Characterizing Nature Videos for an Attention Placebo Control for MBSR: The Development of Nature-Based Stress Reduction (NBSR)

Abstract

Objective:

Various active control interventions for Mindfulness-Based Stress Reduction (MBSR) have been developed, though many can fall short in controlling for non-specific or placebo effects. We developed a Nature-Based Stress Reduction (NBSR) program based on previously reported positive results from virtual natural environment exposure on mental health. In the present study, we validated the nature video component of NBSR as an attention-matched activity equivalent to the formal meditation practice components of MBSR.

Methods:

Key components of MBSR were matched in NBSR to ensure equality in structure, duration, contacts, and intensity, but not in specific active components (i.e., mindfulness meditation). MBSR meditation during group sessions was matched in NBSR by videos consisting of scenes of nature. Each video had a ratio of pleasant, unpleasant, and neutral valence clips consistent with the documented heterogeneity of affective experiences during mindfulness meditation. Amazon Mechanical Turk Workers (n=127) rated emotional valence and self-reported arousal for each video clip.

Results:

Mean valence ratings significantly differed between the three categories of nature videos clips using an ANOVA test (p<0.001). Follow-up pairwise t-tests revealed significant differences between valence ratings for pleasant vs. unpleasant (p<0.001), neutral vs. unpleasant (p<0.001), and pleasant vs. neutral (p<0.01).

Conclusion:

The subjective experience of NBSR nature videos was reported as pleasant, with higher variability reported for unpleasant clips. This pattern generally parallels the variability and heterogeneity of subjective experiences during mindfulness meditation. These findings demonstrate that the nature video component of NBSR provides a promising attention- and valence-matched placebo activity unrelated to mindfulness meditation.

Introduction

Mindfulness, defined as “paying attention, on purpose, to the present moment in a non-judgmental way”, is a prominent active element of the 8-week group program, Mindfulness-Based Stress Reduction (MBSR) (Kabat-Zinn, 1990). In this program, mindfulness is cultivated through formal and informal practices that engage awareness of present moment experience. Participants are trained in the naturalistic observation of arising phenomena, as well as the acceptance of emotional, physical, or situational states that accompany waking consciousness from moment-to-moment (Grossman et al., 2004). MBSR is the most widely used group intervention for mindfulness research and has been linked to significant improvements in overall health and wellbeing, leading to reductions in stress, anxiety, depression, and chronic pain, as well as enhancements in emotional and behavioral regulation (Grossman et al., 2004). However, a number of studies investigating the benefits of mindfulness-based interventions (MBI’s) are compared with either waitlist control (WC) or treatment as usual (TAU) (Goyal et al., 2014), and lack substantial evidence for mindfulness as the active ingredient in MBIs compared to controls in ways that account for non-specific benefits (e.g., attention, sharing in groups) and/or placebo effects (e.g. improvement from participation in healing rituals). (Fjorback et al., 2011; Shapiro et al., 2006; Kaptchuk, 1998).

Therefore, randomized control trials (RCTs) that utilize attention placebo control (APC) groups as comparators are necessary to validate the efficacy of mindfulness as the “active” component of treatment compared to other non-specific effects. (Chalmers et al., 1981; Kaptchuk, 1998). Unlike pharmacological studies, the “dose” of the active component in MBSR is dependent on the individual’s effort, intention, expectation, etc., and therefore requires a matched control intervention in engagement and structure to establish internal validity and to establish the unique contribution of mindfulness.(Rosenkranz et al., 2019). Attention placebo control (APC) groups are a subset of attention and placebo controls that are likely to have impact on patients’ symptoms, but lack the core, active element(s) of treatment under study (e.g., the practice of mindfulness) (Aycock et al., 2018). In order to compensate for the inability to have double- blinding as would be possible in a pharmacological RCT, implementing a methodologically strong APC requires a convincing therapeutic rationale that guarantees similar distribution across study arms of all (or most) non-specific effects. The APC needs to carefully try and equalize components such as duration, contacts, intensity, relationship building, and expectation of benefit, etc. (Popp & Schneider, 2015; Stewart-Williams & Podd, 2004) as compared to the active intervention. Only carefully matching of these components can properly determine the primary mechanism driving the psychosocial and/or behavioral changes of MBSR.

Our team has created a Nature-Based Stress Reduction program as a credible APC group to test the efficacy of the mindfulness component of MBSR training. This study involves the validation of the nature video component of the NBSR intervention. We assess the subjective valence and affective arousal produced by the nature videos; we chose these measures because previous research has demonstrated that daily time spent in meditation has effects on subjective measures of valence (subjective ratings of pleasantness/unpleasantness) and affective arousal (engagement with salient or evocative stimuli) and was significantly associated with increased instances of low-arousal positive emotions (e.g. feelings of quiet and calm) (Jones et al., 2018). The way that experiential practices in group impact subjective experience and arousal levels has the potential to enhance expectation of benefit and impact group process. Therefore, in addition to matching on duration, contacts, and intensity, subjective valence and arousal levels are key measures to decipher the level of congruency between the mindfulness meditation (MM) arm and the control arm, Nature-Based Stress Reduction (NBSR).

Part of the rationale for using a NBSR treatment as an APC for mindfulness derives from multiple reports suggesting exposure to nature and green environments comprehensive health benefits, including mental well-being (Bratman et al., 2012; Hunter et al., 2019). Similar emotional states and accompanying thought patterns confronted during meditation are likely to arise during exposure to nature, which “is so fascinating that it calls for soft attention”(Djernis et al., 2019), and possibly contains a similar level of expectation bias for positive effect (Djernis et al., 2019). Previous research indicates that exposure to nature is associated with a reduced stress response (Mao et al., 2012) and enhanced immune functioning (Kuo, 2015). Furthermore, it has been reported that increased presence of neighborhood green spaces are correlated with decreased depression and anxiety symptoms (Beyer et al., 2014). Exposure to nature using all 5 senses has been associated with activation of the parasympathetic nervous system and a heightened awareness that leads to a state of relaxation (Hansen et al., 2017). A further proposed psychological pathway through which nature can influence health is attention restoration, which can reduce mental fatigue (Berman et al., 2008). However, with growing urbanization, not everyone has access to nature. The predominant colors in nature (green, blue, brown), the sounds (bird song, water/waves, sound of the wind), the scents and odors of nature, and the light intensity belong to the cues we receive from nature that may contribute to better health and well-being (Depledge et al., 2011). For example, olfactory stimulation by the Hinoki cypress leaf oil was shown to be sufficient to achieve the health benefits of Forest Bathing (physiological relaxation, improvements in immune functions) (Song et al., 2016). Virtual environments give the possibility of taking advantage of the audiovisual stimuli we receive from natural environments. Recent technical developments such as virtual reality (VR) can provide access to the audiovisual elements of nature through 360-degree videos (Browning et al., 2019). While, not a full immersion into all of the sensory dimensions provided by a nature experience (e.g. olfactory, somatosensory, gustatory), VR-based nature exposure has been shown to provide emotional well-being benefits for people who cannot access the outdoors (Browning et al., 2019; Depledge et al., 2011).

Previous literature suggests that exposure to nature and mindfulness practices may share common mechanisms (Djernis et al., 2019). Attention is one cognitive process that may be responsible for the positive effects of both mindfulness practice (Shapiro et al., 2006) and exposure to nature (Kaplan, 1995). In the context of mindfulness practice, paying attention involves being present without emotional reactivity or conceptual engagement. Further, the process of observing the stream of thoughts, feelings and sensations can become similarly intrinsically fascinating as observing natural environments (Kaplan, 1995). Mindfulness practice can cultivate present moment awareness that in turn can facilitate immersion to nature. Therefore, mindfulness meditation appears to synergize with the acquisition of nature-derived health benefits (Djernis et al., 2019). Altogether, common mechanisms and interactions between mindfulness meditation and nature exposure seem to exist.

As opposed to other nature-based treatment options which might be beneficial due to exposure to phytochemicals, olfactory stimulation, as well as activity from walking in nature (e.g. Forest Bathing) (Djernis et al., 2019), NBSR was designed to simulate the inspirational quality of an audiovisual stimuli of natural environments in a controlled environment without actually exposing people to a natural environment.

To our knowledge, this is the first nature-based active control condition incorporating elements of group treatment including instructor attention, group support, and education, but without the specific mindfulness-related components of MBSR (meditation, non-judgement/acceptance, self-awareness). The purpose of the present study was to validate the nature videos component of NBSR, which we proposed to be viable attention-matched activities that parallel the formal meditation practice components of MBSR in their variability of hedonic tone and arousal states and heterogeneity of subjective experiences.

Intervention

NBSR was devised by our study team as an active control for comparison with MBSR. The goal was to parallel the didactic structure of MBSR without the inclusion of any mindfulness components. Class activities, content and homework assignments were designed to match MBSR activities (Table 1) and represent valid and active components. However, they do not include the same active components of MBSR (e.g., meditation, self-acceptance, self-awareness). For example, MBSR consists of eight weekly sessions (2-2.5 hours) and a 7-hour silent retreat between weeks 6 and 7 (Carmody & Baer, 2009). MBSR focuses on cultivating mindfulness through formal practices (e.g., sitting meditation, body scan, mindful yoga and walking meditation) and informal practices (e.g., mindful eating, speaking and listening, and mindfulness of daily activities). MBSR also includes group interactions called inquiry where participants share and investigate pleasant, neutral, and unpleasant experiences from daily life, mindfulness practice, and stressful situations. The total daily home practice in MBSR includes 40-45 minutes of guided formal mindfulness practices (using audio recordings) and other informal practices that do not require a recording (Santorelli, S.F., Kabat-Zinn, Jon, Blacker, M., Meleo-Meyer, F., Koerbel, 2017).

Table 1.

Intervention content comparison.

MBSR		NBSR
In-Class	Home Practice	In-Class	Home Practice
Body scan sitting meditation	Body scan/sitting meditation with provided recording	30-minute nature experience videos	30-minute nature-related podcast
Mindful eating (raisin-eating exercise)	Informal mindfulness practices (tuning into breath, body, sounds, surrounding environment), mindfulness of daily activities	Discussion about the colors, sounds and smells of nature	Nature-related practices: Suggestions to reflect on the individual’s relationship to nature, how they show their appreciation towards nature, how they talk about nature
Yoga	Lying-down and standing yoga exercises	Discussion about the benefits of movement and nature exposure	Nature-related practices: Suggestion to track how much time participants spent outside daily
Silent Retreat day (4 hours): work with all MBSR practices, group discussion & exercises	--	Retreat (4 hours): nature discussion and videos, group process, & exercises	--

In contrast to MBSR, which focuses on changing participants relationship to present moment experience, NBSR focuses on the natural environment and changing participants relationship to it. The manual consists of eight sessions, each with a distinct theme (e.g., 1. What is Nature?; 2. The Natural World and the Mind; 3. Confronting Fears of Nature; 4. How Nature Can Improve Health (Part 1); 5. How Nature Can Improve Health (Part 2); 6. Your Relationship with the Natural World; 7. Caring about our Natural World; and 8. Setting up your Return to Nature Plan). Weekly lessons were designed to engage thoughtful reflection and to create dialogue between the group members and the instructor. The total daily home practice includes 40-45 minutes of guided formal nature-related practices (using audio recordings) and other informal nature-related practices that do not require a recording. The manual evokes similar emotional, affective, and cognitive states as MBSR does during inquiry by encouraging the group to discuss the evolution of their relationship to nature, from childhood to adulthood, and asks them to reflect upon how it has impacted their lives. Within the manual there is a blend of information about nature, as well as more experiential, personal sharing and accounts where the instructor is encouraged to maintain an open and safe space to communicate and share with the group similar to the MBSR inquiry process.

The two groups are structurally identical in the amount of class time and assigned homework time each week. There are 8 weekly classes of 2 hours each, with one weekend silent retreat between weeks 6 and 7 (at least 4 hours required, with 4 hours recommended for live online retreat, and up to 7 hours for in-person context). Both groups have equivalent amounts of didactic content, active practice, group discussion, and social support each week provided by the instructor and the group.

In order to match the formal at-home meditation practice of MBSR, we developed a curriculum of various existing podcast episodes with nature-related content that are readily available in online podcast libraries (e.g., BBC Nature and Environment podcasts). NBSR participants are assigned at-home practices that include listening to the 30-minute nature-related podcasts that they are asked to attend to with their full and undivided attention once a day. To ensure NBSR accounts for Hawthorne effects from monitoring home practice on the likelihood of engaging in the measured behaviors, a weekly survey asks about at-home nature-related practices related to reflecting on the individual’s relationship to nature, i.e., how many times the individual goes out in nature, appreciates nature, talks about nature, etc.(McCambridge et al., 2014). Instructors check in periodically with group members to help them better understand or discuss class materials, as well as offer their expertise and support.

Didactic presentation includes information provided by the instructor on the empirical evidence of each nature-related topic or theme, the rationale for the topic, and how to apply these lessons to daily life. Importantly, rather than “lecturing” to participants, the attention and skill of the instructor is directed toward listening to the feedback provided by participants in order to use participant-generated experiences as a starting point for integrating the didactic material into the structure of each class similar to MBSR inquiry.

Each class begins with an exploration of the participants’ experience of the assigned and recommended at-home practices. The instructor is encouraged to provide a safe and open space for participants to share their experiences freely, without judgment. During week 1, classroom rules are formulated and agreed upon by all members of the group in order to maintain a safe, respectful, and honest environment where each participant’s confidentiality is upheld.

MBSR teachers’ embodiment of mindfulness and commitment to daily practice can be quite impactful on participants. Therefore, the instructor leading the NBSR group is required to embody a passion for nature and be personally committed to the skills and practices that group members are encouraged to implement. The NBSR group leaders’ personal interest in nature and commitment to nature-related activities for stress reduction is encouraged to be modeled. At no time, however, can the NBSR instructor mention mindfulness or share about experiences with meditation, and instructors with substantial mindfulness meditation experience should not be allowed to lead NBSR groups in rigorous clinical trials to prevent drift and unintentional confounding (Waller & Turner, 2016).

To control for the sustained attention component of mindfulness in MBSR (i.e., formal meditation practice), NBSR participants watch a 30-minute nature video each week. Each video contains novel content and adheres to a particular theme based on the earth’s biomes or habitats. These themes include savannah, grasslands, forest, ocean, jungle, coastline, and mountains. During the video, participants are asked to remain in a comfortable and upright posture and are encouraged to be as still as possible while remaining focused on the video. The body position and behavioral expectations incorporated in NBSR mirror the position and conduct typically adopted during mindfulness meditation. Therefore, the allocation of NBSRs video component directly parallels the time spent in MM during MBSR groups. Each nature video contains a balance of pleasant (e.g., jackal puppies playing, jungle sunset), neutral (e.g., bison grazing, river flowing) and unpleasant clips (e.g., rattlesnake nest, glaciers collapsing), in order to mimic the variability and heterogeneity of the experience of MM, which is not always relaxing or pleasant. Importantly, the purpose of MM is not relaxation or to create pleasant experience, but rather to be with and accept present moment experience as it occurs (Kabat-Zinn, 1990),(Schuman-Olivier et al., 2020). Formal meditation is known to produce a range of experiences. Often the experience can be pleasant or neutral, and occasionally difficult thoughts or memories become present. Thus, we sought to create an activity matched to the range of emotional valence (pleasantness/unpleasantness) and arousal (high/low) that can be experienced during meditation, even if the temporal distribution of valence and arousal for any given meditation practice may not be exactly congruent. During prior subjective pilot testing with ecological momentary assessment of mindfulness meditators doing seated mindfulness of breathing, we found roughly a nearly equal representation of momentary experiences deemed pleasant (40%), neutral (30%), and unpleasant (30%). Therefore, a control for MM should not be all pleasant, relaxing nature videos, but rather nature videos with a similar valence ratio. To obtain the nature videos, the research team used YouTube to search for nature footage. Footage with clearly visible human interference (e.g., camera operators, buildings), commentary and music were either excluded or these elements were removed. Appropriate videos, in which permission was granted by the videographer, were downloaded using a video conversion tool and edited using iMovie into clips that were approximately 3-minutes in length. These clips were then compiled into eight 30-minute videos consisting of 10 sections each. In addition, two 15-minute videos consisting of 5 clips each were created for backup clips in case clips needed to be replaced in the standard 8 videos after validation testing, resulting in a total of nine different full-length videos consisting of 90 different 30-minute clips.

Methods

To remove unsuitable clips prior to the validation study, the footage was rated internally by the study team. The valence of each clip was assessed via a google form, with each receiving a minimum of 3 ratings from study staff. Clips with a standard deviation greater than 3 were removed, and those that did not meet the required valence ratio were replaced. Each 30-minute video contained four “pleasant”, three “neutral”, and three “unpleasant” clips, while each 15-minute video contained two “pleasant”, one “neutral”, and two “unpleasant” clips. Each video contained footage from a similar biome (e.g., grassland, desert, mountains, coastline, etc.) (Figure 1), designed to match the varying session-based focus on content found in MBSR courses (e.g., loving kindness, body scan, breath awareness) and to provide a sense of legitimacy and thematic consistency that may not have been present had the videos contained a random assortment of footage. Each clip was then re-uploaded to a private YouTube account and designated “unlisted” to avoid the clip being viewed by the public.

Fig 1.

Still snapshot images of 3-minute video clips a) Pleasant clip from “Ocean” video, b) Neutral clip from “Grasslands” video, c) Unpleasant clip from “Jungle” video, d) Unpleasant clip from “Tundra” video

To validate our initial ratings of valence and arousal, we used Amazon Mechanical Turk (MTurk) to obtain ratings of nature video clips. MTurk is a crowdsourcing marketplace that allows businesses and researchers to pay participants to perform short tasks online through the MTurk server (Hunt, 2015). The MTurk population consists of the large and diverse community of registered “Workers” from around the world. However, to eliminate intercultural differences, participants outside of the U.S. were not included in participant population (Kurdi et al., 2017). We recruited adult MTurk workers through the MTurk platform, and included only workers with an approval rate of 90% or above on all previous human intelligence tasks (HITs), and from all backgrounds (race, gender, marital status, etc.), except those under the age of 18 (Kurdi et al., 2017).

After signing an informed consent that was reviewed by Cambridge Health Alliance’s Institutional Review Board, each participant also provided their MTurk Worker ID, age, and gender. Participants were asked to watch 10 short, 3-minute nature videos and after each 3-minute video they were asked to rate their level of affective arousal and the valence of each video (Kurdi et al., 2017). More specifically, the participants were asked to rate how each video made them feel on a 7-point Likert scale (1 = very unpleasant, 4 = neutral, 7 = very pleasant), and to rate the highest arousal level they felt over the course of the video on a scale from 1 being the most unaroused (slow, still, de-energized) to 7 being the most aroused (excited, alert, activated). After completion of the survey, a unique code was presented to the participants who were then prompted to enter the code onto MTurk to receive compensation of $4.50. All information obtained from participants was recorded using Google Forms and a secure database at Cambridge Health Alliance.

A total of nine surveys were created using Google Forms, containing clips from 10 different biome-specific nature videos, video content questions for quality assurance purposes, and respective rating measures for valence. Each of the nine surveys contained 10 different 3-minute video clips. A video content question was required to be answered correctly after each video to ensure no automated bots completed the surveys and that MTurk workers watched the videos in their entirety (Kurdi et al., 2017). The video content question was unique to the content of each video and resembled the following: “In the video, there was an image of: a) a bear, b) a fish, c) sky, d) just a and b, e) all of the above)”. If respondents answered the video content question incorrectly, they were redirected to the previous page prompting them to re-watch the video and answer this question again. For the purposes of this study, valence measures of pleasantness were the primary aim of the analysis to determine the efficacy of the nature videos as an attention placebo control for MM practices where sustained attention is required.

To detect and exclude outliers on each survey, we calculated the standard deviation of all ratings given by each individual separately. To remove outliers from those who answered nearly all the same responses for every rating or only responded with highly extreme variables, when participants responses across the entire survey had a standard deviation less than 0.65 or greater than 2.85 they were excluded (this excluded those who generally answered the same numbers for both valence and arousal Likert scales across the board, i.e., those who responded with almost all answers as 1s or 7s) (Kurdi et al., 2017).

Results

Adults (N=176) completed one of nine possible surveys via Amazon MTurk (mean age = 35.28 ± 2.85, 110 male, 66 female) with a median of 14 respondents per survey. Exclusion of 49 (35 male, 14 female) participants resulted in a final sample of n=126 (74 male, 52 female, mean age = 36.06 ± 3.13). Participants were excluded if the standard deviation (SD) of the ten total 3-minute nature clip ratings was <0.65 and >2.85 as determined by the homogeneity of ratings for all three categories (pleasant, unpleasant, and neutral).

We used an Analysis of Variance (ANOVA) test to determine whether mean valence ratings significantly differed between the three video clip valence categories (pleasant, neutral, unpleasant). We ran a similar test between the arousal ratings for each of the video categories. To examine any potential gender differences in valence or arousal ratings, we ran separate t-tests to compare both valence ratings and arousal between males and females. We used linear models to determine whether there were associations between either continuous or categorical age (18-25, 26-45, or >45 years) and overall valence and arousal ratings.

Mean valence ratings (Figure 2, Figure 3a) significantly differed between the three categories of nature video clips using an ANOVA test (means: pleasant=5.74±0.71, neutral=5.45±0.90, and unpleasant=4.44±1.5, F(2,375)=49.78, p<0.001). Follow-up pairwise t-tests revealed significant differences between valence ratings for pleasant vs. unpleasant (p<0.001), neutral vs. unpleasant (p<0.001), and pleasant vs. neutral (p=0.005) video clips.

Fig 2.

Mean valence ratings significantly differed in a one way ANOVA test, and the means of each category (pleasant, neutral, unpleasant, combined across all nine surveys) were significantly different from each other category in follow-up t-tests (* = p<0.01,** = p<0.001). Mean reported arousal did not differ between the three categories.

Fig 3.

a) Histograms of valence ratings for each video clip, combined across each of the nine different surveys for Pleasant, Neutral, and Unpleasant clips. Since there were different numbers of total ratings for the three categories (P = 508 ratings, N = 370 ratings, and U = 392 ratings), the number of ratings for each level (1-7) is shown as the percentage of the total number of ratings for each video category. Kolmogorov-Smirnov tests between each combination of distributions revealed that each was significantly different from each other (P vs N: D = 0.10, p = 0.02, P vs U: D = 0.31, p < 0.001, N vs U: D = 0.26, p < 0.001). b) Similar histograms but representing reported arousal instead of valence ratings for each video category. Kolmogorov-Smirnov tests between each combination of distributions revealed no significant differences (P vs N: D = 0.06, p = 0.34, P vs U: D = 0.02, p = 1, N vs U: D = 0.07, p < 0.33).

Mean reported arousal (Figure 2, Figure 3b) did not differ across the three video types (means: pleasant = 4.81±1.35, neutral = 4.52±1.53, and unpleasant = 4.81±1.20, F(2,375)=1.865, p=0.156). We fit a linear mixed effects model with valence rating as the response and with predictors including video category (P,N,U), arousal, a quadratic term for arousal, and interaction terms between all of the main terms. A random effect term for participant was included in the model, to account for the design in which ratings across the different video categories are not independent within an individual, and therefore cannot be treated as fully independent observations. The interaction terms between the unpleasant video category and arousal (p < 0.001), as well as quadratic arousal (p = 0.01, were significant in the model, whereas the equivalent interaction terms for the pleasant and neutral categories were not significant (Figure 4).

Fig 4.

Scatter plot of valence ratings vs. reported arousal with quadratic fit lines for each video category. Based on the outcomes of the linear mixed-effects model described above, only unpleasant videos were statistically nonlinear (quadratic). In the mixed effects linear model there was a significant association between the unpleasant video category and both linear arousal (*=p<0.001) and quadratic arousal (**p=0.01).

Average valence ratings differed between males and females (male mean = 5.45±0.75, female mean = 4.99±0.85, p=0.002). Age was found to be significantly associated with valence ratings (p=0.046), such that increasing age was associated with decreasing valence (Figure 5a). Age was not associated with reported arousal overall or within any of the three video categories (p=0.758) (Figure 5b).

Fig 5.

a) Age vs. Valence: There was a significant association between age and valence rating for the unpleasant video category specifically (r = −0.22, p = 0.01). b) Age vs. Reported Arousal: No significant associations were found between age and reported arousal overall or within any of the three video categories separately (Solid lines are significant, dashed lines are non-significant relationships).

Discussion

In this manuscript, we present a rationale for the development of NBSR as an attention placebo control (APC) group as a comparison to MBSR for use in RCTs. Further, we aimed to assess whether NBSR in-session nature videos serves as a good attention matching comparison for MBSR in-session mindfulness meditation practices by investigating the validity of the video component of NBSR. NBSR seems to exhibit a similar inspirational quality as MBSR with the videos that were specifically created with the goal of eliciting similar feelings of calm, connectedness, and joy, as well as fear, sadness, unpleasantness, and neutrality/boredom in similar frequencies and variability as experienced during mindfulness meditation practices within MBSR groups. Indeed, the participant responses to the videos were distributed across the three predicted categories of sensations (pleasant, neutral, unpleasant) with significantly different mean scores (1-7) for each category. The result of the reported score range being higher than expected could indicate that even “unpleasant” scenes are commonly perceived/accepted as naturally occurring phenomena and thus do not elicit a strong negative reaction or are experienced by younger MTurk viewers as possibly pleasant since they are exciting and fun to watch. Similar variations in experience are present during mindfulness practice, with very unpleasant reactions being less prominent and variability in arousal responses to unpleasant experiences. Interestingly, unpleasant videos were found to have the highest variance amongst raters, suggesting that unpleasant stimuli may have a naturally higher variation in response from person to person than pleasant or neutral stimuli. This variability could be due to differences in culture, environment, personal associations, past experiences, specific traumas, sadistic/masochist traits, etc. These factors tend to be idiosyncratic in nature and less predictable than factors that evoke pleasant or neutral experiences or perceptions. This pattern of increased variance is also reflected in unpleasant phenomena that may arise during mindfulness meditation. For example, an unpleasant experience could onset by various cues such as, the individual’s body position, the setting or circumstance, the time of day, smells in the room, etc. These triggers tend to be more random in nature than stimuli that elicits pleasant or neutral experiences. The increase in variance amongst the unpleasant video clips could also be due to a higher variability amongst the videos we chose as unpleasant and not necessarily reflective of the individual ratings gathered through MTurk.

The gender differences in valence ratings could be related to previous fMRI findings indicating that men show greater levels of automatic emotion regulation than women. A greater automatic regulation of negative affect triggered by the unpleasant nature scenes could serve as a potential explanation for the significantly lower valence ratings in men (McRae, Kateri, Ochsner, Kevin N., Mauss, Iris B., Gabrieli, John J.D., & Gross, 2008).Another potential explanation for the higher valence rating in women is that they have been shown to report their emotional experiences as more intense compared to men (M. Grossman & Wood, 1993). Our findings are in line with previous research, showing higher valence in women and no differences in arousal between the two genders (Rattel et al., 2020). Previous studies on age-related findings regarding emotional functioning describe an information processing bias toward positive versus negative information among older adults. The fact that our findings were not in line with this observation could be explained by the low mean age (35.28 years) of the participants (Reed, A.E., Chan, L., Mikels, 2014). Taken together, our results indicate that viewing nature videos was successful in producing a comparable experience as mindfulness practices without the key ingredients of MM (meditation, acceptance, awareness).

The procedure of choosing or creating a control intervention should include ethical considerations about its nature. Using treatment as usual (TAU) as a control is a robust method for demonstrating the effectiveness of interventions. Nevertheless, in some cases, patients in the TAU group receive substandard treatment, or no treatment at all (Areán & Alvidrez, 2002). Furthermore, TAU controls fail to fully account for the nonspecific effects of an intervention such as MBSR (e.g., regular contact with an empathetic healthcare provider, provision of social support in a group format, creation of the expectation of benefit, etc.). NBSR provides all of those relevant controls for non-specific treatment factors.

A further consideration in the development of a control intervention for MBSR is structural equivalence. Structural variables include number of sessions and their duration, therapist qualifications, and therapy format (e.g., group or individual) (MacCoon et al., 2012). Additionally, a control intervention should include all non-specific factors of MBSR such as therapeutic alliance, the opportunity to express emotions and ventilate problems, positive expectation for change/success. NBSR was developed to structurally match MBSR sessions and to provide information on the health benefits of nature and room for open discussion. Similarly, NBSR encourages emotional expression, self-reflection and a positive attitude towards change. Collectively, NBSR is well-matched to MBSR on numerous non-specific elements but without the key MM ingredients; it can therefore serve as a rigorous active control for MBSR, permitting randomized trials to isolate the specific benefits of mindfulness as an intervention.

There are some distinct advantages and disadvantages in using MTurk platform for data collection. On one hand, MTurk allows for targeted recruitment of specific population segments and provides access to diverse samples in the United States (Buhrmester, M., Kwang, T., Gosling, 2011). In addition, MTurk makes it possible to collect hundreds or even thousands of responses within a few days and gives researchers more control over various study settings. Compensation typically includes gift cards or reward points similar to in person recruitment.

On the other hand, at any given moment MTurk has access to a limited number of participant population or sometimes participants complete many and often related studies. This creates concerns that prior exposure to research materials can decrease data quality. While MTurk is often celebrated as offering more diverse samples than college student subject pools, a second concern is that MTurk workers might not be representative of the US population or the specific clinical population for which NBSR may be used as a control for MBSR in any given study.

A limitation is that we were unable to directly compare ratings of subjective valence and affective arousal during equal duration blocks of mindfulness meditation during MBSR. This was beyond the scope of this initial attempt to characterize and validate the nature videos. This is a key area of future research and targeted data collection as we use the NBSR control intervention in comparison to MBSR in the future. Moreover, a caveat should be mentioned in the adoption of NBSR as a control for MBSR. Placebo effects can be powerful and can make it difficult to distinguish effective active interventions from placebo controls (Kaptchuk, 1998). We have struggled to make our NBSR intervention authentic and compelling, but placebo effects can also be too strong for certain clinical trial designs. For example, some placebo interventions, especially with placebo attention controls, reach a “ceiling” effect for clinical outcomes for both active and control interventions. Thus, many clinical studies have shown that it is sometimes possible for placebo responses to be so large as to obscure any difference from an active intervention known to be effective in other RCTs. For example, a study with asthma patients (n=468 of asthma acerbations) showed that for subjective clinical outcomes, two different types of placebo interventions (placebo aerosol and placebo acupuncture) had similar effects to active albuterol (Wechsler et al., 2011). A similar finding was reported in another asthma study (n=601) (Wise et al., 2010). Furthermore, experimental neuroimaging studies have shown that placebo effects can have similar behavioral outcomes but operate through different regional brain mechanisms. Thus, for clinical outcomes, there is a possibility that our placebo APC may too compelling and lead to an inability to differentiate MBSR from NBSR. However, for studies that will evaluate whether the two interventions are mechanistically different, NBSR may indeed be an excellent control for the specific mindfulness aspects of MBSR.

Conclusion

These findings demonstrate that one of the primary components of NBSR, nature videos, provide a promising attention- and valence-matched placebo activity unrelated to mindfulness meditation. Specifically, they highlight the emotional experience of varying scenes of nature as generally pleasant on average, with higher variability in what is subjectively rated as unpleasant, which generally parallels the variability and heterogeneity of subjective experiences during mindfulness meditation. Overall, our results suggest that the video component of the NBSR was successful in eliciting feelings of pleasantness, unpleasantness and neutrality similar to the 4:3:3 ratios reported during mindfulness meditation from our original unpublished ecological momentary assessment testing. These results demonstrate that NBSR might be a promising comparison group to MBSR as well as similar mindfulness-based interventions that researchers can use in the future. To our knowledge, NBSR is the first active control intervention that was designed to be structurally equivalent to MBSR, including attention control and matching its impacts on subjective hedonic experience, while not including any mindfulness components. Further studies are warranted to investigate the generalizability of these findings across various populations including clinical groups. Data from RCTs will be needed to confirm the expected effectiveness of NBSR as an attention control group for MBSR.