The study aimed to identify the within-person mediating mechanisms linking state mindfulness to mental health, wellbeing, and compassion.
Method
Participants (133 college students going through an eight-week mindfulness intervention, and 131 waitlist controls) provided daily measures of predictors (facets of mindfulness), mediators (rumination, cognitive interference, self-compassion, and self-transcendent emotions), and outcomes (mental health, wellbeing, and compassion). One-day lagged within-person mediation models were applied to test for mediation effects.
Results
We obtained complete or partial mediation for all outcome variables except compassion; cognitive interference was the strongest mediator. Effects of the acceptance facet of mindfulness were mediated through rumination, cognitive interference, and self-compassion; the effects of the monitoring aspect through self-compassion and self-transcendent emotions. Moderation analyses showed that the mindfulness intervention did not modify the strength of existing causal pathways. We also obtained positive correlations between meditation and mindfulness practice and state mindfulness.
Conclusions
The findings strongly suggest a causal flow of influence, where mindfulness (as enhanced through meditation and informal practices) boosts mental health and wellbeing by reducing rumination and cognitive interference and increasing self-compassion and self-transcendence. Mindfulness thus can free the mind from excessive self-preoccupation, thereby improving mental health and wellbeing.
Mindfulness is often defined as explicit awareness of the present moment with a curious and accepting attitude towards that experience (e.g., Bishop et al., 2004). Mindfulness is beneficial: It is well known to be associated with decreased stress, anxiety, and depression, and with higher quality of life (for meta-analyses, see Goyal et al., 2014; Gu et al., 2015; Khoury et al., 2015). Mindfulness can successfully be enhanced through training and intervention, with salutary results. In non-clinical populations, mindfulness interventions have been shown to increase focus, ameliorate mood, and enhance overall wellbeing while decreasing self-perceived stress, anxiety, negative affect, and rumination (Eberth & Sedlmeier, 2012).
An important question in the field concerns the mechanisms behind these beneficial changes. Knowing what these mechanisms are can allow for a deeper understanding of the causal structure of the different potential effects, ultimately leading to interventions that can target effective mechanisms more directly (Kazdin, 2009). On the theoretical side, many models have been advanced to explain the translation of mindfulness into positive outcomes (e.g., Baer, 2003; Brown et al., 2007; Chiesa et al., 2013; Creswell & Lindsay, 2014; Grabovac et al., 2011; Hölzel et al., 2011; Segal et al., 2013; Shapiro et al., 2006), each with their own emphases and levels of complexity. Despite this variety, the list of proposed mechanisms generally contains three interrelated categories, as pointed out by Vago and Silbersweig (2012): self-awareness, self-regulation, and self-transcendence. Self-awareness refers to a heightened awareness of momentary states of body and mind (“monitoring”; Lindsay & Creswell, 2017), as well as an increased acceptance of whatever thoughts, feelings, or sensations are present in awareness (“acceptance”; Lindsay & Creswell, 2017)—what is typically meant by trait mindfulness. In these frameworks, this meta-awareness of the self might give rise to a more effective management of one’s thoughts and emotions, that is, better self-regulation. Such self-regulatory mechanisms include better control over self-directed negative thought as well as heightened self-compassion, that is, treating oneself with kindness and understanding. Self-awareness and self-regulation can ultimately lead to self-transcendence, that is, the development of a positive relationship with oneself and others that transcends self-focused needs and becomes increasingly prosocial.
Empirically, the field has mostly focused on mindfulness and self-regulation as the mechanisms that translate meditation or other mindfulness practices into beneficial outcomes. Specifically, in their 2015 review and meta-analysis on the mechanisms of mindfulness interventions, Gu et al. identified 18 intervention studies that examined mechanisms for the interventions’ effects on stress, anxiety, and depression. The four most frequently probed mechanisms were mindfulness, repetitive negative thinking (including rumination and worry), self-compassion, and emotional/cognitive reactivity. The validity of such mechanisms is typically assessed through mediation analysis, where the influence of a predictor variable (e.g., mindfulness) on an outcome (e.g., mental health and wellbeing) is said to be mediated by an intervening variable (the “mediator”; e.g., rumination) when the strength of the relationship between predictor and outcome is significantly reduced after controlling for the influence of the mediator (e.g., Baron & Kenny, 1986; MacKinnon, 2012).
One issue, however, is that most mediation studies use a two-occasion (i.e., pre-post intervention) design without intermediate assessments of mechanism or outcomes (for an exception, see Mirabito & Verhaeghen, 2023). A stronger design would include more frequent measurements: The hypothesis of mechanistic causality would be greatly strengthened if a temporal sequence from predictor to mediator to outcomes could be established.
Frequent-measurement designs have an additional and vital advantage: They place the locus of causality at the intra-individual, that is, within-subject level rather than the inter-individual, between-subject level. Traditional inter-individual mediation analyses can answer questions such as “Do people who are more mindful ruminate less and therefore feel happier?” Although this is a worthwhile question, it can be argued that the intra-individual level (i.e., “When I am more mindful, will I also be less ruminative and therefore happier?”) is the level some subfields of psychology should arguably be more interested in. That is, the intra-individual level is where therapy and other interventions operate—the realm of intrapersonal salutary change. Although the assumption in much psychological (including mindfulness) research is that conclusions reached through inter-individual-level analysis also generalize to the intra-individual level, this is not necessarily the case (this assumption is known as the ergodicity problem or ergodic fallacy; Molenaar & Campbell, 2009; Speelman et al., 2024). In other words, the validity of this assumption needs to be empirically verified through the examination of within-person couplings between the variables of interest: are day-to-day fluctuations in mindfulness indeed coupled with day-to-day changes in outcomes, and can those be linked through day-to-day changes in the purported mechanisms?
Additionally, frequent-measurement designs allow for the use of state measures, that is, measures that capture the current state of a person—their experience in-the-moment. In contrast, traditional two-occasion designs typically employ trait measures, that is, measures that tap into presumably stable, habitual, or slow-changing characteristics. States and traits do interact (Schmitt & Blum, 2020), but generally, unlike the two-occasion method, the state-measure approach allows for the examination of fast-moving processes and their dynamic interplay that are missed when using trait measures (Shiffman et al., 2008). This is also the level at which people’s perception of their own mindfulness, wellbeing, and/or mental health likely operates—how they feel in this particular moment.
Recently, Enkema et al. (2020) identified 15 studies deploying frequent measurement of state variables in the context of a mindfulness intervention. Out of these 15, six examined mechanisms, or at least associations between variables. The general conclusions from these studies were that mindfulness was associated with improved affect (Brown & Ryan, 2003; Gotink et al., 2016; Shoham et al., 2017), with greater emotional stability (Quaglia et al., 2015), and with greater differentiation of emotions (Van der Gucht et al., 2019). In a more recent study, cognitive interference (a measure akin to rumination, but also arguably tinged with worry) was found to mediate between state mindfulness and depression and wellbeing (Mirabito & Verhaeghen, 2023). The literature, then, gives credence to Vago and Silbersweig’s (2012) assertion that state mindfulness is key in explaining the effects of mindfulness interventions, but leaves, for the most part, the question of the mechanistic link between mindfulness and its mechanisms open.
The study of couplings of within-person change within mindfulness interventions, then, is arguably sparse. It also, in our opinion, largely suffers from a number of shortcomings. First, these studies invariably examined mindfulness as a unitary construct. Mindfulness, however, is often considered as consisting of both a monitoring/observing and an accepting/nonjudging aspect. This dichotomy is inherent in most definitions of the construct, including Bishop et al.’s (2004). Some have asserted that the attention monitoring aspect of mindfulness might lead to negative emotional consequences (especially in non-meditators; Baer et al., 2006) and that acceptance of one’s experience is necessary to regulate one’s affective experience towards positive outcomes (Monitoring and Acceptance Theory [MAT]; Lindsay & Creswell, 2017; for a critical reading, see Simione & Saldarini, 2023).
Second, very few studies (Mirabito & Verhaeghen, 2023, is an exception) examined whether group membership (i.e., being part of the intervention group vs. the control group) moderates the mediation effects. This is crucial to determine whether mindfulness interventions introduce a new mechanism, perhaps by changing the strength of one or more of the pathways within the general predictor-mediator-outcome mechanisms, for instance by creating a stronger sensitivity to mindfulness and/or the mediator, or because the intervention inserted a second, unmeasured mediator in the pathway (e.g., group processes; Goldberg, 2022). Such new mechanisms would introduce a moderation of pathways by group. Conversely, a lack of moderation would signify that the mechanisms translating mindfulness into beneficial outcomes are identical for both groups. In that case, the mindfulness intervention would be effective in changing mental health and wellbeing simply by raising state mindfulness, which would then influence mental health and wellbeing by known pathways. These findings have implications for therapy or other forms of intervention: If mindfulness training would introduce new mechanisms, it would be irreplaceable; if it would not, its effects could also be achieved through alternative interventions that target mindfulness through other means than the traditional formal mindfulness practices (e.g., Acceptance and Commitment Therapy), which may be helpful for clients who have difficulty practicing such formal techniques.
In the present study we investigated and documented the couplings of day-to-day changes in predictors, mediators, and outcomes through ecological momentary assessments (EMA), that is, frequent measures collected via smartphones in the moment at quasi-random times. We did so over the course of an eight-week mindfulness intervention; each measure was collected at least daily. The intervention was conducted in 2021–2022 in groups of college students in remote format. An earlier report on the same data set (Aikman et al., 2026) established that trait measures of mindfulness, mental health, wellbeing, and compassion showed stronger beneficial changes over three time points (pre-, mid-, and postintervention) in the mindfulness-trained group than the waitlist control group. Our outcome measures were measures of mental health (depression, anxiety, stress, and negative affect) and wellbeing (flourishing/eudemonic wellbeing and positive affect), variables that typically benefit from mindfulness interventions (Goyal et al., 2014; Gu et al., 2015; Khoury et al., 2015), as well as compassion, which is often claimed as a beneficial outcome of mindfulness interventions, but rarely investigated (Kreplin et al., 2018). Our mediator variables follow from the theoretical and empirical literature cited above: state self-directed negative thought (operationalized here as cognitive interference, as in Mirabito & Verhaeghen, 2023, and rumination), state self-compassion, and state self-transcendence (here operationalized, as in Verhaeghen, 2019, as self-transcendent emotions of love, awe, and joy). Based on the literature cited above, we expected that day-to-day changes in the observing and nonjudging aspects of state mindfulness (as well as a total state mindfulness score) would be associated with beneficial changes in cognitive interference, rumination, self-compassion, and self-transcendent emotions, and that these four variables would mediate some of the effects of mindfulness on mental health, wellbeing, and compassion.
Method
Participants
A total of 264 participants (131 in the control group and 133 in the experimental group; mean age = 20.38; SD = 3.31; 58 men and 206 women) provided EMA data (subset of the original sample of 332 who provided pretest data). Inadvertently, race and ethnicity information were not collected; however, 70% of students at the University of North Georgia are White, 16% are Hispanic, 4.3% Asian, Native Hawaiian, or Pacific Islander, 3.8% Black or African-American, 3.5% Multiracial; 35% of students at the Georgia Institute of Technology are White, 8.2% are Hispanic, 34% Asian, 8% are Black or African-American, and 4.6% are Multiracial. Participants were recruited via announcements and flyers shared with classes across both campuses as well as select listservs.
Participants received $200 in return for their time and effort. They could earn a $100 bonus, dependent on program attendance (for the intervention group) and survey completion rates (for both groups). Participants were randomly assigned to the intervention or the waitlist control group. Data were collected over the course of 3 semesters: Fall 2021 and Spring and Fall 2022.
Procedure
The intervention consisted of a manualized 8-week protocol (MIEA parts 1 and 2, each 4 weeks in duration) developed by the Mindfulness Institute for Emerging Adults (formerly Koru Mindfulness, Rogers & Maytan, 2019; we will refer to the protocol hereafter as MIEA). In accordance with pandemic protocols, all classes were conducted remotely and online.
The MIEA program is tailored specifically towards college students in two major aspects. First, the frequency and amount of formal and informal home practice is lower than in other programs such as MBSR—amounts that students typically perceive as manageable. Second, the program has an initial emphasis on practices aimed at stress reduction before moving into more insight-based practices because stress is an important complaint by college students. The program was conducted in small groups of about 5–12 participants to maximize interaction among students and between students and instructor.
The program is broad, involving body scan meditation, breathing meditation, belly breathing, dynamic breathing, gatha meditation, labeling-of-thought and labeling-of-feelings meditation, loving-kindness meditation, chair yoga, mindful eating, and walking meditation. Practices were introduced in eight weekly 75-min sessions that also included check-ins and group discussions. Participants were instructed to practice these techniques, as they learned them, for ten minutes per day for the first four weeks, increased to 20 min per day for the final four weeks; hereafter, we label this as meditation. Each week, participants were also instructed to perform one daily activity of their choice (such as showering, getting dressed, drinking one’s morning coffee, etc.) mindfully; hereafter we call this mindful activities. The homework load is low compared to other MBIs to decrease attrition, increase compliance, and allow participants to weave these practices into their schedules. Participants were taught by one of two of the authors, Aikman or Verhaeghen, both of whom are certified MIEA instructors. In addition to the group sessions, participants had access to resources through the MIEA mobile app and MIEA website, which include guided meditations. They were also provided with digital copies of the accompanying books The Mindful Twenty-Something (Rogers, 2016) and Real Happiness (Salzberg, 2010).
The measures and analyses presented in this paper represent a subset collected in a larger intervention study. In this paper, we present findings concerning state variables obtained through ecological momentary assessment (EMA) over the course of the intervention. Results concerning pre, mid, and post-measurement of trait variables were reported in Aikman et al. (in press).
Measures
EMA data were collected via the ExpiWell (www.expiwell.com) smartphone app on the participant’s own device. Participants were cued to answer an EMA assessment four times each day. Preset cueing intervals were 10 am–1 pm for the morning survey, 1:30–4 pm for the early afternoon survey, 4:30–7 pm for the late afternoon survey, and 7:30–10 pm for the evening survey. If they wished, participants were able to select a 30-min window within each interval in which they preferred to be cued. Once a cue was presented, participants were given a ten-minute window to start the survey, with a reminder at the four-minute mark.
We tried to make all assessments about equal length (estimated to be between 3 and 6 min). The morning survey contained the go/no-go task, mind wandering probes, and the state mindfulness measures; the early afternoon survey measured mind wandering, state mindfulness, meditation and mindful activities, depression and anxiety, negative and positive affect, and flourishing; the late afternoon survey measured mind wandering, state mindfulness, meditation and mindful activities, compassion and self-compassion, and self-transcendent emotions; the evening survey measured mind wandering, state mindfulness, meditation and mindful activities, stress, cognitive interference, and rumination. Surveys were active over the period of the intervention, that is, over 56 days. Due to technical difficulties, EMA cues were presented to some participants for a longer period of time (max = 64 days). All 264 participants included in the analysis completed at least one EMA survey. On average participants responded to 103.88 EMA cues (46% of 224 possible), SD = 70.03, min = 1, max = 219. Adherence did not significantly correlate with age, gender, condition, or any of the Big Five personality factors (measured here by the IPIP; Donnellan et al., 2006): largest pretest |r| = 0.06. We did obtain significant correlations between the total numbers of cues answered and the pretest DASS depression and anxiety subscales (Henry & Crawford, 2005), r = −0.13 and −0.14, respectively, suggesting that dysphoria and anxiety negatively impacted adherence.
State Mindfulness
The state mindfulness measures were constructed by taking two items each from each of the subscales of the Five Facets Mindfulness Questionnaire (FFMQ; Baer et al., 2006): Nonjudging of inner experience, hereafter called Nonjudging (“In the past few hours, I made judgements about whether my thoughts are good or poor”, “In the past few hours, I thought some of my emotions are poor or inappropriate and I shouldn’t feel them”; McDonald’s ω = 0.88; for the FFMQ scales, McDonald’s ω was calculated over the four measurement occasions per day because the data were analyzed at the day-by-day level), Describing (“It would be easy for me to describe in detail how I’ve been feeling over the past few hours”, “It would be hard for me to find the words to describe what I’ve been thinking over the past few hours”; McDonald’s ω = 0.85), Observing (“In the past few hours, I have paid attention to how my emotions affect my thoughts and behavior”, “In the past few hours, I have paid attention to sensations, such as the wind in my hair or the sun on my face”; McDonald’s ω = 0.84), Non-reactivity to inner experience, hereafter called Non-reactivity (“In the past few hours, I watched my feelings without getting lost in them”, “In the past few hours, I perceived my feelings and emotions without having to react to them”; McDonald’s ω = 0.88) and Acting with awareness (“In the past few hours, it seemed I was ‘running on automatic’ without much awareness of what I was doing”, “In the past few hours, I haven’t paid attention to something I’m doing because I’m daydreaming, worrying, or otherwise distracted”; McDonald’s ω = 0.86). All subscales of state mindfulness were measured at all occasions; all were scored such that higher scores denoted higher levels of mindfulness. A total score was calculated as the mean of all five subscales (McDonald’s ω = 0.83).
Proposed Mediators: Self-preoccupation, Self-compassion, and Self-transcendent Emotions
Self-Preoccupation was measured using the 3-item state rumination scale devised by Puterman et al., 2010; McDonald’s ω = 0.92 as well as an adapted version of the 6-item Short Cognitive Interference Test, hereafter called cognitive interference (Stawski et al., 2011; “Did you think about personal worries?”, “Did you have trouble concentrating?”, “Did you try to avoid certain thoughts?”, “Did you try to put problems out of your mind?”, “Did you think about something you didn’t mean to?”, “Did you have thoughts that kept jumping into your head?”; McDonald’s ω = 0.90).
Self-compassion was measured using a state-worded version (i.e., “Please indicate how you are feeling about yourself right now…”) of the Self-Kindness, Common humanity, and Mindfulness subscales of the Self-Compassion Scale Short-Form (Raes et al., 2011), 6 items in total (McDonald’s ω = 0.78).
Self-Transcendent Emotions were measured using a state-worded version (i.e., “How well does the statement describe what you just experienced, just now?”) of 6 items from the Dispositional Positive Emotions Scale (DPES, Shiota et al., 2006; “I felt bursts of joy”, “good things happened to me”, “I saw beauty around me”, “I felt awe”, “I found it easy to trust others”, “I experienced feelings of closeness to people around me”; McDonald’s ω = 0.86).
Proposed Outcomes: Mental Health, Stress, Wellbeing, and Compassion
Mental health was measured using a state version of the four-item Patient-Reported Outcomes Measurement Information System (PROMIS) short-form depression scale (McDonald’s ω = 0.91) and the four-item PROMIS short-form anxiety scale (Moore et al., 2016; McDonald’s ω = 0.89), as well as a ten-item version of the Positive And Negative Affect Schedule (Ruscio et al., 2016; Positive affect: McDonald’s ω = 0.83; negative affect: McDonald’s ω = 0.87).
Stress was measured using the Daily Inventory of Stressful Events (seven items; Sliwinski et al., 2006), which taps into common stressors; hereafter called stressful events (McDonald’s ω = 0.81). Additionally, we included a sliding scale to indicate an overall perceived stress score for the day (“On a scale from 0–100, how stressful would you say your last 24 h have been?”)
Flourishing (sometimes also known as eudemonic wellbeing) was measured using the four-item Runyan et al. (2019) state adaptation of the Diener et al. (2009) Flourishing Scale (McDonald’s ω = 0.87).
Compassion was tapped by Runyan et al. (2016) four-item state adaptation of the Cameron and Payne (2011) Compassion Scale (McDonald’s ω = 0.97).
Meditation/Mindful Activity Probes
Meditation and mindful activities were probed using the questions “Since we last checked in, did you do any formal meditation?”, “How many minutes did you meditate for?”, “Since we last checked in, did you do a mindful activity?”, and “How many minutes did that activity take?” We used the second and fourth questions in our data analysis, averaged within day.
Additional EMA Measures
Additional EMA measures not reported here concerned mind wandering probes and a go/no-go task.
Data Analyses
Within-person correlations between the state variables of interest were calculated using the misty package in R. Moderation by group of these correlations was analyzed using multilevel regression implemented using the lmer function in R. In these models, variables were within-person centered and a group by predictor variable was included to test for moderation; additionally, the corresponding person-mean of the predictor variable and time were included as control variables (as in Bolger & Laurenceau, 2013). Two regression analyses were run for each correlation, using each of the variables of the correlation pair as the predictor and outcome in turn. Moderation was considered significant when the interaction term was significant in both regressions.
Within-person mediation analyses were set up to examine the temporal sequence across days, such that the predictors, or X variables (i.e., the FFMQ Observing and Nonjudging subscales and the total mindfulness score) influenced the mediators, or M variables (i.e., rumination, cognitive interference, self-compassion, and self-transcendent emotions) on the subsequent day, and the outcomes, or Y variables (the remaining variables), on the day after that. Introducing this time lag in the analyses allows for stronger causal inference than an examination of within-day mediation would. A series of 1–1–1 within-person mediation models, one for each combination of X, M, and Y variables, was conducted as specified in Bolger and Laurenceau (2013, as further specified at https://thechangelab.stanford.edu/tutorials/specifying-1-1-1-mediation-models-r/), using the lme function in R. All models included time as a control variable. Variables were within-person centered and z-transformed to obtain standardized regression coefficients. We also examined moderation of the X→M, M→Y, and X→M links by group (https://thechangelab.stanford.edu/tutorials/specifying-1-1-1-mediation-models-r/).
Leveraging the frequent measurements design, lagged within-person mediation analyses were also conducted. This was done by extending the temporal sequence to examine how long the effects of state mindfulness on the mediators lasted. Across a series of models, the lag between predictor and mediator was systematically varied (e.g., total mindfulness on day t predicting rumination on day t + 1, t + 2, t + 3, and so on).
Results
Within-Person Correlations Among the Variables of Interest
Figure 1 presents a heatmap of the whole-sample within-person correlations between the state variables of interest; significant correlations are indicated in boldface. As can be seen and as expected, both mindfulness and the mental health/flourishing variables form a manifold, with many significant correlations within each cluster (median r within mindfulness scales = 0.18; median absolute r within mental health and flourishing variables = 0.34). There are also many correlations (all in the expected direction) between mindfulness and the assumed stress/mental health/wellbeing/compassion outcomes (median absolute r = 0.15), between mindfulness and the purported mediators (median absolute r = 0.18), and between the purported mediators and the assumed stress/mental health/wellbeing/compassion outcomes (median absolute r = 0.21). The existence of significant correlations between predictors and mediators, mediators and outcomes, and predictors and outcomes justifies within-person mediation analyses, which are described in the next section.
Fig. 1
Heat map of within-person correlations between the state variables of interest. Red colors denote negative relationships; blue colors represent positive relationships. Boldfaced coefficients are significant at p < 0.05
As can be seen in Fig. 1, time spent meditating and/or time spent in mindful activities (which are intercorrelated) shows small (between −0.02 and 0.06) but reliable correlations with the mindfulness variables, rumination, flourishing, and self-transcendent emotions, but none of the other variables. Note that these correlations were calculated within the intervention group only.
Moderation analysis showed that five (out of 153) correlations were significantly different between groups. At an alpha level of 0.05, about eight correlations would be expected to be spuriously flagged for moderation. Thus, the number of statistically significant correlations was not larger than expected under the null hypothesis. Given the risk that those correlations were likely to be false positives, it seems prudent not to discuss them further.
Within-Person Mediation Analyses
Parameters for the within-person mediation analyses (i.e., path coefficients and significance level, as well as group moderation effects) are reported in Online Resource 1; Fig. 2, 3, and 4 schematize those results, one figure for each predictor variable. In the figures, line width is proportional to the size of the coefficients. Note that all analyses involved a single X, M, and Y variable; the figure represents a schematized synthesis of these results.
Fig. 2
Schematized results from within-person 1–1–1 mediation models with the total state mindfulness score as the predictor. Red lines denote negative relationships; blue lines represent positive relationships; line width is proportional to the standardized regression coefficients. Only significant paths are represented
Schematized results from within-person 1–1–1 mediation models with the Observing subscale of the state mindfulness scale as the predictor. Red lines denote negative relationships; blue lines represent positive relationships; line width is proportional to the standardized regression coefficients. Only significant paths are represented
Schematized results from within-person 1–1-1 mediation models with the Nonjudging subscale of the state mindfulness scale as the predictor. Red lines denote negative relationships; blue lines represent positive relationships; line width is proportional to the standardized regression coefficients. Only significant paths are represented.
Rumination and cognitive interference were significant mediators for both the total score and the Nonjudging aspect of mindfulness, except for the variable of positive affect; mindfulness and cognitive interference significantly mediated compassion from Nonjudging. Cognitive interference mediated all of the variance between the total mindfulness score and the Y variables it was associated with, with the exception of self-compassion; there were more direct paths (i.e., partial rather than full mediation) for rumination as a mediator. Rumination and cognitive interference did not mediate any variance from Observing. Self-compassion was a significant mediator, with some exceptions. For Observing, it fully mediated all relationships, except for that to self-transcendent emotions. The variable of self-transcendent emotions was a significant mediator for total mindfulness score and Observing, but not Nonjudging. For Observing, it fully mediated all relationships but one.
The data thus show that all mediators are ultimately interconnected. That is, each mediator also transmits some of the variance from mindfulness (if any) to the other mediators. The implication is that there is a flowing cascade where mindfulness affects mental health and wellbeing through one type of mechanism that also affects a second mechanism that in turn affects mental health and wellbeing and another mediator, and so on.
Group as Moderator of Within-Person Mediation
Results for the moderator analyses are presented in Online Resource 1. Overall, 18 of the links, or 4.55%, showed moderation by group. Thus, the number of statistically significant coefficients is not larger than expected under the null hypothesis. Given the risk that those coefficients might be false positives, it seems prudent not to discuss them further.
Lagged Within-Person Mediation Analysis
Results of the lagged within-person mediation analyses are schematized in Fig. 5, 6, and 7; parameters are reported in Online Resource 2. Some effects lasted only a single day (viz., total score to rumination; observing to self-transcendent emotions); some lasted for two days (viz., total score to cognitive interference; nonjudging to rumination and self-compassion); some for three days (nonjudging to self-compassion); some for four (total score to self-compassion; total score to self-transcendent emotions; observing to self-compassion); none were significant beyond four days.
Fig. 5
Schematized results from analysis of delayed predictor-to-mediator paths for within-person 1–1-1 mediation models with the total state mindfulness score as the predictor. Red lines denote negative relationships; blue lines represent positive relationships; line width is proportional to the standardized regression coefficients. Only significant paths are represented. Paths to outcomes are only schematized here (and hence indicated as dashed, leading to dashed boxes); the estimated paths are as in Fig. 2
Schematized results from analysis of delayed predictor-to-mediator paths for within-person 1–1-1 mediation models with the Observing subscale for state mindfulness as the predictor. Red lines denote negative relationships; blue lines represent positive relationships; line width is proportional to the standardized regression coefficients. Only significant paths are represented. Paths to outcomes are only schematized here (and hence indicated as dashed, leading to dashed boxes); the estimated paths are as in Fig. 3
Schematized results from analysis of delayed predictor-to-mediator paths for within-person 1–1-1 mediation models with the Nonjudging subscale for state mindfulness as the predictor. Red lines denote negative relationships; blue lines represent positive relationships; line width is proportional to the standardized regression coefficients. Only significant paths are represented. Paths to outcomes are only schematized here (and hence indicated as dashed, leading to dashed boxes); the estimated paths are as in Fig. 4
The aim of our study was to investigate, within-person, the mechanism by which mindfulness translates into salutary outcomes. Specifically, we examined whether day-to-day changes in the observing and nonjudging aspects of state mindfulness (as well as a total state mindfulness score) were beneficially associated with cognitive interference, rumination, self-compassion, and self-transcendent emotions, and whether these four purported mechanisms would mediate some of the effects of mindfulness on mental health, wellbeing, and compassion. Summarized very briefly, we found that rumination, cognitive interference, self-compassion, and self-transcendence indeed mediated at least some of the variance of mindfulness to mental health and wellbeing, but with some important nuance around the aspects of mindfulness probed—observing or nonjudging. Some of these effects could be traced over multiple days. We elaborate on these and other results below.
We selected four variables from the literature that we hypothesized would be good potential mediators to translate the effects of state mindfulness to state mental health, wellbeing, and compassion, namely rumination, cognitive interference, self-compassion, and self-transcendence. We investigated the mediation hypothesis using across-day within-person multilevel regression models. We introduced a one-day temporal lag between predictors, mediators, and outcomes to allow for strong causal inference; measuring mediation at the within-person level (as opposed to the more usual between-subject level) allowed us to answer the questions at the level typically sought by an individual, that is, for instance: “If I become more mindful, will I also become less ruminative and will that dampen my anxiety?” rather than “Are people who are more mindful also less ruminative and therefore less anxious?”.
We considered mindfulness under three guises. The first was an aggregate score, that is, the average of all five subscales of a state version of the FFMQ (Nonjudging, Describing, Observing, Nonreactivity, and Acting with Awareness). This aggregate level is what is typically reported in mindfulness intervention research; doing so here allows for across-study comparisons. The second and third aspects concern the Observing and Nonjudging subscales, respectively. The reason to examine these two subscales separately stems from the prediction from Monitoring and Acceptance Theory (Lindsay & Creswell, 2017) that these two aspects of mindfulness might lead to differential outcomes. Mindfulness definitions (see, for instance, the consensus definition by Bishop et al., 2004, cited at the very head of this paper) typically include both aspects: a deliberate monitoring of awareness and an open, nonjudging, equanimous attitude towards that experience. Monitoring and Acceptance Theory posits that enhanced monitoring of one’s experience (measurable via, among others, the Observing subscale of the FFMQ) can lead to positive outcomes in the attention system, but may also increase affective reactivity, which may have negative consequences, for instance when negative states get intensified. This effect might be especially pronounced in non-meditators (Baer et al., 2006). What is needed for a beneficial outcome is then not just observation, but a modification of one’s relationship to experience through acceptance (captured in, among others, the Nonjudging subscale of the FFMQ).
At the aggregate level, all four proposed mediators indeed reliably mediated at least some of the variance for most of the outcomes. The effects of mindfulness on mental health and flourishing were significantly mediated by all four proposed mediators, with cognitive interference showing full mediation for these variables (i.e., there are no direct paths), and all other mediators showing at least partial mediation. Positive affect and compassion were each only mediated by one mechanism, self-compassion and self-transcendent emotions, respectively.
The Observing and Nonjudging aspects of mindfulness did indeed operate through different mechanisms. Within-person changes in Observing were not related to changes in rumination or cognitive interference, but they were to self-compassion and self-transcendent emotion. The latter mechanisms fully explained the relationship between Observing and mental health and between Observing and flourishing (as well as positive affect for self-transcendent emotion). These results suggest that, within-person, the negative view on monitoring expressed in the Monitoring and Acceptance Theory is incorrect: When people are more observant, they become more self-compassionate and experience more self-transcendent emotions, which leads to beneficial outcomes on mental health and flourishing, and this effect is likely causal.
Within-person changes in Nonjudging operated through rumination and cognitive interference (with full mediation for all variables except positive affect) and through self-compassion (with partial mediation for anxiety and depression, and full mediation for all other mental health and wellbeing variables); self-transcendent emotions were not connected to Nonjudging. Note that there is a within-day correlation between Nonjudging and self-transcendent emotions (as seen in Fig. 1), so the lack of effect in the lagged mediation analysis suggests that this relationship is more fleeting than the relationship between Nonjudging and the other potential mediators. These relationships are in line with the expectations from Monitoring and Acceptance Theory.
The differential mechanisms originating from Observing and Nonjudging are worth noting and perhaps themselves in need of explanation. Some links are easy to understand: It makes intuitive sense that being able to accept one’s experience without judgment would decrease the intensity and/or number of intrusive thoughts and/or the amount of interference experienced from such thoughts. It also makes sense that this would not be the case for the mere monitoring or observing of such thoughts. It is less clear, however, why a sharper monitoring or observing of experience would make one more compassionate towards oneself.
Figure 8 provides for interpretation of the results as they pertain to the total mindfulness score, and packaging rumination and cognitive interference as self-preoccupation. The figure illustrates how all proposed mechanisms are significant mediators and how most of the effects of mindfulness are indirect. This indicates that the mediators indeed mediate most (and sometimes all) of the variance and thus offer a strong explanation for the effects of mindfulness. The figure further illustrates that day-to-day fluctuations in mental health and wellbeing are well understood through these mechanisms; compassion is one outcome that is less touched by this cascade.
Fig. 8
Schematic representation of the within-person mediation effects in all state variables, using total mindfulness score as the predictor. Rumination and cognitive interference are packaged as (lack of) self-preoccupation; anxiety, depression, stressful events, perceived stress, and negative effect are packaged as mental health; positive affect and flourishing as wellbeing. Line width is proportional to the size of the standardized effect; the size of the mediator to outcome and the size of the direct effects is averaged across variables within each package. Only significant effects are represented
A novel finding from the EMA analysis was that all mediators are ultimately interconnected, both as within-day within-person correlates and as mediators of the effects of mindfulness. That is, each mediator also transmits some of the variance from mindfulness (if any) to the other mediators. The implication is that there is not just a manifold of mechanisms, but also a cascade of effects, where each mediator or mechanism in turn affects the other mechanisms, thus creating a cascade or feedback loop.
This reverberating cascade of multiple mediators likely explains why mindfulness interventions tend to have broad impact across a wide variety of variables: While most of these mechanisms provide a broad coverage of effects, no single mechanism is associated with all effects, but together they are. For instance, neither rumination nor cognitive interference provides links to positive affect, but self-transcendent emotions do.
The cascade may also explain why the effects of state mindfulness, at least in some cases, can extend across multiple days, as shown in Figs. 5, 6, and 7: Day-to-day changes in the mediators themselves trigger changes in other mediators. Especially, the effects of cognitive interference (with its mediating effects extending over three days as a mediator for Nonjudging) and self-compassion (with its mediating effects extending over four days as a mediator for Observing) are long-lasting.
A methodological consequence of this finding is that research singling out any mechanism within this family (and, obviously, many more mechanisms might be at play than those that are investigated here) is likely to find significant mediation, but would be, at the same time, incomplete in its description of mechanism. We suggest that future research studies should adopt a multiple-mechanism approach to more fully explore the interplay among mechanisms.
Because every potential mediator on Day t + 1 was associated with every other potential mediator on Day t + 2, there is no obvious order or sequencing of these mediators, and no prime mover among them. That being said, the cognitive interference variable appears to be the stronger mediator, since it fully mediated the effects of both the mindfulness aggregate and Nonjudging on mental health and flourishing as well as on most of the other mediators (self-compassion for the mindfulness aggregate is the only exception). This, then, would be one possible narrative around these effects: Mindfulness, both in general and specifically its nonjudgmental, accepting aspect, can free the mind from excessive self-preoccupation, which is a recipe for mental health, wellbeing, and even compassion.
A closer look at the data suggests that rumination and cognitive interference tend to have similar effects and tend to be mediators for each other, and so do self-compassion and self-transcendent emotions. This suggests that mindfulness interventions might want to concentrate on these two families of mechanisms (i.e., self-preoccupation on the one hand and self-compassion/self-transcendence on the other). Conversely, if one were looking for an alternative for mindfulness interventions, boosting self-compassion and decreasing self-preoccupation concurrently would likely be a fruitful endeavor.
The results discussed above pertain to the full sample. We additionally examined whether group (i.e., control vs. intervention) moderated the predictor-mediator-outcome relationships. As mentioned above, a lack of moderation would imply that, within the set of variables examined here, mindfulness interventions work through the same mechanisms as day-to-day fluctuations in state mindfulness do. That is, mindfulness interventions would not activate a new mechanism or modify the strength of existing causal pathways; rather, they would simply enhance mindfulness, which then naturally gives rise to the cascade of effects discussed in the previous section. That is indeed what we found here, both for correlations and mediational analysis. For both types of analysis, some correlation coefficients or pathways were moderated by group, but the number of those did not exceed what would be expected under the standard p < 0.05 statistical threshold for statistical significance.
The correlation matrix (Fig. 1) shows that the mindfulness intervention yielded its effects at least partially through the mindfulness practices: The number of minutes spent in meditation and/or the number of minutes spent in informal practice in real life correlated with the state mindfulness variables—a dose–response relationship. This is not surprising (such practices are the means by which mindfulness is traditionally trained), and it is in line with evidence that shows that the maintenance of treatment gain after training is conditional on the continuation of a mindfulness practice (Galante et al., 2021; Mirabito et al., 2025). Both types of findings (the dose–response relationship and the need for maintenance of practice) suggest that mindfulness training is not so much the training of a skill or the changing of an attitude, and more akin to training a muscle—the actual practice is necessary to keep the cascade of effects going, at least within the bounds of this eight-week intervention.
The within-participant mediation effects echo those found in the literature on between-subject data (for rumination and worry, which is akin to cognitive interference, see a meta-analysis by Gu et al., 2015; for a cautiously optimistic review on self-compassion, see Golden et al., 2021); we can now also add self-transcendent emotions to the list of mechanisms. The finding that mindfulness interventions do not change the mediated pathways suggests that this mechanism is universal, that is, independent of whether one is undergoing or underwent a mindfulness intervention or not. Part of the reason why mindfulness interventions work, then, is that they elevate levels of state mindfulness.
Limitations and Future Directions
This study has obvious limitations. First, it was conducted within a population of college students, using a single mindfulness intervention approach tailored to this population. This limits the potential generalizability. Second, the results are almost certainly tinged by the reality of the pandemic, and so they might not generalize to in-person or remote delivery of the MIEA curriculum (or other mindfulness curricula) under more normal circumstances. Third, the study is also limited by our choice of measures—it is possible we missed important mechanisms altogether. Fourth, our across-day analysis imposed a strict (24 h) time-scale on which the mechanisms operate—some mechanisms may be more short-lived but still important. Fifth, one reviewer pointed out, rightfully, that our study did not independently verify that our state version of the FFMQ is indeed valid. We do, however, note that the pattern of correlations between the state measures of mindfulness and the other state measures is very much what one would expect from the correlations between the trait measures of the same constructs. Sixth, we did not obtain measures in-the-moment for meditation or other mindfulness practices. Our probes asked whether participants had practiced each of these “since the last time we checked in”, which may have missed some practice points and introduced noise in the data set. One final concern is the relatively low adherence rate—participants responded on average to 46% of the probes. We ascribe this result to the frequent nature of our probing, the length of each survey, and perhaps also the length of the intervention, which all make it harder for participants to respond to cues in time. For instance, Mirabito and Verhaeghen (2023) probed participants three times a day over a four-week MIEA intervention, measured only eight variables in total, and obtained a 91% adherence rate. Empirically speaking, we might have traded adherence for a data set rich in variables. We feel this trade-off helped us gain important knowledge about mechanism.
Future research studies could fruitfully employ a multiple-mechanism approach to more fully explore the interplay between different mechanisms. Open questions concern the minimal dosing of mindful practices to yield meaningful effects, and how those would be mediated, as well as whether non-mindfulness interventions would indeed yield similar effects. Finally, little is known about how formal and informal mindful practices differentially engage the different mechanisms, which would be directly relevant for clinical practice.
One practical implication of these findings for mindfulness practitioners would be that maintaining or restoring high levels of mindfulness, for instance by engaging in meditation or in more informal mindfulness exercises, would be of obvious and direct benefit to mental health and wellbeing. For the minimalist practitioner, the lagged analyses suggest that daily maintenance might not be necessary. Depending on the facet of mindfulness and the mediator, some of the effects of state mindfulness on the mediating mechanisms are still noticeable two to four days out. Note that we obtained these effects with an intervention that requires minimal homework practice from participants (certainly compared to standard interventions such as MBSR and MBCT which suggest 40 min per day from the very start): a recommended ten to twenty minutes of formal meditation and some additional informal practice each day, with our actual data suggesting the average is closer to four minutes per day. Combined with the lagged analyses, this would suggest that even five minutes every other day might lead to salutary outcomes. Obviously, a closer examination of how different doses and lags would affect the benefits of mindfulness training would be an interesting avenue for further research.
A second practical implication follows from the finding that in many cases the mediators effectively explain all of the variance transmitted from mindfulness. This result implies that non-mindfulness interventions that would effectively boost some of these mediators, such as self-compassion training (which boosts self-compassion; for a meta-analysis, see Wilson et al., 2019) or Cognitive Behavioral Therapy (which is as effective at downregulating ruminations as mindfulness-based interventions are; for a meta-analysis, see Mao et al., 2023), would likely be useful alternatives to mindfulness interventions.
Declarations
Ethics
This study was approved by the Institutional Review Boards of the University of North Georgia and the Georgia Institute of Technology.
Informed Consent
All participants provided a waiver of documentation of consent.
Use of Artificial Intelligence
AI was not used.
Conflict of interest
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Aikman, S. N., Verhaeghen, P., & Ram, N. (2026). Mindfulness from me to we: A randomized control trial on the effects of an 8-week mindfulness intervention on mindfulness, well-being, mental health, compassion, and prosociality. American Journal of Orthopsychiatry, 96(2), 197–209. https://doi.org/10.1037/ort0000828CrossRefPubMed
Baer, R. A. (2003). Mindfulness training as a clinical intervention: A conceptual and empirical review. Clinical Psychology: Science and Practice,10, 125–143. https://doi.org/10.1093/clipsy.bpg015CrossRef
Baer, R. A., Smith, G. T., Hopkins, J., Krietemeyer, J., & Toney, L. (2006). Using self-report assessment methods to explore facets of mindfulness. Assessment,13(1), 27–45. https://doi.org/10.1177/1073191105283504CrossRefPubMed
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology,51, 1173–1182. https://doi.org/10.1037/0022-3514.51.6.1173CrossRefPubMed
Bishop, S. R., Lau, M., Shapiro, S., Carlson, L., Anderson, N. D., Carmody, J., Segal, Z. V., Abbey, S., Speca, M., Velting, D., & Devins, G. (2004). Mindfulness: A proposed operational definition. Clinical Psychology: Science and Practice,11, 230–241. https://doi.org/10.1093/clipsy.bph077CrossRef
Bolger, N., & Laurenceau, J. P. (2013). Intensive longitudinal methods: An introduction to diary and experience sampling research. Guilford press.
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. https://doi.org/10.1037/0022-3514.84.4.822CrossRefPubMed
Brown, K. W., Ryan, R. M., & Creswell, J. D. (2007). Mindfulness: Theoretical foundations and evidence for its salutary effects. Psychological Inquiry,18, 211–237. https://doi.org/10.1080/10478400701598298CrossRef
Cameron, C. D., & Payne, B. K. (2011). Escaping affect: How motivated emotion regulation creates insensitivity to mass suffering. Journal of Personality and Social Psychology,100(1), 1–15. https://doi.org/10.1037/a0021643CrossRefPubMed
Creswell, J. D., & Lindsay, E. K. (2014). How does mindfulness training affect health? A mindfulness stress buffering account. Current Directions in Psychological Science,23, 401–407. https://doi.org/10.1177/0963721414547415CrossRef
Diener, E. (2009). Subjective well-being. In E. Diener (Ed.), The science of well-being (pp. 11–58). Springer.CrossRef
Donnellan, M. B., Oswald, F. L., Baird, B. M., & Lucas, R. E. (2006). The mini-IPIP scales: Tiny-yet-effective measures of the Big Five factors of personality. Psychological Assessment,18(2), 192–203. https://doi.org/10.1037/1040-3590.18.2.192CrossRefPubMed
Enkema, M. C., McClain, L., Bird, E. R., Halvorson, M. A., & Larimer, M. E. (2020). Associations between mindfulness and mental health outcomes: A systematic review of ecological momentary assessment research. Mindfulness,11, 2455–2469. https://doi.org/10.1007/s12671-020-01442-2CrossRefPubMedPubMedCentral
Galante, J., Stochl, J., Dufour, G., Vainre, M., Wagner, A. P., & Jones, P. B. (2021). Effectiveness of providing university students with a mindfulness-based intervention to increase resilience to stress: 1-year follow-up of a pragmatic randomised controlled trial. Journal of Epidemiological Community Health,75(2), 151–160. https://doi.org/10.1136/jech-2020-214390CrossRef
Golden, H. L., Vosper, J., Kingston, J., & Ellett, L. (2021). The impact of mindfulness-based programmes on self-compassion in nonclinical populations: A systematic review and meta-analysis. Mindfulness,12(1), 29–52. https://doi.org/10.1007/s12671-020-01501-8CrossRef
Gotink, R. A., Meijboom, R., Vernooij, M. W., Smits, M., & Hunink, M. G. (2016). 8-week mindfulness based stress reduction induces brain changes similar to traditional long-term meditation practice - A systematic review. Brain and Cognition,108, 32–41. https://doi.org/10.1016/j.bandc.2016.07.001CrossRefPubMed
Goyal, M., Singh, S., Sibinga, E. M. S., Gould, N. F., Rowland-Seymour, A., Sharma, R., Berger, Z., Sleicher, D., Maron, D. D., Shihab, H. M., Ranasinghe, P. D., Linn, S., Saha, S., Bass, E. B., & Haythornthwaite, J. A. (2014). Meditation programs for psychological stress and well-being. JAMA Internal Medicine,174, 357. https://doi.org/10.1001/jamainternmed.2013.13018CrossRefPubMedPubMedCentral
Gu, J., Strauss, C., Bond, R., & Cavanagh, K. (2015). How do mindfulness-based cognitive therapy and mindfulness-based stress reduction improve mental health and well-being? A systematic review and meta-analysis of mediation studies. Clinical Psychology Review,37, 1–12. https://doi.org/10.1016/j.cpr.2015.01.006CrossRefPubMed
Henry, J. D., & Crawford, J. R. (2005). The short-form version of the Depression Anxiety Stress Scales (DASS-21): Construct validity and normative data in a large non-clinical sample. British Journal of Clinical Psychology,44(2), 227–239. https://doi.org/10.1348/014466505X29657CrossRefPubMed
Hölzel, B. K., Lazar, S. W., Gard, T., Schuman-Olivier, Z., Vago, D. R., & Ott, U. (2011). How does mindfulness meditation work? Proposing mechanisms of action from a conceptual and neural perspective. Perspectives on Psychological Science,6, 537–559. https://doi.org/10.1177/1745691611419671CrossRefPubMed
MacKinnon, D. (2012). Introduction to statistical mediation analysis. Routledge.CrossRef
Mao, L., Li, P., Wu, Y., Luo, L., & Hu, M. (2023). The effectiveness of mindfulness-based interventions for ruminative thinking: A systematic review and meta-analysis of randomized controlled trials. Journal of Affective Disorders,321, 83–95. https://doi.org/10.1016/j.jad.2022.10.022CrossRefPubMed
Mirabito, G., Koganti, L. S., & Verhaeghen, P. (2025). One-year follow-up of a remote delivered Koru Mindfulness RCT intervention for college students during the COVID-19 pandemic. Journal of American College Health,73(4), 1306–1309. https://doi.org/10.1080/07448481.2024.2409694CrossRefPubMed
Mirabito, G., & Verhaeghen, P. (2023). Changes in state mindfulness are the key to success in mindfulness interventions: Ecological momentary assessments of predictors, mediators, and outcomes in a four-week Koru Mindfulness intervention. Psychological Reports, , Article 00332941231216899. https://doi.org/10.1177/00332941231216899CrossRef
Moore, R. C., Depp, C. A., Wetherell, J. L., & Lenze, E. J. (2016). Ecological momentary assessment versus standard assessment instruments for measuring mindfulness, depressed mood, and anxiety among older adults. Journal of Psychiatric Research,75, 116–123. https://doi.org/10.1016/j.jpsychires.2016.01.011CrossRefPubMedPubMedCentral
Puterman, E., DeLongis, A., & Pomaki, G. (2010). Protecting us from ourselves: Social support as a buffer of trait and state rumination. Journal of Social and Clinical Psychology,29, 797–820. https://doi.org/10.1521/jscp.2010.29.7.797CrossRef
Quaglia, J. T., Brown, K. W., Lindsay, E. K., Creswell, J. D., & Goodman, R. J. (2015). From conceptualization to operationalization of mindfulness. In K. W. Brown, J. D. Creswell, & R. M. Ryan (Eds.), Handbook of mindfulness: Theory, research, and practice (pp. 151–170). The Guilford Press.
Raes, F., Pommier, E., Neff, K. D., & Van Gucht, D. (2011). Construction and factorial validation of a short form of the self‐compassion scale. Clinical Psychology & Psychotherapy,18(3), 250–255. https://doi.org/10.1002/cpp.702CrossRef
Rogers, H. B. (2016). The mindful twenty-something: Life skills to handle stress... and everything else. New Harbinger Publications.
Rogers, H., & Maytan, M. (2019). Mindfulness for the next generation: Helping emerging adults manage stress and lead healthier lives. Oxford University Press.
Runyan, J. D., Fry, B. N., Steenbergh, T. A., Arbuckle, N. L., Dunbar, K., & Devers, E. E. (2019). Using experience sampling to examine links between compassion, eudaimonia, and pro‐social behavior. Journal of Personality,87, 690–701. https://doi.org/10.1111/jopy.12426CrossRefPubMed
Ruscio, A. C., Muench, C., Brede, E., & Waters, A. J. (2016). Effect of brief mindfulness practice on self-reported affect, craving, and smoking: A pilot randomized controlled trial using ecological momentary assessment. Nicotine & Tobacco Research,18, 64–73. https://doi.org/10.1093/ntr/ntv074CrossRef
Salzberg, S. (2010). Real happiness: The power of meditation: A 28-day program. Workman Publishing.
Shiota, M. N., Keltner, D., & John, O. P. (2006). Positive emotion dispositions differentially associated with Big Five personality and attachment style. Journal of Positive Psychology,1, 61–71. https://doi.org/10.1080/17439760500510833CrossRef
Shoham, A., Goldstein, P., Oren, R., Spivak, D., & Bernstein, A. (2017). Decentering in the process of cultivating mindfulness: An experience-sampling study in time and context. Journal of Consulting and Clinical Psychology,85, 123–134. https://doi.org/10.1037/ccp0000154CrossRefPubMed
Speelman, C. P., Parker, L., Rapley, B. J., & McGann, M. (2024). Most psychological researchers assume their samples are ergodic: Evidence from a year of articles in three major journals. Collabra: Psychology, 10(1), Article 92888. https://doi.org/10.1525/collabra.92888CrossRef
Stawski, R. S., Mogle, J., & Sliwinski, M. J. (2011). Intraindividual coupling of daily stressors and cognitive interference in old age. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences,66B, i121–i129. https://doi.org/10.1093/geronb/gbr012CrossRef
Vago, D. R., & Silbersweig, D. A. (2012). Self-awareness, self-regulation, and self-transcendence (S-ART): A framework for understanding the neurobiological mechanisms of mindfulness. Frontiers in Human Neuroscience, 6, Article 296. https://doi.org/10.3389/fnhum.2012.00296CrossRefPubMedPubMedCentral
Van der Gucht, K., Dejonckheere, E., Erbas, Y., Takano, K., Vandemoortele, M., Maex, E., Raes, F., & Kuppens, P. (2019). An experience sampling study examining the potential impact of a mindfulness-based intervention on emotion differentiation. Emotion,19(1), 123-131. https://doi.org/10.1037/emo0000406CrossRefPubMed
Verhaeghen, P. (2019). The mindfulness manifold: Exploring how self-preoccupation, self-compassion, and self-transcendence translate mindfulness into positive psychological outcomes. Mindfulness,10, 131–145. https://doi.org/10.1007/s12671-018-0959-3CrossRef
Wilson, A. C., Mackintosh, K., Power, K., & Chan, S. W. (2019). Effectiveness of self-compassion related therapies: A systematic review and meta-analysis. Mindfulness,10, 979–995. https://doi.org/10.1007/s12671-018-103CrossRef
