserves executive func-
.001, were statistically significant. The cross-lagged paths from tion and emotion regulation, matures during adolescence (Casey et
perceived stress to nonreactivity, b ! $.18, SE ! .06, p ! .005, al., 2008; Steinberg, 2014). The pattern of developmental growth
and from positive affect to nonreactivity, b ! .11, SE ! .05, p ! in nonreactivity was similar across gender, SES, and ethnicity.

Figure 2. Random intercept cross-lagged panel model with nonreactivity and perceived stress. Unstandardized
regression coefficients and covariances are presented. Auto-regressive and cross-lagged paths were constrained
to be time-invariant, and within-wave covariances were freely estimated. Demographic variables of gender,
race/ethnicity, eligibility for free or reduced-price lunch, and school attended were included as missing data
correlates. The mean structure for the manifest variables was freely estimated, but is not displayed to aid visual
clarity. NR ! nonreactivity; PS ! perceived stress. Bold lines (that are not constrained to 1) are significant at
p " .05.
 THE MINDFUL ADOLESCENT 359
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Figure 3. Random intercept cross-lagged panel model with nonreactivity and positive affect. Unstandard-
ized regression coefficients and covariances are presented. Auto-regressive and cross-lagged paths were
constrained to be time-invariant, and within-wave covariances were freely estimated. Demographic vari-
ables of gender, race/ethnicity, eligibility for free or reduced-price lunch, and school attended were included
as missing data correlates. The mean structure for the manifest variables was freely estimated, but is not
displayed to aid visual clarity. NR ! nonreactivity; PA ! positive affect. Bold lines (that are not
constrained to 1) are significant at p " .05.

Parallel process latent growth curve models showed that natu- Implications for Theory and Practice
ralistic developmental changes in nonreactivity tracked concomi-
tant decreases in perceived stress and negative affect and increases This study is supportive of the broader idea that voluntarily
in positive affect. Put differently, adolescents who more rapidly approaching (in the form of “opening up to”) rather than reflex-
matured in their capacity for observing difficult experiences in a ively avoiding difficult experiences is associated with greater
nonreactive, balanced way were more likely to simultaneously well-being over the long-term (Roth & Cohen, 1986). How? Exact
experience decreases in the degree to which they felt sad and angry mechanisms of nonreactivity remain to be studied, but we specu-
as well as stressed, worried, and overwhelmed by the problems in late that nonreactive awareness may encourage the appraisal of
their life. Likewise, these same adolescents were also more likely difficult experiences as challenges rather than threats (Weinstein,
to report increases in their experience of joy and happiness. Brown, & Ryan, 2009). From a biopsychosocial perspective (Ja-
Finally, random intercept cross-lagged panel models indicated mieson, Hangen, Lee, & Yeager, 2018), nonreactivity might en-
that within-person nonreactivity prospectively predicted within- hance appraisals of coping resources, reduce appraisals of de-
person changes in perceived stress and positive affect, but not mands, or both. Appraising situational demands as challenges to be
negative affect. Adolescents who improved in their capacity for overcome rather than as threats to be avoided may in turn promote
nonreactive observation of difficult inner experiences reported less more adaptive responding and, ultimately, greater well-being. If
perceived stress and more positive affect 6 months later. These so, it may be possible to improve adolescents’ performance and
within-person associations rule out unmeasured, time-invariant affective experience in stressful situations through the training of
confounds, a concern for many studies examining correlations mindful nonreactivity. Future experimental research is needed to
between individual differences in mindfulness and well-being. test these propositions.
These results, therefore, provide strong evidence that naturalistic The discovery that dimensions of mindfulness can change in the
changes in the capacity for nonreactive observation of difficult absence of formal meditation training suggests that environmental
experiences are prospectively associated with improvements in factors may play a role in shaping the development and expression
positive dimensions of well-being. of mindfulness. Parents, for example, are powerful socializers of
 360 GALLA, TSUKAYAMA, PARK, YU, AND DUCKWORTH

their children’s emotion regulation strategies (Eisenberg, Cumber- anxiety) and appraisals of environmental demands and resources
land, & Spinrad, 1998; Eisenberg, Fabes, & Murphy, 1996). Like- (Jamieson et al., 2018), and how demographic characteristics may
wise, early temperamental dispositions, including effortful control, alter growth trajectories of other dimensions of mindfulness.
contribute to self-regulatory capacities (Eisenberg, Duckworth, Third, this study was nonexperimental. The random intercept
Spinrad, & Valiente, 2014). It seems plausible that these and other cross-lagged panel models ruled out time-invariant between-
factors (e.g., teacher and peer interactions) may help explain person confounds, but could not rule out the possibility that un-
individual differences in the naturalistic development of mindful- measured time-varying variables that mirror changes in nonreac-
ness across adolescence (Colaianne, Galla, & Roeser, 2019; Riggs tivity and causally influence outcomes might have accounted for
& Brown, 2017). Our own finding that perceived stress and pos- the observed relationships. Experiments are needed to demonstrate
itive affect prospectively predicted changes in adolescents’ nonre- causal relationships among nonreactivity, perceived stress, and
activity further reinforces the idea that mindfulness may be sensi- affect.
tive to different inputs. As the social ecology of mindfulness is Finally, our results are based on self-report data. Nonreactivity,
clarified in future studies, there may be new opportunities to perceived stress, and affect are fundamentally about individuals’
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

harness these external and internal processes in interventions. subjective perceptions of their own internal experience, and we are
This document is copyrighted by the American Psychological Association or one of its allied publishers.

What practical implications can be drawn from this study? unaware of any validated performance or informant-report mea-
Specific intervention recommendations are beyond the scope of sures of mindfulness, and in particular nonreactivity, for adoles-
this investigation. However, the discovery that adolescents re- cents. Future studies could incorporate daily diary and ecological
ported increases in perceived stress and negative affect and de- momentary assessments to minimize potential method effects.
creases in positive affect during the two-year study supports the
view that the transition to high school may be a period of height-
ened risk (Benner, 2011). This suggests, at a minimum, that the
Conclusion
transition to high school be given careful consideration when Our results show that the capacity to become aware of difficult
timing mindfulness-based interventions. Over four million stu- experiences in a nonreactive manner—a core dimension of mind-
dents make the transition to high school each year (National Center fulness—matures during parts of adolescence. Adolescents who
for Education Statistics, 2017), and many may lack the knowledge improved in this mental faculty reported less perceived stress and
and skills necessary to thrive academically, socially, and emotion- negative affect and more positive affect during the transition from
ally. Mindfulness interventions—if properly timed and delivered middle school to high school. The present study provides an initial
with fidelity, using curricula aligned with adolescents’ develop- step toward a social ecology of the mindful adolescent. Future
mental needs and motivational priorities (Yeager, Dahl, & Dweck, research is needed to determine external and internal processes that
2018)—may play a role in helping more students succeed during naturally alter the development of nonreactivity and of mindful-
the high school transition and beyond. ness more generally. Such an understanding may reveal new
avenues for augmenting this quality of mind during a critical
Limitations and Future Directions period of social-emotional development.

This study has several limitations that suggest important direc-

tions for future research. The first is external validity. It is unclear References
how nonreactivity in particular, and mindfulness in general, might Abujaradeh, H., Colaianne, B., Roeser, R. W., Tsukayama, E., & Galla,
change during other points in adolescence and the life span. B. M. (2019). Evaluating a short-form Five Facet Mindfulness Ques-
Relatedly, though this study included a racially and socioeconom- tionnaire in adolescents: Evidence for a four-factor structure and invari-
ically diverse sample of adolescents, it was not representative of ance by time, age, and gender. International Journal of Behavioral
the population of American adolescents. Future research is needed Development. Advance online publication. http://dx.doi.org/10.1177/
to determine whether the observed pattern of developmental 0165025419873039
change in nonreactivity replicates across samples that do not share Allensworth, E. M., & Easton, J. Q. (2005). The on-track indicator as a
predictor of high school graduation. Chicago, IL: Consortium on Chi-
the demographic characteristics of the adolescents in our study.
cago School Research, University of Chicago.
Second, our study focused on one dimension of mindfulness, Analayo. (2003). Satipatthana: The direct path to realization. Birming-
nonreactivity to difficult inner experiences. Though this dimension ham, AL: Windhorse.
has been hypothesized to be a key ingredient underlying the Baer, R. A., Carmody, J., & Hunsinger, M. (2012). Weekly change in
salutary effects of mindfulness training (Lindsay & Creswell, mindfulness and perceived stress in a mindfulness-based stress reduction
2017), mindfulness is a multidimensional, complex construct, and program. Journal of Clinical Psychology, 68, 755–765. http://dx.doi.org/
other dimensions have also been shown to be associated with 10.1002/jclp.21865
perceived stress and well-being (Brown et al., 2011). Future re- Baer, R. A., Smith, G. T., Hopkins, J., Krietemeyer, J., & Toney, L. (2006).
search should examine how other dimensions of mindfulness, Using self-report assessment methods to explore facets of mindfulness.
including self-regulation of attention, develop across adolescence, Assessment, 13, 27– 45. http://dx.doi.org/10.1177/1073191105283504
Baer, R. A., Smith, G. T., Lykins, E. L. B., Button, D., Krietemeyer, J., Sauer,
and whether changes in these dimensions are related to improve-
S., . . . Williams, J. M. G. (2008). Construct validity of the five facet
ments in well-being. It would also be interesting to explore the mindfulness questionnaire in meditating and nonmeditating samples. As-
codevelopment of multiple mindfulness dimensions simultane- sessment, 15, 329 –342. http://dx.doi.org/10.1177/1073191107313003
ously to see whether changes in one facilitate changes in the other. Benner, A. D. (2011). The transition to high school: Current knowledge,
We also encourage studies that examine how changes in nonreac- future directions. Educational Psychology Review, 23, 299 –328. http://
tivity may relate to mental health outcomes (e.g., depression, dx.doi.org/10.1007/s10648-011-9152-0
 THE MINDFUL ADOLESCENT 361

Bentler, P. M., & Bonett, D. G. (1980). Significance tests and goodness of Ciesla, J. A., Reilly, L. C., Dickson, K. S., Emanuel, A. S., & Updegraff,
fit in the analysis of covariance structures. Psychological Bulletin, 88, J. A. (2012). Dispositional mindfulness moderates the effects of stress
588 – 606. http://dx.doi.org/10.1037/0033-2909.88.3.588 among adolescents: Rumination as a mediator. Journal of Clinical Child
Bernstein, A., Hadash, Y., Lichtash, Y., Tanay, G., Shepherd, K., & Fresco, and Adolescent Psychology, 41, 760 –770. http://dx.doi.org/10.1080/153
D. M. (2015). Decentering and related constructs: A critical review and 74416.2012.698724
metacognitive processes model. Perspectives on Psychological Science, 10, Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global measure of
599 – 617. http://dx.doi.org/10.1177/1745691615594577 perceived stress. Journal of Health and Social Behavior, 24, 385–396.
Bishop, S. R., Lau, M., Shapiro, S. L., Carlson, L. E., Anderson, N. D., http://dx.doi.org/10.2307/2136404
Carmody, J., . . . Devins, G. (2004). Mindfulness: A proposed opera- Colaianne, B., Galla, B. M., & Roeser, R. W. (2019). Perceptions of
tional definition. Clinical Psychology: Science and Practice, 11, 230 – mindful teaching are associated with longitudinal change in adolescents’
241. http://dx.doi.org/10.1093/clipsy.bph077 mindfulness and self-compassion. International Journal of Behavioral
Blair, C., & Raver, C. C. (2012). Child development in the context of Development. Advance online publication. http://dx.doi.org/10.1177/
adversity: Experiential canalization of brain and behavior. American 0165025419870864
Psychologist, 67, 309 –318. http://dx.doi.org/10.1037/a0027493 Collins, L. M., Schafer, J. L., & Kam, C. M. (2001). A comparison of
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Bluth, K., & Blanton, P. (2014). Mindfulness and self-compassion: Ex- inclusive and restrictive strategies in modern missing data procedures.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

ploring pathways to adolescent emotional well-being. Journal of Child Psychological Methods, 6, 330 –351. http://dx.doi.org/10.1037/1082-
and Family Studies, 23, 1298 –1309. http://dx.doi.org/10.1007/s10826- 989X.6.4.330
013-9830-2 Cortazar, N., & Calvete, E. (2019). Dispositional mindfulness and its
Bluth, K., & Eisenlohr-Moul, T. A. (2017). Response to a mindful self- moderating role in the predictive association between stressors and
compassion intervention in teens: A within-person association of mind- psychological symptoms in adolescents. Mindfulness, 10, 2046 –2059.
fulness, self-compassion, and emotional well-being outcomes. Journal http://dx.doi.org/10.1007/s12671-019-01175-x
of Adolescence, 57, 108 –118. http://dx.doi.org/10.1016/j.adolescence Creswell, J. D., & Lindsay, E. K. (2014). How does mindfulness training
.2017.04.001 affect health? A mindfulness stress buffering account. Current Direc-
Bodhi, B. (2011). What does mindfulness really mean? A canonical perspec- tions in Psychological Science, 23, 401– 407. http://dx.doi.org/10.1177/
tive. Contemporary Buddhism, 12, 19 –39. http://dx.doi.org/10.1080/ 0963721414547415
14639947.2011.564813 Diener, E., Wirtz, D., Tov, W., Kim-Prieto, C., Choi, D.-W., Oishi, S., &
Bögels, S. M., Hoogstad, B., van Dun, L., de Schutter, S., & Restifo, K. Biswas-Diener, R. (2010). New well-being measures: Short scales to
(2008). Mindfulness training for adolescents with externalizing disorders assess flourishing and positive and negative feelings. Social Indicators
and their parents. Behavioural and Cognitive Psychotherapy, 36, 193– Research, 97, 143–156. http://dx.doi.org/10.1007/s11205-009-9493-y
209. http://dx.doi.org/10.1017/S1352465808004190 Dumontheil, I. (2014). Development of abstract thinking during childhood
Brown, K. W., & Ryan, R. M. (2003). The benefits of being present: and adolescence: The role of rostrolateral prefrontal cortex. Develop-
Mindfulness and its role in psychological well-being. Journal of Per- mental Cognitive Neuroscience, 10, 57–76. http://dx.doi.org/10.1016/j
sonality and Social Psychology, 84, 822– 848. http://dx.doi.org/10.1037/ .dcn.2014.07.009
0022-3514.84.4.822 Eisenberg, N., Cumberland, A., & Spinrad, T. L. (1998). Parental social-
Brown, K. W., West, A. M., Loverich, T. M., & Biegel, G. M. (2011). ization of emotion. Psychological Inquiry, 9, 241–273. http://dx.doi.org/
Assessing adolescent mindfulness: Validation of an adapted mindful 10.1207/s15327965pli0904_1
attention awareness scale in adolescent normative and psychiatric pop- Eisenberg, N., Duckworth, A. L., Spinrad, T. L., & Valiente, C. (2014).
ulations. Psychological Assessment, 23, 1023–1033. http://dx.doi.org/10 Conscientiousness: Origins in childhood? Developmental Psychology,
.1037/a0021338 50, 1331–1349. http://dx.doi.org/10.1037/a0030977
Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model Eisenberg, N., Fabes, R. A., & Murphy, B. C. (1996). Parents’ reactions to
fit. In K. A. Bollen & J. S. Long (Eds.), Testing structural models (pp. children’s negative emotions: Relations to children’s social competence
136 –162). Newbury Park, CA: Sage. and comforting behavior. Child Development, 67, 2227–2247. http://dx
Calvete, E., Gámez-Guadix, M., & Cortazar, N. (2017). Mindfulness facets .doi.org/10.2307/1131620
and problematic Internet use: A six-month longitudinal study. Addictive Enders, C. K., & Bandalos, D. L. (2001). The relative performance of full
Behaviors, 72, 57– 63. http://dx.doi.org/10.1016/j.addbeh.2017.03.018 information maximum likelihood estimation for missing data in struc-
Casey, B. J., Getz, S., & Galvan, A. (2008). The adolescent brain. Devel- tural equation models. Structural Equation Modeling, 8, 430 – 457.
opmental Review, 28, 62–77. http://dx.doi.org/10.1016/j.dr.2007.08.003 http://dx.doi.org/10.1207/S15328007SEM0803_5
Chadwick, P., Hember, M., Symes, J., Peters, E., Kuipers, E., & Dagnan, Eschenbeck, H., Kohlmann, C.-W., & Lohaus, A. (2007). Gender differ-
D. (2008). Responding mindfully to unpleasant thoughts and images: ences in coping strategies in children and adolescents. Journal of Indi-
Reliability and validity of the Southampton mindfulness questionnaire vidual Differences, 28, 18 –26. http://dx.doi.org/10.1027/1614-0001.28
(SMQ). British Journal of Clinical Psychology, 47, 451– 455. http://dx .1.18
.doi.org/10.1348/014466508X314891 Galla, B. M. (2016). Within-person changes in mindfulness and self-
Chaplin, T. M., Cole, P. M., & Zahn-Waxler, C. (2005). Parental social- compassion predict enhanced emotional well-being in healthy, but
ization of emotion expression: Gender differences and relations to child stressed adolescents. Journal of Adolescence, 49, 204 –217. http://dx.doi
adjustment. Emotion, 5, 80 – 88. http://dx.doi.org/10.1037/1528-3542.5 .org/10.1016/j.adolescence.2016.03.016
.1.80 Garland, E. L., Farb, N. A., Goldin, P., & Fredrickson, B. L. (2015). Mind-
Chen, F. F. (2007). Sensitivity of goodness of fit indexes to lack of fulness broadens awareness and builds eudaimonic meaning: A process
measurement invariance. Structural Equation Modeling, 14, 464 –504. model of mindful positive emotion regulation. Psychological Inquiry, 26,
http://dx.doi.org/10.1080/10705510701301834 293–314. http://dx.doi.org/10.1080/1047840X.2015.1064294
Ciarrochi, J., Kashdan, T. B., Leeson, P., Heaven, P., & Jordan, C. (2011). Greco, L. A., Baer, R. A., & Smith, G. T. (2011). Assessing mindfulness
On being aware and accepting: A one-year longitudinal study into in children and adolescents: Development and validation of the Child
adolescent well-being. Journal of Adolescence, 34, 695–703. http://dx and Adolescent Mindfulness Measure (CAMM). Psychological Assess-
.doi.org/10.1016/j.adolescence.2010.09.003 ment, 23, 606 – 614. http://dx.doi.org/10.1037/a0022819
 362 GALLA, TSUKAYAMA, PARK, YU, AND DUCKWORTH

Gu, J., Strauss, C., Crane, C., Barnhofer, T., Karl, A., Cavanagh, K., & Riggs, N. R., & Brown, S. M. (2017). Prospective associations between peer
Kuyken, W. (2016). Examining the factor structure of the 39-item and victimization and dispositional mindfulness in early adolescence. Preven-
15-item versions of the Five Facet Mindfulness Questionnaire before tion Science, 18, 481– 489. http://dx.doi.org/10.1007/s11121-017-0750-z
and after mindfulness-based cognitive therapy for people with recurrent Roeser, R. W., & Eccles, J. S. (2015). Mindfulness and compassion in
depression. Psychological Assessment, 28, 791– 802. http://dx.doi.org/ human development: Introduction to the special section. Developmental
10.1037/pas0000263 Psychology, 51, 1– 6. http://dx.doi.org/10.1037/a0038453
Hamaker, E. L., Kuiper, R. M., & Grasman, R. P. P. P. (2015). A critique Roth, S., & Cohen, L. J. (1986). Approach, avoidance, and coping with
of the cross-lagged panel model. Psychological Methods, 20, 102–116. stress. American Psychologist, 41, 813– 819. http://dx.doi.org/10.1037/
http://dx.doi.org/10.1037/a0038889 0003-066X.41.7.813
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance Royuela-Colomer, E., & Calvete, E. (2016). Mindfulness facets and de-
structure analysis: Conventional criteria versus new alternatives. Struc- pression in adolescents: Rumination as a mediator. Mindfulness, 7,
tural Equation Modeling, 6, 1–55. http://dx.doi.org/10.1080/10705 1092–1102. http://dx.doi.org/10.1007/s12671-016-0547-3
Satorra, A., & Bentler, P. M. (2001). A scaled difference chi-square test
519909540118
statistic for moment structure analysis. Psychometrika, 66, 507–514.
Jamieson, J. P., Hangen, E. J., Lee, H. Y., & Yeager, D. S. (2018).
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

http://dx.doi.org/10.1007/BF02296192
Capitalizing on appraisal processes to improve affective responses to
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Schafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the state of
social stress. Emotion Review, 10, 30 –39. http://dx.doi.org/10.1177/175
the art. Psychological Methods, 7, 147–177. http://dx.doi.org/10.1037/1082-
4073917693085
989X.7.2.147
Liauw, I., Baelen, R. N., Borah, R. F., Yu, A., & Colby, A. (2018).
Schneider, W. (2008). The development of metacognitive knowledge in chil-
Gratitude for teachers as a psychological resource for early adolescents:
dren and adolescents: Major trends and implications for education. Mind,
A mixed-methods study. Journal of Moral Education, 47, 397– 414. Brain and Education, 2, 114 –121. http://dx.doi.org/10.1111/j.1751-228X
http://dx.doi.org/10.1080/03057240.2017.1415872 .2008.00041.x
Lindsay, E. K., & Creswell, J. D. (2015). Back to the basics: How attention Schumacker, R. E., & Lomax, R. G. (2010). A beginner’s guide to struc-
monitoring and acceptance stimulate positive growth. Psychological tural equation modeling (3rd ed.). New York, NY: Routledge.
Inquiry, 26, 343–348. http://dx.doi.org/10.1080/1047840X.2015 Shin, H.-S., Black, D. S., Shonkoff, E. T., Riggs, N. R., & Pentz, M. A.
.1085265 (2016). Associations among dispositional mindfulness, self-compassion,
Lindsay, E. K., & Creswell, J. D. (2017). Mechanisms of mindfulness and executive function proficiency in early adolescents. Mindfulness, 7,
training: Monitor and Acceptance Theory (MAT). Clinical Psychology 1377–1384. http://dx.doi.org/10.1007/s12671-016-0579-8
Review, 51, 48 –59. http://dx.doi.org/10.1016/j.cpr.2016.10.011 Shulman, E. P., Smith, A. R., Silva, K., Icenogle, G., Duell, N., Chein, J.,
Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications & Steinberg, L. (2016). The dual systems model: Review, reappraisal,
for cognition, emotion, and motivation. Psychological Review, 98, 224 – and reaffirmation. Developmental Cognitive Neuroscience, 17, 103–117.
253. http://dx.doi.org/10.1037/0033-295X.98.2.224 http://dx.doi.org/10.1016/j.dcn.2015.12.010
McRae, K., Gross, J. J., Weber, J., Robertson, E. R., Sokol-Hessner, P., Silk, J. S., Steinberg, L., & Morris, A. S. (2003). Adolescents’ emotion
Ray, R. D., . . . Ochsner, K. (2012). The development of emotion regulation in daily life: Links to depressive symptoms and problem
regulation: An fMRI study of cognitive reappraisal in children, adoles- behavior. Child Development, 74, 1869 –1880. http://dx.doi.org/10
cents, and young adults. Social Cognitive and Affective Neuroscience, 7, .1046/j.1467-8624.2003.00643.x
11–22. http://dx.doi.org/10.1093/scan/nsr093 Silvers, J. A., McRae, K., Gabrieli, J. D., Gross, J. J., Remy, K. A., &
Muthén, L. K., & Muthén, B. O. (2017). Mplus user’s guide (8th ed.). Los Ochsner, K. (2012). Age-related differences in emotional reactivity,
Angeles, CA: Author. regulation, and rejection sensitivity in adolescence. Emotion, 12, 1235–
National Center for Education Statistics. (2017). Table 203.10. Enrollment 1247. http://dx.doi.org/10.1037/a0028297
in public elementary and secondary schools, by level and grade: Se- Steinberg, L. (2008). A social neuroscience perspective on adolescent
lected years, fall 1980 through fall 2027. Retrieved from https://nces risk-taking. Developmental Review, 28, 78 –106. http://dx.doi.org/10
.ed.gov/programs/digest/d17/tables/dt17_203.10.asp .1016/j.dr.2007.08.002
Oberle, E., Schonert-Reichl, K. A., Lawlor, M. S., & Thomson, K. C. Steinberg, L. (2014). Age of opportunity: Lessons from the new science of
adolescence. New York, NY: Houghton Mifflin Harcourt.
(2012). Mindfulness and inhibitory control in early adolescence. The
Steinberg, L., Albert, D., Cauffman, E., Banich, M., Graham, S., &
Journal of Early Adolescence, 32, 565–588. http://dx.doi.org/10.1177/
Woolard, J. (2008). Age differences in sensation seeking and impulsivity
0272431611403741
as indexed by behavior and self-report: Evidence for a dual systems
Park, D., Yu, A., Baelen, R. N., Tsukayama, E., & Duckworth, A. L.
model. Developmental Psychology, 44, 1764 –1778. http://dx.doi.org/10
(2018). Fostering grit: Perceived school goal-structure predicts growth in
.1037/a0012955
grit and grades. Contemporary Educational Psychology, 55, 120 –128.
Steinberg, L., Graham, S., O’Brien, L., Woolard, J., Cauffman, E., &
http://dx.doi.org/10.1016/j.cedpsych.2018.09.007
Banich, M. T. (2009). Age differences in future orientation and delay
Park, D., Yu, A., Metz, S. E., Tsukayama, E., Crum, A. J., & Duckworth, discounting. Child Development, 80, 28 – 44. http://dx.doi.org/10.1111/
A. L. (2018). Beliefs about stress attenuate the relation among adverse j.1467-8624.2008.01244.x
life events, perceived distress, and self-control. Child Development, 89, Tang, Y.-Y., Holzel, B. K., & Posner, M. I. (2015). The neuroscience of
2059 –2069. http://dx.doi.org/10.1111/cdev.12946 mindfulness meditation. Nature Reviews Neuroscience, 16, 213–225.
Qu, Y., & Telzer, E. H. (2017). Cultural differences and similarities in http://dx.doi.org/10.1038/nrn3916
beliefs, practices, and neural mechanisms of emotion regulation. Cul- Tran, U. S., Glück, T. M., & Nader, I. W. (2013). Investigating the Five
tural Diversity and Ethnic Minority Psychology, 23, 36 – 44. http://dx Facet Mindfulness Questionnaire (FFMQ): Construction of a short form
.doi.org/10.1037/cdp0000112 and evidence of a two-factor higher order structure of mindfulness.
Riggs, N. R., Black, D. S., & Ritt-Olson, A. (2015). Associations between Journal of Clinical Psychology, 69, 951–965. http://dx.doi.org/10.1002/
dispositional mindfulness and executive function in early adolescence. jclp.21996
Journal of Child and Family Studies, 24, 2745–2751. http://dx.doi.org/ Van Dam, N. T., van Vugt, M. K., Vago, D. R., Schmalzl, L., Saron, C. D.,
10.1007/s10826-014-0077-3 Olendzki, A., . . . Meyer, D. E. (2018). Mind the hype: A critical
 THE MINDFUL ADOLESCENT 363

evaluation and prescriptive agenda for research on mindfulness and Wieczner, J. (2016, March 12). Meditation has become a billion-dollar
meditation. Perspectives on Psychological Science, 13, 36 – 61. http:// business. Fortune. Retrieved from http://fortune.com/2016/03/12/
dx.doi.org/10.1177/1745691617709589 meditation-mindfulness-apps/
Weil, L. G., Fleming, S. M., Dumontheil, I., Kilford, E. J., Weil, R. S., Yeager, D. S., Dahl, R. E., & Dweck, C. S. (2018). Why interventions to
Rees, G., . . . Blakemore, S.-J. (2013). The development of metacogni- influence adolescent behavior often fail but could succeed. Perspectives
tive ability in adolescence. Consciousness and Cognition, 22, 264 –271. on Psychological Science, 13, 101–122. http://dx.doi.org/10.1177/
http://dx.doi.org/10.1016/j.concog.2013.01.004 1745691617722620
Weinstein, N., Brown, K. W., & Ryan, R. M. (2009). A multi-method
examination of the effects of mindfulness on stress attribution, coping,
and emotional well-being. Journal of Research in Personality, 43, 374 –
385. http://dx.doi.org/10.1016/j.jrp.2008.12.008
White, R. E., Kross, E., & Duckworth, A. L. (2015). Spontaneous self- Received September 28, 2018
distancing and adaptive self-reflection across adolescence. Child Devel- Revision received November 8, 2019
opment, 86, 1272–1281. http://dx.doi.org/10.1111/cdev.12370 Accepted November 14, 2019 !
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.