Abstract
In recent years, mindfulness-based interventions for the treatment of schizophrenia spectrum disorders (SSD) have yielded promising outcomes. Concurrently, growing evidence supports the mental health benefits of nature exposure, leading to increased interest in nature-based interventions. Nature connectedness—how strongly one feels connected to nature—is associated with improved well-being, particularly stress reduction. Research points toward a reciprocal relationship between mindfulness and nature connectedness, suggesting increased benefits of interventions combining them. Additionally, the neuropeptide oxytocin has gained attention for its potential in alleviating SSD symptomatology. This exploratory study examined the combined effects of mindfulness, nature, and oxytocin on SSD in the context of a mindfulness-based group therapy (MBGT) with a nature-focus. Forty-one (age range 24–64 years) participants with SSD were randomly allocated to two groups and received either intranasal oxytocin (24 I.U. Syntocinon®) or placebo 30 min before they underwent two 50-min MBGT sessions consisting of an introduction, mindfulness exercises, sharing and goal-setting. The Connectedness to Nature Scale (CNS), the Inclusion of Nature in Self Scale (INS) and visual analogue scales for stress were administered before and after sessions. ANCOVA revealed no significant between-group differences; however, within-subject analyses revealed significant increases in nature connectedness and reductions in stress in both groups. Additionally, increases in nature connectedness were significantly associated with reduced stress. While no significant differences were found between the MBGT modules (breathing vs. senses in nature), within-group analyses revealed significant improvements of more outcome measures after the nature-focused session. These findings offer preliminary support for integrating mindfulness, nature, and oxytocin in SSD treatment. Future research with larger samples is needed to clarify their individual and combined effects.
Introduction
Schizophrenia spectrum disorders (SSD) are heterogeneous and encompass complex and severe mental disorders. The broad symptomatology includes positive symptoms such as hallucinations and delusions, and negative symptoms such as social withdrawal, diminished emotional range, alogia, avolition, and anhedonia. Antipsychotic medications are typically the first-line treatment for SSD, and clinical guidelines recommend their continued use even after symptom remission to minimize relapse risk and poor long-term outcomes (Correll, Rubio, & Kane, 2018; Gaebel, Stricker, & Riesbeck, 2020; German Association for Psychiatry, Psychotherapy & Psychosomatics, and DGPPN, 2019; Hui et al., 2018). While antipsychotics are effective in treating positive symptoms (Haddad and Correll, 2018; Leucht et al., 2017), their impact on negative symptoms is less clear, with some studies suggesting that improvements in negative symptoms may be secondary to positive symptom reduction (Krause et al., 2018; Leucht et al., 2017). Cognitive-behavioral therapy is another recommended treatment for SSD, primarily targeting positive symptoms and harmful behaviors associated with delusions and hallucinations (Bighelli et al., 2018; Jauhar, Laws, and McKenna, 2019; Kart, Özdel, and Türkçapar, 2021).
Over the past 30 years, mindfulness-based interventions have gained popularity in Western psychology, also for treating SSD. Mindfulness, defined as nonjudgmental awareness of one’s moment-to-moment experiences (Kabat-Zinn, 2003), has shown potential to decrease negative symptoms and stress, enhance well-being, and increase social functioning in individuals with SSD (Böge et al., 2021; Hodann-Caudevilla, Díaz-Silveira, Burgos-Julián, and Santed, 2020; Jansen, Gleeson, Bendall, Rice, and Alvarez-Jimenez, 2020; Lee, 2019; Louise, Fitzpatrick, Strauss, Rossell, and Thomas, 2018; Özdemir and Kavak Budak, 2022; Shen et al., 2023). The mechanisms underlying these improvements are thought to involve acceptance of distressing psychotic experiences, reduction of negative thinking, and improvements in metacognition, which may mitigate negative symptoms (Böge et al., 2020; Cramer, Lauche, Haller, Langhorst, and Dobos, 2016; Feruglio et al., 2021; İnce and Üçok, 2018; Johnson et al., 2011; Shen, Chen, and Cui, 2020a; Shen et al., 2023).
Nature-based interventions are emerging as promising complementary therapies for individuals with SSD. According to the biophilia hypothesis proposed by E.O. Wilson, humans’ innate affinity for nature, stemming from their shared evolutionary history with animals and plants, explains nature’s salutary effects (Wilson, 1984). Recent studies confirm that nature exposure can positively affect mental and physical health (Bratman, Olvera‐Alvarez, and Gross, 2021; Kotera, Richardson, and Sheffield, 2022; Mygind et al., 2019). Interventions such as forest bathing—which involves sensory immersion in nature—have been associated with reduced anxiety, improved mood, and well-being for both general and clinical populations (Hansen, Jones, and Tocchini, 2017; Keenan, Lumber, Richardson, and Sheffield, 2021; Leibold, 2021; Moeller, King, Burr, Gibbs, and Gomersall, 2018; Rogerson et al., 2020; Wen, Yan, Pan, Gu, and Liu, 2019). A crucial concept underpinning nature-based interventions is nature connectedness, referring to the connection an individual feels with the natural environment. Greater nature connectedness correlates with enhanced well-being, vitality, positive affect, and better stress regulation (Bakir-Demir, Berument, and Akkaya, 2021; Capaldi, Dopko, and Zelenski, 2014; Martin et al., 2020; Pritchard, Richardson, Sheffield, and McEwan, 2020). The urban stress hypothesis supports the therapeutic potential of nature interventions for SSD. Psychosis rates are higher in urban areas than in rural areas, and urban upbringing can increase the risk of developing psychosis (Abrahamyan et al., 2020; Adli and Schöndorf, 2020). Urban stressors, including sensory overload, environmental pollution, and social stress, may contribute to the development of SSD in genetically predisposed individuals (Murray, Bhavsar, Tripoli, and Howes, 2017) and can lead to sensory gating deficits common in SSD (Atagun et al., 2020; Shen et al., 2020b; Xia et al., 2020). Additionally, brain regions showing functional abnormalities in SSD, such as the dorsolateral prefrontal cortex and perigenual anterior cingulate cortex, have been linked to urban stress exposure (Besteher, Gaser, Spalthoff, and Nenadić, 2017; Lammeyer, Dietsche, Dannlowski, Kircher, and Krug, 2019; López-González et al., 2019; Meyer-Lindenberg and Tost, 2019). Nature’s therapeutic effects for SSD may also arise from its impact on self-related processing. Ebisch (2020) distinguishes between intrinsic and extrinsic self-networks, which describe the experience of mental activity as related to oneself, and the perception of oneself as an autonomous entity with agency, respectively. It is proposed that in SSD, these networks function in a fragmented manner, which gives rise to the experience of a disrupted sense of self (Ebisch and Aleman, 2016; Parnas and Zandersen, 2018). Exposure to nature, which offers a dynamic environment supporting mental and bodily self-experiences, may help reintegrate these networks, leading to a more coherent sense of self.
Mindfulness—the practice of sustained, present-moment awareness—appears to strengthen this bond with nature. Evidence suggests a reciprocal relationship, in which mindfulness enhances nature connectedness, while nature fosters mindfulness (Howell, Dopko, Passmore, and Buro, 2011; Muneghina, Van Gordon, Barrows, and Richardson, 2021; Schutte and Malouff, 2018). Interventions that combine mindfulness with nature exposure yield particularly strong effects: Participants in guided outdoor mindfulness walks showed greater mood improvements, lower negative affect, and stronger connectedness to their surroundings compared with those who walked indoors or without mindfulness (Nisbet, Zelenski, and Grandpierre, 2019). These synergistic effects seem to extend to individuals with psychosis. A forest walk followed by contemplative exercises significantly reduced anxiety, depression, anger, and confusion in patients with psychotic disorders (Bielinis, Jaroszewska, Łukowski, and Takayama, 2019), and sensory-focused forest activities improved depressive symptoms and stress in psychiatric inpatients (Kim et al., 2015). Finally, a 10-week nature-based program for individuals having experienced a first episode of psychosis led to increased nature and social connectedness, emotional well-being and calmness (Cuthbert, Sharp, and Berry, 2021).
Oxytocin, a neuropeptide involved in social behavior, has garnered attention for its potential in treating social deficits in SSD (William and Bürkner, 2017). People with SSD tend to have lower endogenous oxytocin levels, linked to poorer social cognition and cognitive impairments (Jobst et al., 2014; Strauss et al., 2019). While some studies show benefits of exogenous oxytocin on SSD symptoms, results are inconsistent, possibly due to differences in study design and individual responses (Bradley and Woolley, 2017). For instance, combining oxytocin with social skills training for SSD showed no added effect in one study (Buchanan et al., 2021), but meta-analytic evidence suggests it may have selective effects on high-level social cognition such as theory of mind, eye gazing and perspective taking (Bürkner, Williams, Simmons, and Woolley, 2017; De Coster, Lin, Mathalon, and Woolley, 2019; Halverson, Jarskog, Pedersen, and Penn, 2019; Pedersen et al., 2011; Shilling and Feifel, 2016; Woolley et al., 2014). According to the Social Salience Hypothesis, oxytocin modulates the salience of social cues, with effects being dependent on contextual aspects (Shamay-Tsoory and Abu-Akel, 2016).
Whether intranasal oxytocin reaches behaviorally effective central nervous system levels remains debated (Leng and Ludwig, 2016), yet growing evidence shows it crosses the blood–brain barrier with corresponding effects (Coleman et al., 2025; Erdő, Bors, Farkas, Bajza, and Gizurarson, 2018; Quintana et al., 2021). In SSD, the administration of oxytocin has shown benefits such as stress reduction when combined with mindfulness-based interventions as a positive social context (Zierhut et al., 2024). Mindfulness can influence oxytocin levels (Bellosta Batalla et al., 2020; Böge et al., 2024), and oxytocin can promote mindfulness-related traits like self-compassion (Wang et al., 2019), suggesting synergistic effects of their combination. Mindfulness in group settings may enhance oxytocin’s effects by providing a safe, socially supportive context. Simultaneously, oxytocin may facilitate engagement and receptivity to mindfulness training.
Combining mindfulness with nature-based interventions might further amplify benefits (Bielinis et al., 2019; Cuthbert et al., 2021; Kim et al., 2015), as mindfulness fosters nature connectedness and oxytocin, by potentiating the effects of mindfulness could additionally promote nature connectedness. Building on this evidence, the integration of both nature and oxytocin with mindfulness may lead to positive outcomes regarding stress reduction.
Oxytocin has been proposed to regulate behaviors that foster social connectedness (Campbell, 2010; Itskovich, Bowling, Garner, and Parker, 2022; Melton, Hodge, Haron, and Boccia, 2024), which—defined as a subjective sense of connection and oneness with humanity (Lee and Robbins, 1995; Townsend and McWhirter, 2005)—strongly correlates with nature connectedness (Aspy and Proeve, 2017).
Mindfulness likewise enhances social connectedness (Amel, Manning, and Scott, 2009; Gootjes and Rassin, 2014; Tipsord, 2009), suggesting that oxytocin, particularly when combined with mindfulness, may improve nature connectedness, potentially through its effects on social connectedness. Figure 1 provides a schematic overview of these proposed relationships.

Triadic model of relationship between mindfulness, oxytocin and nature connectedness. Bidirectional, reinforcing relationship of mindfulness and oxytocin and mindfulness and nature connectedness. Unidirectional, reinforcing effect of mindfulness on social connectedness. Reinforcing effect of oxytocin on nature connectedness, mediated by social connectedness.
The concept of the ecological self —an identity extending beyond the individual ego to include other beings and nature—provides a useful framework when considering the relationship between oxytocin and nature. Oxytocin may promote interconnectedness and soften self–other boundaries, as shown in studies demonstrating its role in blurring self–other distinctions (Pfundmair, Rimpel, Duffy, and Zwarg, 2018; Yue, Xu, Xue, and Huang, 2020; Zhao et al., 2016). Oxytocin can also increase self-reported spirituality, defined as a sense of relatedness and interconnectedness with the world and all living beings (Van Cappellen, Rimé, and Saroglou, 2014; Van Cappellen, Way, Isgett, and Fredrickson, 2016). Oxytocin may therefore facilitate psychological processes aligned with the ecological self, promoting a more relational and nature-connected sense of identity.
This pilot study examined the effects of oxytocin on stress and nature connectedness in the positive social context of mindfulness-based group therapy (MBGT) in individuals with SSD. We hypothesized that oxytocin plus MBGT would increase nature connectedness and reduce stress compared with MBGT with placebo. As a feasibility and acceptability study, only two MBGT sessions were conducted, focusing on social context rather than MBGT’s therapeutic effects, which typically require eight sessions (Böge et al., 2022). We also compared two MBGT modules, expecting that a nature-focused sensory module would produce greater improvements than a breathing-focused module. As breathing awareness is a well-established element of many mindfulness-based interventions (Blanck et al., 2018; Zhang, Lee, Mak, Ho, and Wong, 2021), it provides an appropriate reference for evaluating nature-focused mindfulness.
Materials and Methods
Ethical and regulatory compliance
The study was conducted in compliance with the ethical standards of the Declaration of Helsinki and relevant EU and national regulations. The study protocol received approval from the ethical committee of the Charité–Universitätsmedizin Berlin (EA4/196/19) and was registered at clinicaltrials.org (NCT05247151). Time constraints related to the COVID pandemic only allowed registration after the completion of recruitment and data collection. Preparation and handling of the intranasal sprays followed Good Clinical Practice and Good Manufacturing Practice standards under the supervision of a certified pharmacist. The study focused more on a mechanistic and less on a therapeutic understanding.
Design and randomization
A double-blinded (participants, psychotherapists), placebo-controlled, randomized pilot trial was employed, comprising the administration of intranasal oxytocin versus placebo combined with MBGT. Participants were randomly allocated to the experimental condition (oxytocin + MBGT) or the placebo condition (placebo + MBGT). Randomization was performed by independent researchers with a 1:1 scheme with a fixed block size and occurred on an individual level. Although participants were not explicitly matched before allocation, the randomization procedure ensured balanced distribution of oxytocin and placebo participants across MBGT groups. Mixed-group allocation was handled sequentially at study entry while maintaining full blinding. Post-hoc analyses showed no significant group differences in demographic or clinical variables, including age, gender, and diagnostic distribution (see Table 1).
Sociodemographic Variables at Baseline for Both Groups
P-values are based on Chi-square tests for categorical and t-tests for continuous variables; oxytocin: Oxytocin Spray; Placebo: Saline Solution; SD: standard deviation; df: degrees of freedom. aChi-square test; bIndependent samples t-test; *F20: Schizophrenia, F22: Persistent Delusional Disorders, F23: Acute and transient Psychotic Disorders, F25: Schizoaffective Disorders.
Participants engaged in two MBGT sessions over the course of 1 week. All MBGT groups were mixed, containing participants in the experimental and the placebo condition. Nasal sprays were administered by physicians to all participants at two time points (T0, T2). Negative symptoms, nature connectedness and stress were assessed through validated self-rated psychometric questionnaires. Assessments took place at baseline (T0) before the first session and after the treatment period (T3), and after the first (T1) and before the second session (T2) to measure within-session effects. Figure 2 presents an overview of the study design.

Study design. Baseline (T0) measures: Sociodemographics; CNS, Connectedness to Nature Scale; INS, Inclusion of Nature in Self Scale; PANSS, Positive and Negative Syndrome Scale; Stress: Visual analogue scale. Postintervention (T1), Preintervention (T2), Postintervention (T3) measures: CNS, INS, stress.
Procedures
Oxytocin and placebo
At T0 and T2, participants received either a nasal spray containing oxytocin (24 I.U. Syntocinon®) or a matching placebo containing the same ingredients as the oxytocin spray except for the active compound. To account for oxytocin’s effect latency (30–80 min) and its peak efficacy window (up to 150 min; Gossen et al., 2012), the sprays were administered by physicians 30 min prior to each session. This timing was selected to enhance the potential for reinforcing positive experiences in the context of the MGBT in line with the social salience hypothesis (Shamay-Tsoory and Abu-Akel, 2016).
The sprays were indistinguishable regarding their appearance and taste. The chosen dosage reflected optimal effects on social cognition according to the literature (Halverson et al., 2019; Peled-Avron, Abu-Akel, and Shamay-Tsoory, 2020) and the delivery strategy was utilized for its demonstrated safety (Bartholomeusz, Ganella, Labuschagne, Bousman, and Pantelis, 2015).
MBGT
Participants engaged in two 50-min manual-based MBGT sessions in small groups of three to six individuals. MBGT has been developed iteratively and participatorily by our research group and offers an innovative treatment program that integrates mindfulness and group-based interventions for SSD, with no expected adverse effects (Böge et al., 2020, 2021; Böge and Hahn, 2021). The program has been shown to promote social cohesion and reduce loneliness among individuals with SSD and was therefore believed to create a positive social context (Böge et al., 2020). Each of the two sessions was facilitated by two psychotherapists and followed a standardized procedure tailored to the study context, commencing with an introduction, followed by 15 min of mindfulness exercises, a 10 min inquiry process for sharing experiences, and 5 min for individual goal-setting. The first session introduced mindfulness concepts and breathing awareness, while the second session focused on sensory-based exercises in nature, and a mindful nature walk, ending with participant reflections. Both sessions were conducted outdoors in a natural setting and participants were seated in a circle, fostering a collaborative and engaging environment.
Participants
Between July and October 2021, participants were recruited from the outpatient clinic at Charité–Universitätsmedizin Berlin, Campus Benjamin Franklin, Germany by a multiprofessional team of clinical psychologists, physicians and nurses. Inclusion criteria required participants to be aged 18–65, to meet SSD diagnostic criteria (ICD-10), to have adequate German proficiency and to have no recent (<6 weeks) changes in psychopharmacologic medication. Exclusion criteria included a Positive and Negative Syndrome Scale positive scale score > 6 (Kay, Fiszbein, and Opler, 1987), acute suicidality (Clinical Decision Support System, CDSS item 8 > 1 (Horrocks, Michail, Aubeeluck, Wright, and Morriss, 2018)), neurological disorders or current substance use except nicotine. After accounting for dropouts, the final sample size for our pilot study was 41 participants. This sample size is in line with recommendations for pilot studies that advocate for a total sample size of at least 20 participants (Cocks and Torgerson, 2013) and was achievable within the available recruitment period. See Figure 3 for an overview of the CONSORT flow diagram. All participants provided written informed consent and received 30€ upon completion of the final assessment.

CONSORT flow diagram of recruitment process. *Multiple options are possible per participant.
Assessment
Symptom severity
To provide a description of symptom severity in the patient sample at baseline, positive and negative symptoms were assessed by a clinician using the PANSS subscales. Each scale contains seven statements on a seven-point Likert format from 1 (absent) to 7 (extreme) (Kay et al., 1987). Due to the brief one-week data collection period, postintervention measurements of symptom severity were not conducted, as no significant changes in this timeframe were expected.
Primary outcomes
Negative symptoms as primary outcomes of the current study have already been reported (Zierhut et al., 2024). For the present paper, the pre-specified secondary outcomes of nature connectedness and stress of the original study, constitute the primary focus of this analysis.
Nature connectedness
Nature connectedness was assessed by the Connectedness to Nature Scale (CNS) (Cervinka, Röderer, and Hefler, 2012; Mayer and Frantz, 2004), a multi-item scale designed to measure an individual’s affective and experiential connection to nature (Supplementary Appendix A1). Answers are given in a five-point Likert format from 1 (strongly disagree) to 5 (strongly agree). Instructions were adapted to ask for participants’ momentary experience. The Inclusion of Nature in Self (INS) scale additionally assessed participants’ degree of connection to the natural world (Supplementary Appendix A2). The INS is a graphical, single item measure, comprising a series of seven pairs of circles labeled “self” and “nature” that are overlapping to different degrees (Schultz, 2001).
Stress
Acute stress (general stress and symptom-related distress) was assessed using self-report visual analogue scales with seven circles of increasing size representing different levels of perceived stress (Jacobsen, Richardson, Harding, and Chadwick, 2019; Supplementary Appendix A3). General stress and symptom-related distress were assessed with the following questions: How stressed do you feel at the moment? and How distressing do you find unwanted thoughts, images, or voices at the moment? Alongside the INS, the stress scale was administered immediately before and after each MBGT session to capture participants’ momentary experience.
Data and safety monitoring
In line with the Federal Data Protection Act of Germany, data collection and management were conducted pseudo-anonymously with electronic case report files (eCRF) software. Data were stored on secure servers hosted by the Charité–Universitätsmedizin Berlin.
Statistical analysis
Statistical analyses were performed using IBM SPSS Statistics Version 26 Mac. This pilot study estimated effect sizes for between- and within-group effects, including corresponding confidence intervals. Mean group differences in baseline scores (T0) were calculated with Chi-square tests and independent samples t-tests. Between-group changes in stress and nature connectedness for the whole intervention as well as for both MBGT modules separately were assessed with three separate 2 × 2 mixed-model ANCOVA designs (for both modules, module 1 and module 2), with group (oxytocin vs. placebo) as the between-group factor and time (pre-/post-intervention) as the within-group factor. Respective baseline scores were controlled for as covariates due to the small sample size. Paired samples t-tests were used to assess within-group changes for the whole intervention and for both MBGT modules separately, as well as to compare the change scores of stress and nature connectedness between both modules. Relationships between the changes in nature connectedness and changes in stress were analyzed by linear regression models. In each model, change in nature connectedness was used as the independent variable (predictor), and either change in stress or change in symptom-related distress served as the dependent variable (outcome). As this study is a pilot study, no a priori power analysis was conducted, as this is not mandatory for pilot studies according to Kunselman and Fertility and sterility (2024). Given the exploratory nature of the study, no alpha correction for multiple testing was performed. Significance level was set at α = 0.05 (two-tailed). Analyses were conducted per-protocol, only including participants who completed both modules and all assessment time points. As some participants did not complete all questionnaires at each assessment point, analyses were conducted using all available data for each outcome measure, resulting in slightly varying sample sizes across analyses (see Table 3). No imputation was performed.
Results
Sample description
The final sample consisted of N = 41 participants, with n = 22 in the oxytocin and n = 19 in the placebo condition (Fig. 2). An overview of sociodemographic and clinical data is shown in Table 1. No significant differences were observed between both conditions on baseline measures. Mean scores for nature connectedness and stress for all timepoints are displayed in Supplementary Appendix A4.
Between- and within-group comparisons
A summary of between- and within-group comparisons from baseline to postintervention (T0 to T3) and for module 1 (T0 to T1) and module 2 (T2 to T3) can be found in Tables 2 and 3. Means and SDs for all outcomes and timepoints per group can be found in the Supplementary Table S1. Relevant findings are summarized again below.
Between-Group Changes from Baseline to Post-Intervention (T0 to T3) and per Module (T0 to T1 and T2 to T3)
CNS, Connectedness to Nature Scale; INS, Inclusion of Self in Nature; T, Time point; df, degrees of freedom; η2p, partial eta-squared; Between-group test: ANCOVA with respective baseline score as covariate.
Within-Group Comparisons from Baseline to Postintervention (T0 to T3) and per Module (T0 to T1 and T2 to T3)
CNS, Connectedness to Nature Scale; INS, Inclusion of Self in Nature; T, Time point; CI, 95% confidence interval of the difference T3 – T0; Δ: mean difference score (T3 – T0, T1 – T0, T3 – T2); d: Cohen’s d; Within-group test: paired-sample t-test.
Nature connectedness
No between-group effects were found for CNS and INS at T3, F(1, 31) = 0.21, p = 0.65, η2p = 0.01; F(1, 29) = 0.02, p = 0.88, η2p = 0.00. Post-hoc within-group t-testing indicated significant increases from T0 to T3 of CNS, t(17) = 2.26, p < 0.05, d = 0.53 and INS, t(17) = 2.19, p < 0.05, d = 0.52 in the oxytocin group. No significant within-group changes in CNS and INS scores were observed in the placebo group from T0 to T3, t(15) = 1.42, p = 0.18, d = 0.35; t(13) = 1.59, p = 0.14, d = 0.43.
General stress and symptom-related distress
No between-group effects were found for stress at T3, F(1, 34) = 0.52, p = 0.48, η2p = 0.02; F(1, 34) = 1.60, p = 0.22, η2p = 0.05. Post-hoc within-group t-testing indicated a significant decrease of general stress in the oxytocin group from T0 to T3, t(20) = −4.99, p < 0.001, d = −1.09, and a significant decrease of symptom-related distress in the placebo group from T0 to T3, t(15) = −2.54, p < 0.05, d = −0.63. Changes of general stress in the placebo and symptom-related distress in the oxytocin group were not significant, t(15) = −1.93, p = 0.07, d = −0.48; t(20) = −1.73, p = 0.10, d = −0.38. Figure 4 visualizes the changes for both groups from T0 to T3.

Change in means for Oxytocin and Placebo Group from T0 to T3. CNS, Connectedness to Nature Scale; INS, Inclusion of Nature in Self Scale; SRD, Symptom-related Distress. Changes in mean scores for the Oxytocin and Placebo Group from T0 to T3.
Relationship between nature connectedness and stress
Linear regression analysis showed that, regardless of group, increases in nature connectedness from T0 to T3 significantly predicted decreases in general stress, R2 = 0.19, F(1, 29) = 6.73, p < 0.05, but not in symptom-related distress, R2 = 0.005, F(1, 29) = 0.14, p = 0.72. When analyzed separately by module, increases in nature connectedness predicted decreases in general stress in both module 1, R2 = 0.13, F(1, 30) = 4.37, p < 0.05 and module 2, R2 = 0.18, F(1, 30) = 6.42, p < 0.05. Notably, in module 2, nature connectedness additionally predicted reductions in symptom-related distress, R2 = 0.13, F(1, 31) = 7.12, p < 0.05, but not in module 1, R2 = 0.06, F(1, 30) = 1.75, p = 0.20.
Comparison of both MBGT modules (1 and 2)
Paired samples t-tests with change scores of both modules indicated that the changes in general stress, symptom-related distress, CNS and INS did not differ significantly between MBGT module 1 (focus on breathing) and MBGT module 2 (focus on senses in nature), t(31) = 0.58, p = 0.56, d = 0.10; t(32) = 0.00, p = 1.00, d = 0.00; t(33) = −0.89, p = 0.38, d = −0.15 and t(25) = −0.74, p = 0.46, d = −0.15.
ANCOVAs conducted for both modules separately revealed no significant effects between groups in either module (see Table 2). To assess within-subject effects, paired samples t-tests were conducted for both modules separately. For module 1 (T0 to T1), significant increases of CNS, t(17) = 2.17, p < 0.05, d = 0.51 and significant decreases of general stress, t(20) = −5.06, p < 0.001, d = −1.10), and symptom-related distress, t(20) = −4.11, p < 0.01, d = −0.90) could be shown in the oxytocin group. No significant changes were found for INS, t(17) = 0.90, p = 0.38, d = 0.21. The placebo group showed no significant changes for module 1 on CNS, t(16) = 0.97, p = 0.34, d = 0.24, INS, t(13) = 0.86, p = 0.41, d = 0.23, general stress, t(15) = −1.46, p = 0.17, d = −0.36, and symptom-related distress, t(15) = −0.90, p = 0.38, d = −0.23. For module 2 (T2 to T3), the oxytocin group showed significant increases of CNS, t(18) = 4.13, p < 0.05, d = 0.95 and INS, t(14) = 3.76, p < 0.05, d = 0.97, with larger effect sizes than in module 1, and significant decreases for general stress, t(18) = −4.65, p < 0.001, d = −1.07 and symptom-related distress, t(19) = −2.37, p < 0.05, d = −0.53. The placebo group showed significant increases of INS, t(13) = 4.09, p < 0.05, d = 1.06, and significant decreases of general stress, t(16) = −3.67, p < 0.01, d = −0.89 and symptom-related distress, t(16) = −2.76, p < 0.05, d = −0.67 for module 2, and no changes of CNS, t(16) = 1.48, p = 0.16, d = 0.36. Supplementary Figures S1 and Figure S2 (Supplementary Appendix A5) visualize changes for both groups from T0 to T1 and T2 to T3.
Discussion
The present study aimed at investigating the effects of intranasal oxytocin administration within an MBGT intervention on nature connectedness and stress in individuals with SSD. Additionally, the differences in effectiveness between two MBGT modules focusing on breathing and nature, respectively, were analyzed.
Nature connectedness
Recent studies suggest a mutually enhancing relationship between mindfulness and oxytocin (Bellosta Batalla et al., 2020; Qi, Bing, and Lai, 2020; Wang et al., 2019). It was expected that combining oxytocin with MBGT would increase nature connectedness compared with MBGT alone. No significant differences were found between groups postintervention in CNS and INS scores. Within-group changes indicated significant increases for the oxytocin group over the full intervention but not for the placebo group. This points toward a potential added benefit of oxytocin, which would align with previous research suggesting that oxytocin may enhance mindfulness effects by improving social salience within positive group settings (Böge et al., 2020; Shamay-Tsoory and Abu-Akel, 2016). The lack of significant between-group differences may reflect the limited statistical power resulting from the small sample size in this pilot study. Furthermore studies with larger groups are warranted to explore this relationship.
Stress
The oxytocin group was hypothesized to show greater stress reduction than the placebo group post-MBGT, as oxytocin has shown stress-mitigating effects in healthy individuals (Love, 2018). Between-group differences, however, were not significant. Within-group analyses showed significant reductions in general stress in the oxytocin group and significant reductions in symptom-related distress in the placebo group across the full intervention. This aligns with MBGT’s known effects on stress reduction in both healthy individuals and in SSD (Bellosta Batalla et al., 2020; Engert, Kok, Papassotiriou, Chrousos, and Singer, 2017; Hodann-Caudevilla et al., 2020; Schumer, Lindsay, and Creswell, 2018).
Relationship between nature connectedness and stress
Based on previous findings confirming the stress-reducing effects of nature connectedness (Bakir-Demir et al., 2021; Capaldi et al., 2014), it was hypothesized that increases of nature connectedness would predict reductions of stress and symptom-related distress. This hypothesis was confirmed for general stress across the full intervention and specifically within module 1. However, only module 2 showed a significant predictive relationship between nature connectedness and symptom-related distress, which suggests a stronger stress-reducing effect when a nature element is integrated into MBGT.
Comparison of MBGT modules (breathing vs. nature)
Based on prior findings suggesting that mindfulness and nature connectedness reinforce each other, module 2, which involved sensory exercises in nature, was expected to lead to greater increases of CNS and INS and reductions of stress and symptom-related distress than module 1, which focused on breathing without explicit nature-engagement (Muneghina et al., 2021; Nisbet et al., 2019; Schutte and Malouff, 2018). Differences in change scores of nature connectedness and stress between both modules, regardless of group, were not statistically significant. Between-group comparisons for each module also showed no significant differences. However, when analyzing within-group changes, module 2 led to significant increases in INS scores in both the oxytocin and the placebo groups. CNS scores increased significantly only in the oxytocin condition across both modules, with stronger effects in module 2, suggesting that this module fostered stronger nature connectedness.
In module 1, only the oxytocin group showed significant reductions in general stress and symptom-related distress. In contrast, module 2 led to significant decreases in both outcomes across both conditions, though effect sizes remained larger in the oxytocin group. These findings suggest that the integration of nature in module 2 may enhance the stress-reducing effects of MBGT, particularly when combined with oxytocin. However, since module 2 was delivered second, order-related effects cannot be ruled out. Familiarization with the procedure and the group as well as cumulative practice effects might be additional causes for the stronger improvements after the second module, rather than the specific content of the module itself. The absence of significant between-group effects highlights the need for further studies with larger samples to explore the potential synergistic effects of nature, mindfulness, and oxytocin on stress reduction.
Strengths and limitations
Strengths include the randomized, placebo-controlled, double-blinded study design and the comparability of both groups across all baseline measures. The quality of MBGT sessions was ensured by experienced therapists and a peer-reviewed manual. This pilot study provides initial insights into the effects of combining MBGT with a nature-focus with oxytocin, to be explored in larger future RCTs. Given the exploratory nature of this study, only two MBGT sessions were conducted, and no correction for multiple testing was applied. Standard MBGT protocols for SSD suggest a 4-week intervention period with three sessions per week (Böge et al., 2021; Böge and Hahn, 2021). Future studies could adopt this protocol and combine it with oxytocin and nature. The small sample size limits the interpretability of within-group effects and warrants caution when interpreting the findings. Future studies should additionally address this by using larger, adequately powered samples. For a future between-subject study, a priori power analysis using G*Power (version 3.1.9.7, Faul, Erdfelder, Buchner, and Lang, 2009) for an ANCOVA with two groups and a medium effect size (Cohen’s f = 0.25), α = 0.05, and 1 − β = 0.80 indicated a required sample size of N = 128 per group (total n = 256). Limitations further include the use of individual-level randomization, which may have introduced contamination effects. Future research should consider group-level randomization, extend the duration of the intervention, and incorporate follow-up assessments to better evaluate the stability and generalizability of observed effects. Additionally, to capture participants’ individual experiences, qualitative assessments should be incorporated.
Conclusions
This pilot study suggests that combining MBGT with a nature focus and oxytocin administration may benefit individuals with SSD. Increases in nature connectedness and reductions in stress were observed regardless of oxytocin, though within-group effects indicate these benefits may be more pronounced in the oxytocin group. However, no significant between-group effects emerged.
Importantly, nature connectedness was found to predict reductions in stress, emphasizing nature’s role in fostering mental health. These findings support exploring MBGT enhancements through both natural settings and oxytocin in future studies with larger samples. Clinically, this could translate into holding therapy sessions outdoors, integrating mindfulness into standard treatment, and considering oxytocin as an adjunct intervention.
Future research should also explore the long-term effects of this integrative approach and employ real-time tracking to assess individual experiences, as the full therapeutic impact may unfold over several months. These preliminary findings underscore the potential of combining mindfulness, nature and oxytocin to support well-being and alleviate SSD symptoms.
Authors’ Contributions
M.Z., N.H., K.B., E.H., and MalekB. formulated the research questions. M.Z., N.H., V.E.R., K.B., N.B., and I.H. designed and carried out the study. M.Z., N.H., V.E.R., K.B., N.B., I.H., A.B., J.K., and S.R. conducted the data analyses. M.Z. and N.H. wrote the article. MareikeB., T.T., S.R., MalekB., and E.H. collaborated on the design and editing of the article. All authors contributed to the article and approved the submitted version.
Footnotes
Data Availability
The data and analysis syntax can be made available upon request.
Declaration of Competing Interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this article.
Supplemental Material
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
