**Stress regulation via self-administered mindfulness and biofeedback intervention for adults: A pre-registered meta-analysis** Stress can influence how we experience emotions, how our body functions, and how we think (Aldwin 2007; Lazarus & Folkman, 1984). When stress gets excessive, it can even compromise people’s everyday functioning (Jiaxuan, Jiali, Yuanyuan, & Wei, 2018). In cases when excessive stress lasts for extended periods of time, it can become an important driver in developing depression (Yang et al., 2015) and/or anxiety disorders (Pynoos, Steinberg, & Piacentini, 1999). Thus, knowing how to regulate stress effectively is vital. An increasing number of people have started relying on medications to stifle the build-up of stress and anxiety (Olfson et al., 2015). But it is not always desirable to utilize pharmacological interventions, as these may come with side effects. Further, pharmacological interventions may simply not be the most cost-effective way to deal with stress. It is thus essential to develop cost-effective (and non-pharmacological) interventions in order to cope with stress (q.v., de Witte, Spruit, van Hooren, Moonen, & Stams, 2019). Two very popular and non-pharmacological ways to regulate stress are **self-administered mindfulness** (a type of meditation that does not require an instructor) and **“biofeedback”** (a self-regulation training based on feedback on physiological mechanisms). Here, we plan to synthesize the evidence on these two strategies via a meta-analysis. Our primary reason to choose these interventions are because they are non-invasive and do not require the presence of other people. However, the choice is also partly arbitrary as the synthesis of these strategies will be a first step towards building a more comprehensive database to understand various stress-regulation strategies and their efficacy. Overall, our motivation to conduct this meta-analysis is to assess the state-of-the-art in stress-regulation research and to provide directions on how to improve stress-regulation research moving forward. To accomplish this goal, we reviewed the existing literature and addressed some important questions: For which components (physiological, emotional, cognitive) underpinning biofeedback and self-administered mindfulness is there adequate empirical support? In addition, are individual differences taken into account when it comes to the efficacy of different stress regulation intervention? Is it possible to identify for whom certain strategies work and for whom they don’t? We intend to shed light on the mechanisms underpinning stress regulation by employing a workflow incorporating various publication bias-detection techniques. To be as inclusive as possible, we also included in our meta-analysis affective states that are consequences of stress. **Stress Regulation** Stress is generally understood as a non-specific response of the body, which occurs when external demands exceed internal resources (Lazarus & Folkman, 1984; Selye, 1956). The response to stress can be thought of as separated in three different components: Affective (Watson, & Clark, 1988), physiological (Schneiderman, Ironson, & Siegel, 2005) and cognitive (Du, Huang, An, & Xu, 2018). We use these components as tools for our meta-analysis; we nevertheless agree with the theoretical position that these different types of response do not really present conceptual distinctions (Pessoa, 2008; Phelps, 2006) and that the systems underlying these three components influence each other during stress (De Witte, Spruit, van Hooren, Moonen, & Stams, 2019; McEwen & Gianaros, 2010). But yet, splitting them into three different categories allows us to better understand the potential applied value of certain stress-regulation strategies. The first of the three components, the affective one, is characterized by feelings of nervousness, strain, and tensions that arise when individuals are overrun by external demands (Ratanasiripong, Ratanasiripong, & Kathalae, 2012). The second, the physiological component, is characterized by an activation of the hypothalamic–pituitary–adrenal axis (HPA axis; Stephens, & Wand, 2012). Physiological responses include, but are not limited to, the activation of the autonomic nervous system, which can be assessed through changes in heart-rate, Heart-Rate Variability, systolic and diastolic blood pressure, skin conductance, and cortisol (Bally, Campbell, Chesnick, & Tranmer, 2003). At the level of the cognitive response, stress has been found to be associated with changes in cognitive functions (which include processes like reflection and rumination; McFarland, Buehler, Von Rüti, Nguyen, & Alvaro, 2007). For the present meta-analysis, we decided to focus on measurements of perseverative thinking and rumination. The lack of a true separation between these components also means that stress does not stand on its own; when stress exceeds what one can handle, longer-term affective consequences may emerge, like depression or chronic anxiety (Cohen, Kamarck, & Mermelstein, 1983). Regulation strategies for stress oftentimes include ways to shield oneself from such longer-term consequences of stress. In his model, Russel (1980) classifies affect (and longer-term affective consequences) into valence (positive vs. negative) and arousal (high vs. low). The former is related to the degree of pleasantness of the affective experience while the latter is related to the level of arousal of the affective experience (Feldman, 1995; Russel, 1980). Crossing these two foci lets us categorize the majority of affective experiences. To downregulate stress people may rely on strategies, like self-administered mindfulness meditation and Heart-Rate Variability biofeedback. Self-administered mindfulness shares features, such as a non-judgmental attitude and an acceptance of inner experience, with other mindfulness protocols. In contrast with other protocols, however, self-administered mindfulness does not require the presence of an instructor, is available 24/7 to people, and tends to be one of the least costly ones (Spijkerman, Pots, & Bohlmeijer, 2016). Self-administered mindfulness can be administered via smartphone applications, audio files, and books which can guide the user through self-administered mindfulness exercises. In the empirical literature, a two week self-administered mindfulness meditation intervention (compared to a passive control group), has been found to influence the affective component by reducing self-reported stress (Cavanagh et al., 2013; Cavanagh et al., 2018). We have found no studies on self-administered mindfulness that address the physiological component (probably due to the fact that self-administered protocols are mostly administered online). For traditional mindfulness interventions, Sanada et al. (2016) found slightly lower cortisol levels after intervention. For what concerns the cognitive component, a brief protocol of self-administered mindfulness compared to a waitlist control, reduced maladaptive cognitions like perseverative thinking (Cavanagh et al., 2018). Finally, a daily guided self-administered mindfulness intervention via an audio CD (compared to a passive control group) has been found to lead to a significant reduction in depression (Barry et al., 2018). The other strategy, biofeedback, is a technique that allows people to have immediate feedback on a specific physiological function (such as heart-rate or muscle tension), that is under the control of the autonomic nervous system (Gartha, 1976). Biofeedback interventions recruit the parasympathetic branch of the nervous system, thereby creating inhibition of the sympathetic action (Jerčić & Sundstedt, 2019). Based on the biofeedback received, people can learn how to change a behavior (e.g., their breathing rate) in order to improve the particular function they are monitoring. Biofeedback is typically delivered via computers or smartphones and is thought to improve self-regulatory capacities (Gross, 1998). For the present meta-analysis we have decided to focus on one type of biofeedback called Heart-Rate Variability biofeedback. Heart-Rate Variability biofeedback is, after all, the most used and thought to be the most efficacious in reducing stress (Lehrer & Gevirtz 2014). In one study, female nursing students participated in a 5-week intervention relying on three biofeedback sessions per day. The experimental (as compared to a passive control) group showed a significant reduction in self-reported levels of stress (Ratanasiripong, Ratanasiripong, & Kathalae, 2012). In another study evidence was more mixed: participants in a biofeedback (compared to an active control group) condition showed improvements in respiratory rate, but no changes in blood pressure were found (Prinsloo et al., 2011). For what concerns the cognitive component, several authors found a negative association between Heart-Rate Variability and perseverative thinking (Eddie, Kim, Lehrer, Deneke, & Bates, 2014; Thayer & Friedman, 2002). Finally, in one study, six sessions of Heart-Rate Variability biofeedback (compared with an active control group) interventions improved symptoms of depression at post testing and at 1 month follow-up (Lin et al., 2019). **Precision in estimating the efficacy of stress regulation** And yet, it is important we understand stress regulation well. It is by now no secret that the psychological sciences have been confronted with a replication crisis (the fact that replication studies have failed to find the same results as original studies; Klein et al. 2018; Maxwell, Lau, & Howard, 2015; Open Science Collaboration, 2015). It is reasonable to assume that some proportion of the literature on stress – like any other literature in psychology – suffers from replication issues. To be able to provide people with advice on appropriate stress-regulation strategies, a meta-analytic assessment is thus needed to assess the extent of how replicable (or not) the literature is (cf., IJzerman et al., 2020). Arguably one of the biggest culprits of the replication crisis is publication bias (a state of affairs in which studies with positive results are more likely to be published than studies with negative results; Rosenthal 1979; Sutton, Duval, Tweedie, Abrams, & Jones, 2000). Past meta-analyses on self-administered mindfulness interventions (Cavanagh, Strauss, Forder, & Jones 2014) and biofeedback (Goessl et al. 2017) have tried to estimate the level of publication bias in the literature. These meta-analyses have found that the given regulation strategies are moderately effective in reducing stress, depression, and anxiety. However, neither meta-analysis has dealt with publication bias adequately. Goessl et al. (2017), for example, addressed publication bias using trim-and-fill, which is known to have an excessive false-positive rate under most realistic conditions (Carter et al., 2019). To deal with publication bias we employ a combination of state-of-the-art publication bias correction methods, assuming a more realistic data-generating process behind the published effects of stress-regulation strategies. In doing so, we followed the workflow of IJzerman et al (2020), assessing the evidential value of the literature using a permutation-based p-curve analysis and by estimating a naive meta-analytic effect size (and its heterogeneity) using a hierarchical random-effect meta-analytic model with robust sandwich-type variance estimation (RVE; Hedges, Tipton, & Johnson, 2010). Lastly, we employed a tandem procedure involving the 4-parameter selection model (McShane, Böckenholt, & Hansen, 2016) and the regression-based PET-PEESE method (Stanley & Doucouliagos, 2014) to try to correct for publication bias as well as possible. We considered an effect to be present only if all of the meta-analytic techniques detected one. This approach was conservative, but allowed to exclude publication bias as a sole explanation of the target effect. Moreover, it also provided a more realistic estimate of the effect that could be expected in close replications of the studies included in this meta-analysis. **Research Overview** To appraise the available evidence on the effects of self-administered mindfulness and biofeedback on stress in a more detailed manner, we have conducted a meta-analysis with the following objectives: 1) to assess the evidential value of identified studies in both literatures, 2) for either regulation strategy, to estimate mean effect sizes for the three components of stress (affective, physiological, cognitive), 3) for either regulation strategy, to estimate the mean effect sizes for the affective consequences of stress, 4) to adjust the target estimates for publication bias using various techniques and 5) to determine whether personality traits were taken into account in stress regulation studies.