Cognitive and Physiological Impacts of Adventure Activities: Beyond Self-Report Data

Abstract

Outdoor adventure activities have been used to facilitate a variety of positive outcomes. However, the practical challenge of collecting data in the field and a heavy reliance on self-report data render it difficult to understand the process of the experience. This study examined the association between self-reported valence and arousal and electroencephalography (EEG)-measured anxiety, focus, and approach motivation to determine the physiological and cognitive response to stressful stimuli and compare those objective measures with self-report assessments. Data were collected from 10 participants fitted with an EEG headset during rappelling. Spearman correlations and repeated-measures ANOVA were used to analyze the data. Results indicated significant changes in EEG readings for anxiety and approach motivation, and significant correlations between self-reported valence and EEG-measured approach motivation. The findings illustrated the acute internal response to a common adventure activity and demonstrated the influence of a novel challenge on the mind and body of participants. Significant changes in self-report assessments were reflected in similar changes in objective measures, indicative of the mind/body connection.

Keywords

valence electroencephalography approach motivation arousal anxiety

Introduction

Adventure education has revealed many insights about the physical, psychological, and social antecedents and outcomes of outdoor activities. Outdoor environments and adventure activities (e.g., rock-climbing) have been used to facilitate a variety of positive outcomes, including improved self-concept, self-efficacy, academic motivation, physical and mental health, and improved behavior (Gillis & Speelman, 2008; Neill, 2003). However, the practical challenge of collecting data in the field and a heavy reliance on post hoc self-report data render it difficult to understand how those activities are experienced (Ewert & Sibthorp, 2009; van de Mortel, 2008). Researchers have explored other methods of measurement in outdoor research, including experience sampling (Csikszentmihalyi & Hunter, 2003), biomarkers such as heart rate and cortisol levels (Ewert, 2015), and participant observations and qualitative methods (Fulmer & Frijters, 2009; Remington & Legge, 2017), among others. All methods have advantages and limitations, and it could be argued that multiple methods are always preferable. Our study builds upon previous research by investigating the use of brainwave data along with established biomarkers. This literature review provides a concise description of the outdoor adventure “stimulus,” as well as physiological and cognitive responses associated with such activities, laying an empirical foundation for our study.

Literature Review

Outdoor Adventure Activities

Outcomes associated with adventure activities are well-documented (Gillis & Speelman, 2008; Neill, 2003). Moving beyond evidence of impact, researchers are now calling for a deeper understanding of how the process works (Henderson, 2004; Sibthorp, 2003). The process of adventure learning has been explored with a variety of methods, connecting outcomes to program components (Bailey & Fernando, 2011; McKenzie, 2003; Paisley, Thurman, Sibthorp, & Gookin, 2008). Elements commonly associated with positive outcomes include the physical and social environment, appropriate physical and mental challenges, and cognitive processing.

Central to any adventure program is the assumption that growth occurs through experience. For this reason, experiential learning theory (Kolb, 1984) is a recurrent theme in the outdoor literature. Theoretically, this experience should be novel and challenging, facilitating cognitive dissonance and inducing growth toward mastery of the experience (Beames & Brown, 2016; Walsh & Golins, 1976). The novel environment and adventure activity provide challenges that must be negotiated to re-establish mental equilibrium. A side effect (or perhaps a catalyst) of these challenges is stress. An optimal level of stress, associated with perceived risk, may enhance the development of psychosocial resilience, mental toughness, and character (Dienstbier, 1989; Ewert & Yoshino, 2011; Neill, 2003).

Recently, the utility of stress for inducing growth has come under scrutiny (Brown, 2008; Mackenzie, Son, & Hollenhorst, 2014). However, research indicates that one’s response to stress may be a better predictor of health than the mere existence of stress (McGonigal, 2015). Those who accept calculated risks, persist at difficult challenges, and view failure as a minor setback are also healthier, happier, and more successful by many measures (Dweck, 2006). Thriving through the recurring stressors of life, often referred to as “hardiness” (Mutz & Müller, 2016), is thought to result from successfully negotiating incremental challenges. Positive stress, or “eustress” (McGonigal, 2015), from challenging activities may enhance healthy outcomes (e.g., physical training), while chronic, unresolved “distress” impedes growth. Acknowledging the debate over risk in adventure education, the purpose of this study is to explore the physiological and cognitive effects of perceived risky activities. The following sections will describe known physiological and neurological responses to challenge, stress, and risk.

Physiological Indicators

Stress is the body’s natural reaction to challenges (McGonigal, 2015). When confronted with a mental or physical challenge, the body reacts with a series of well-described changes, such as increased breathing rate, heart rate, skin temperature, and sweat (Mai & Paxinos, 2011). These physiological markers are evidence that the body is preparing to confront the challenge or to escape. Such indicators can be tracked by individuals and researchers through various research instruments and fitness trackers (Ahuja, Ozdalga, & Aaronson, 2017).

Acute stress, as encountered through adventure activities, eventually subsides and the body returns to homeostasis with a parasympathetic response (Mai & Paxinos, 2011). Physical stress increases heart rate and sweating, but certain mental challenges may induce an even stronger physiological response. Rappelling is physically less demanding than rock-climbing, for example, but may expose participants to a much bigger cognitive challenge. Previous research revealed that novice participants commonly expend more energy rappelling than rock-climbing (Bailey, 2014). The mechanism behind this phenomenon originates in the brain, where the sympathetic response is triggered. This pilot study aimed to illustrate the physiological changes (skin temperature, sweat, and heart rate) indicative of a stress response during the rappel.

Cognitive Indicators

Well-established physiological reactions accompany a mental shift that occurs when confronted with a challenge. Challenges, especially novel ones, require a shift away from intuitive, “System 1” thinking directed by the limbic system (Kahneman, 2011). Novel environments and activities require one to evaluate the situation and determine if pre-established modes of thinking (e.g., heuristics) can be applied to the current encounter. This deeper investigation, “System 2 thinking,” renders the person more alert, evidenced by the physiological indicators previously described (Kahneman, 2011). A burgeoning amount of neuroscience research provides a glimpse into the inner workings of the mind during these various phases.

Recent functional magnetic resonance imaging (fMRI) and electroencephalography (EEG) have helped connect thought processes to behavior and physiological activity. Findings relevant to adventure education indicate that physical activity increases mental relaxation, positive affect, and the ability to focus (Schneider et al., 2009; Woo, Kim, Kim, Petruzzello, & Hatfield, 2009). Outdoor environments can decrease frustration and increase meditative states whether they are experienced personally (Aspinali, Mavros, Coyne, & Roe, 2015) or vicariously (Roe, Aspinall, Mavros, & Coyne, 2013). Being in outdoor environments also increases the comprehensive mapping of acute cognitive and emotional reactions to stressful stimulus (Johnstone, Gunkelman, & Lunt, 2005). While fMRI provides a more complete picture of brain activity, the size and expense associated with these machines make them impractical for research outside of a laboratory. New developments in EEG technology, however, have made it feasible to conduct research in remote environments. EEG research relies on highly sensitive devices to collect electrical signals from the surface of the scalp. The power, frequency, and location of these signals provide valuable information about cognitive processes occurring in the brain. The following sections will provide a brief review of neuroscientific findings relevant to EEG. It should be noted that the need for brevity precludes a comprehensive discussion of neuroscience. Here, we present a general overview necessary to interpret our study.

Brain Mapping

Although neuroscientists are only beginning to unlock the brain’s mysteries, there is general consensus regarding the functions associated with brain lobes (Purves, 2012). The posterior areas, for instance, are associated with vision and image perception (occipital lobe), processing sensations from the muscles and skin (parietal lobe), and evaluation of weight, texture, and sensations (somatosensory area). Temporal lobes process short-term memories and equilibrium. The frontal lobe processes executive functions such as judgment, planning, concentration, and emotional expression. These higher order processes are associated with learning, decision making, and leadership. Finally, the amygdala, buried deep beneath the parietal and temporal lobes, processes pain, hunger, and fear.

Brainwaves

To determine the type and intensity of activity in the various areas of the brain, EEG researchers translate electrical signals into discrete frequency bands through spectral analysis. Fast Fourier Transformation (FFT) in EEG research is used to divide the signals from various sensors into measures of power in five distinct frequency ranges: delta (δ = 0.5-3 Hz), theta (φ = 4-7 Hz), alpha (α = 8-15 Hz), beta (β = 16-31 Hz), and gamma (γ = 32-100 Hz). Delta is most often associated with sleep or subdued levels of brain activity, but can also be pronounced in people with depression or attention-deficit/hyperactivity disorder (ADHD; Clarke et al., 2008). Theta waves indicate mind-wandering or meditation. This frequency has been associated with creativity and ideation, as it allows for an open flow of ideas (Purves, 2012). Alpha waves represent the mind’s natural state of readiness or nonarousal, indicating a relaxed state of mind. The beta frequency denotes arousal and mental engagement. The final category, gamma, has been associated with high-level mental tasks, and conceptually linked to flow theory (Lutz, Greischar, Rawlings, Ricard, & Davidson, 2004).

New discoveries linking various brainwaves to areas of the brain are documented daily, and the understanding of these connections will likely undergo much change in the coming years. For the current study, three main cognitive states were assessed: focus (concentration), anxiety, and approach motivation (AM). Focus is a general measure of mental engagement evidenced by beta activity in the frontal lobe (Coelli et al., 2015). Anxiety is associated with gamma waves in the parietal and occipital lobes (Oathes et al., 2008). Finally, AM was measured to compare brainwave data with an established self-report Arousal/Valence scale. A participant with AM presents more high-frequency waves (or an absence of alpha activity) on the left frontal lobe than the right (i.e., frontal asymmetry). AM is related to high levels of arousal, which not only may indicate positive emotion (e.g., happiness, excitement) but could also indicate aggression or anger (Coelli et al., 2015). Previous research has validated the use of frontal asymmetry to measure AM, associating the EEG markers with self-report items, such as “I go out of my way to get things I want” and “I crave excitement and new sensations” (Amodio, Master, Yee, & Taylor, 2008). Accordingly, lack of AM could indicate negative emotion, or a general desire to withdraw from the situation. This would be associated with scale items such as “I worry about making mistakes” and “I have many fears as compared to my friends” (Oathes et al., 2008). This pilot study aimed to demonstrate that cognitive indicators (focus, AM, and anxiety) would show signs of a stress response during the rappel.

Arousal and Valence

The self-report measure collected in this study was a pre, during, and post activity assessment of participants’ level of arousal and valence based on a multi-axis measure of emotion (Russell & Barrett, 1999). The two-dimensional model provides feedback on the level of arousal (e.g., bored or tense) and valence (e.g., happy or upset) experienced at a given point in time, via selection of descriptive words (Bradley & Lang, 1994). Arousal and valence are empirically tied to AM measured as frontal asymmetry (Coelli et al., 2015). Focus is conceptually related to arousal, and anxiety could be associated with negative emotion (i.e., low AM). This study examined the correlation between self-reported valence and arousal and AM, anxiety, and focus as measured through EEG.

Based on previous research as described in this literature review, three hypotheses were developed:

Hypothesis 1 (H1): Physiological indicators (skin temperature, galvanic skin response [GSR], and heart rate variability) will show signs of a stress response during the rappel.

Hypothesis 2 (H2): Cognitive indicators (focus, AM, and anxiety) will show signs of a stress response during the rappel.

Hypothesis 3 (H3): Self-report valence and arousal will correlate with AM and anxiety as measured through EEG.

Method

A total of 12 students enrolled in an introductory outdoor recreation class were fitted with an Emotiv Insight EEG headset and a Microsoft Band 2 fitness tracker during a required class activity. Five of these students were female, and the average age was 21. Of these students, one declined to participate and technical difficulties resulted in the loss of data for another, resulting in a total number of 10 participants. All students were novice rappellers, with none reporting previous rappelling experience outside. A rappel was set up at a popular overlook affording aesthetic views and an intimidating launch. Prior to rappelling, the participants were fitted with the research devices at an area distant from the edge of the rappel. Once outfitted, participants reported their level of arousal and valence by choosing two words from the Arousal/Valence scale to indicate their current affective state (Time 1). Participants were then guided through the rappelling process. After reaching the bottom, participants were again asked to select two words from the Arousal/Valence scale to indicate the way they felt during the rappel (Time 2) and immediately after (Time 3).

Measures

Arousal and valence scores were created by assigning a value to each word on the scale based on its location on the continuum (Bradley & Lang, 1994). Arousal was measured on a scale of −4 (bored, calm) to 4 (tense, alert), based on the average score of the two words selected at each time point. Valence was measured using the same method, with scores ranging from −4 (upset, sad) to 4 (happy, contented). In this way, each participant was given a separate score for valence and arousal at each time point.

Physiological indicators were recorded from the Microsoft Band 2 using an iPhone and third-party application. As with many fitness trackers, the Microsoft Band 2 has been tested in previous research with varying results (Li et al., 2016). This specific device was chosen for its inclusion of multiple sensors (e.g., GSR) that provide information about subtle physiological changes associated with a stress response. The third-party application collected data on GSR, skin temperature, and heart rate interval (HRI) at regular intervals. HRI was used in lieu of heart rate, given its stronger association with a stress response (Choi & Gutierrez-Osuna, 2009). All data were collected on the iPhone using the device’s internal clock to create a timestamp for each collection point. The fitness tracker also collected GPS and elevation data every few seconds, allowing for an assessment of the participants’ position during the rappel for comparison of physiological and EEG data. The 5-s epochs of data were averaged for each of four time points for analysis.

EEG data were collected using the Emotiv Insight headset, streamed at a rate of 2 times per second to the same iPhone. Emotiv headsets have been verified through a variety of previous studies, including the measurement of outdoor environments on mental states (Badcock et al., 2015). A custom application was developed to stream all brainwave data to the phone, then to a server. Spectral analysis (FFT) provided power measures for each frequency (e.g., alpha, beta, etc.). EEG data were given a timestamp using the same internal clock on the device, to accommodate synchronization with other data for analysis.

The EEG data were cleaned for analysis to avoid inclusion of artifacts (i.e., irrelevant signals) from major muscle movements during the activity. Raw values were then transposed into cognitive states using validated formulas. Focus was measured as the presence of beta waves and the absence of alpha waves across left and right frontal lobe sensors (Coelli et al., 2015). AM was measured as frontal asymmetry (Harmon-Jones, Gable, & Peterson, 2010), and anxiety was measured using gamma wave activity in the parietal lobe (Oathes et al., 2008).

Physiological, GPS, and elevation data were then aligned with EEG data by timestamp. Using time and elevation data, 5-s epochs of time were chosen to represent snapshots of mental and physical states during the activity. A 5-s epoch was chosen to accommodate the high level of variability in EEG data (Ayabe, Kumahara, Morimura, & Tanaka, 2013). Epochs were selected from four distinct points for each participant—T1: initial baseline assessment before rappelling, T2: as they leaned over the edge, T3: after negotiating the edge and while still on rappel, and T4: immediately after they reached the bottom (Figure 1). EEG data for each epoch were averaged into a single score to be compared with physiological data at the same time point. Four time points were chosen for EEG data to distinguish between the anecdotal crux of the experience (i.e., the edge) versus the actual descent (i.e., middle). All data were then analyzed with spearman correlations due to small sample size (Puth, Neuhäuser, & Ruxton, 2014) and repeated-measures ANOVA in SPSS. Power analyses determined that a sample size of 10 participants was appropriate for detecting a medium effect size with within-subjects analysis of four time points (Faul, Erdfelder, Lang, & Buchner, 2007).

Figure 1.

Example of time point selection for a single participant.

Results

The purpose of this study was to determine the physiological and cognitive response to stressful stimuli (i.e., rappelling) and compare those objective measures with self-report assessments. The findings will be reported in order of hypotheses. Physiological results will be followed by cognitive findings, and then a comparison of objective measures with self-reported valence and arousal.

Physiological Indicators

Although the data trended consistently with Hypothesis 1, they failed to reach a level of significance, rendering that hypothesis unsupported. As shown in Table 1, GSR and skin temperature increased during the activity, though not enough to merit statistical significance. HRI also shortened during rappelling, indicative of a standard stress response, though not statistically significant.

Table 1.

Descriptive Statistics for Cognitive and Physiological Variables.

	Top		Edge		Middle		Bottom
	M	SD	M	SD	M	SD	M	SD
Focus	2.596	1.495	2.680	1.851	2.467	1.374	2.233	1.008
Approach	0.026	0.176	−0.087	0.220	0.109	0.303	−0.004	0.211
Anxiety	1.011	0.716	1.283	0.684	1.047	0.448	1.092	0.687
GSR	8,675.125	7,159.969	19,798.875	39,212.916	28,898.125	70,764.880	4,031.750	5,110.125
HRI	470.625	518.011	383.625	432.032	392.000	465.932	304.875	374.238
Skin temperature	84.850	2.066	84.850	2.326	84.562	2.273	84.538	2.203

Note. GSR = galvanic skin response; HRI = heart rate interval.

Table 2.

Repeated Measures ANOVA Comparisons of Cognitive States at Four Time Points During the Rappel.

(I) Time	(J) Time	Approach motivation			Anxiety			Focus
(I) Time	(J) Time	Mean difference (I − J)	F	η2	Mean difference (I − J)	F	η2	Mean difference (I − J)	F	η2
Top	Edge	0.113	5.241^†	0.428	−0.272	9.371*	0.572	−0.084	0.094	0.013
	Middle	−0.083	0.911	0.115	−0.036	0.017	0.002	0.129	0.213	0.03
	Bottom	0.030	0.131	0.018	−0.081	0.054	0.008	0.363	0.692	0.09
Edge	Middle	−0.197	7.969*	0.532	0.236	0.960	0.121	0.213	0.767	0.099
Edge	Bottom	−0.083	1.246	0.151	0.191	0.331	0.045	0.447	0.941	0.118
Middle	Bottom	0.114	0.969	0.122	−0.045	0.019	0.003	0.234	0.86	0.109

Note. Top = initial baseline assessment before rallelling; Edge = participants leaned over the edge; Middle = after negotiating the edge and while still on rappel; Bottom = immediately after they reach the bottom.

†

p = .056. *p < .05.

Cognitive Indicators

All three cognitive measures demonstrated changes during the rappel. Two of the three variables achieved significance, providing evidence to support Hypothesis 2. Participants demonstrated decreased AM (F = 5.241; p = .056) when reaching the edge of the rappel. This was followed by a swing back toward increased AM during the middle of the rappel (F = 7.969; p = .026). Finally, participants reverted to a level similar to baseline when reaching the bottom of the rappel.

Anxiety peaked at the edge of the rappel (F = 9.371; p = .018) then returned to baseline in the middle and bottom of the rappel. Focus levels were highest at baseline (M = 2.59) and at the edge of the rappel (M = 2.68). While focus dropped off during the middle (M = 2.46) and bottom of the rappel (M = 2.23), these measurements were not statistically significant.

Finally, self-report measures of valence and AM were related, in general, but the relationships were not consistent for each time point. As seen in Table 3, arousal and valence at the top of the rappel were not related to focus, AM, or anxiety at the top of the rappel. Self-report ratings at the top were, however, related to AM at the edge of the rappel. Finally, self-report valence at the bottom of the rappel was related to AM at the edge and middle of the rappel.

Table 3.

Correlations of EEG Readings and Self-Report Scores for Valence and Arousal.

EEG readings	ArousalTop	ValenceTop	ArousalEdge	ValenceEdge	ArousalMiddle	ValenceMiddle	ArousalBottom	ValenceBottom	OverallArousal	OverallValence
Focus Top	−.187	−.255	.193	−.073	.193	−.073	−.549	.110	.228	−.071
Approach Top	−.234	.447	.602	.220	.602	.220	.664	.700	.563	.786*
Anxiety Top	−.187	−.051	.494	−.244	.494	−.244	.166	−.012	.503	−.024
Focus Edge	−.187	−.051	.542	.317	.542	.317	−.281	.282	.563	.381
Approach Edge	−.733*	.753*	.615	.122	.615	.122	.575	.884**	.563	.881**
Anxiety Edge	−.078	−.192	.361	−.415	.361	−.415	.026	−.012	.395	−.214
Focus Middle	−.078	.013	.639	.366	.639	.366	−.370	.012	.683	.286
Approach Middle	−.592	.587	.169	−.195	.169	−.195	.664	.945**	.120	.595
Anxiety Middle	.655	−.128	.120	−.049	.120	−.049	−.370	−.393	.168	−.238
Focus Bottom	.156	.383	.651	.561	.651	.561	−.166	.037	.659	.571
Approach Bottom	−.094	.792*	.289	.024	.289	.024	.396	.233	.299	.381
Anxiety Bottom	−.047	.192	−.060	−.122	−.060	−.122	−.140	−.344	−.012	−.286

Note. EEG = electroencephalography.

p < .05. **p < .01.

To determine if self-report measures were more consistent with overall evaluation of the experience, new variables were created based on the sum of all time points for each concept (e.g., total focus, total anxiety, total valence). Overall self-report valence demonstrated a significant correlation with AM. Neither focus nor anxiety showed a significant relationship with overall self-report measures. These findings reveal the complexity of using self-report measures to determine the nature of one’s experience.

Also noteworthy is the lack of correlation between the three cognitive measures. While an overlap of cognitive signals would not be unfounded, the absence of collinearity provides evidence that the measurements are discrete. In addition to the conceptual and empirical support these findings provide for the hypotheses, they also provide further validation of the cognitive states thus measured. The following section will include a discussion of these findings and comparisons with previous research.

Discussion

These findings provide an initial peek into the mind of participants during adventure activities, laying the foundation for future research. While these results should be considered preliminary, they do align with theory, experience, and the stated hypotheses. This final section will connect the results to adventure education theory and propose opportunities for future research.

Two central aspects of adventure activities are the challenges and novelty of the experience (McKenzie, 2003; Paisley et al., 2008). Conceptually, these elements engender cognitive dissonance, thereby requiring a more active thinking style (Walsh & Golins, 1976). The findings from our study illustrate the acute internal response to a common adventure activity, demonstrating the influence of a novel challenge on the mind and body of participants. Significant changes in AM and anxiety were reflected in similar changes in GSR and HRI, indicative of the mind/body connection. Thus, it is evident that the activity has the intended effect of elevating awareness and facilitating a shift in mental function.

Decreased AM accompanied by a rise in anxiety indicates a negative emotional experience at the edge of the cliff. Inducing a negative experience might appear ethically dubious, fueling the debate over the utility of risk in learning (Brown, 2008; Mackenzie, Son, & Hollenhorst, 2014). It is noteworthy, however, that anxiety immediately dropped and AM increased after negotiating the edge and while still on rappel. The negative experience was immediately followed by the affirmation of success (or simply survival). Akin to theories of self-efficacy, resilience, and coping, the successful negotiation of stressful encounters may induce positive emotions and enhance one’s self-evaluation (Schumann & Sibthorp, 2014).

Self-report measures of valence and arousal were generally in agreement with objective EEG and physiological data. Given the technological burden of assessing mental status through brainwaves, these results are welcome. This highlights the importance of asking participants how they feel during the experience. Challenge by Choice (Chase, 2014) is a well-established principle in the outdoor industry, but the process is often reduced to a nod of agreement to participate. Having participants select descriptors to indicate their current status would provide deeper insight into their experience, allowing leaders to better accommodate their growth.

Not all participants may provide honest feedback, as psychosocial influences could corrupt self-report measures (van de Mortel, 2008). Presuming honest feedback in our study, the relationship between self-report and objective emotional data was only consistent with overall measures, not for individual time points. Self-assessment of emotions is a difficult task in any context. Add to that the novelty and stress associated with adventure activities and the post hoc recall of emotions, and the assessment can quickly become murky. Other measurement techniques have addressed shortcomings of self-report in previous research. Experience sampling provides timely assessments at specified intervals, but still relies on self-report (Csikszentmihalyi & Hunter, 2003). Biomarkers such as heart rate, sweat response, and cortisol have shown promise in outdoor research (Ewert, 2015). However, physiological measurements (e.g., heart rate and sweat) are often subtle and delayed, while chemical samples (e.g., cortisol) must be collected and analyzed in a laboratory. Ideally, a combination of methods would be used to triangulate results. Biomarker research (including EEG) relies on equipment that is currently cumbersome and expensive, making it impractical for common field applications. As the technology becomes more portable and user friendly, it will be feasible to measure real-time mind-set and physiological reactions of participants during certain activities. This information could inform one-on-one or group discussions, and methods of facilitating growth and transfer of learning.

Finally, physiological data trended in the same direction as cognitive-emotional data, but did not reach levels of significance. Given the subtle nature of the changes, it is possible that statistical and practical significance are at odds in this regard. For example, while some participants will exhibit clear physiological responses to stressful stimuli (e.g., shaking legs, rigid or tense movements, elevated sweating), others may experience internal dissonance that does not translate into clear physical indicators. Regardless of any display of obvious stress signals (or self-reports), their internal reactions likely influenced their sense of self-efficacy and mastery (Schumann & Sibthorp, 2014). Biofeedback could aid participants in becoming aware of their own mental state to control unconscious stress responses, prevent the wasteful use of caloric resources, and enhance performance.

Limitations and Directions for Future Research

Although instructive, the results of this study should be understood within the context of its limitations. First, the small, heterogeneous sample limits extrapolation of the data. While it is common for fMRI and EEG studies to include smaller samples, the results are often replicated many times to ensure validity. Small sample size could be addressed through correlated components analysis (Aspinali et al., 2015) or autoregressive multivariate time-series analysis (Box, Jenkins, Reinsel, & Ljung, 2015). These analyses would allow for assessment of patterns over time, rather than a comparison of snapshots from selected time points. Such analyses would require detailed activity tracking (i.e., timestamped video) not currently available for these devices. In addition, it is possible that the expensive technology and social conditions (e.g., peer pressure) could be spurious factors. This will need to be addressed in future research. Finally, field research using EEG is a relatively new phenomenon, and the performance of the headsets and fitness trackers in various environments is still being tested. Artifacts produced from muscle movements or electronic devices could interfere with data collection. While that did not appear to be a problem in this study, verification from future research must be accomplished to determine a measure of error. Accepting these limitations, this line of research offers promise as a way to provide personal feedback instantaneously in an objective way. Future research will test, verify, and hopefully replicate these findings.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Andrew W. Bailey is Assistant Professor of Sport, Outdoor Recreation, and Tourism at the University of Tennessee, Chattanooga, USA. His research interests include positive psychology, program modeling and experiential education. A healthy routine of outdoor adventures helps him to maintain focus, develop research ideas and enjoy life to the full.

Josh Johann is currently a project assistant in the Office of Research and Sponsored Programs at the University of Tennessee, Chattanooga, USA. His research focuses on physical activity and health. A personal and professional interest in outdoor adventure activities led him to collaborate on the line of inquiry for this study

Hyoung-Kil Kang is Assistant Professor of Department of Physical Education at Kyungnam University, Korea. He specializes in quality-of-life research, demonstrating expertise in quantitative methodologies and analyses. He abides by his research findings, promoting subjective well-being in academia and in daily life.

References

Ahuja

Ozdalga

Aaronson

(2017). Integrating mobile fitness trackers into the practice of medicine. American Journal of Lifestyle Medicine, 11(1), 77-79. doi:10.1177/1559827615583643

Amodio

D. M.

Master

S. E.

Yee

C. M.

Taylor

S. E.

(2008). Neurocognitive components of the behavioral inhibition and activation systems. Psychophysiology, 45, 11-19.

Aspinali

Mavros

Coyne

Roe

(2015). The urban brain: Analysing outdoor physical activity with mobile EEG. British Journal of Sports Medicine, 49, 272-276. doi:10.1136/bjsports-2012-091877

Ayabe

Kumahara

Morimura

Tanaka

(2013). Epoch length and the physical activity bout analysis: An accelerometry research issue. BMC Research Notes, 6, Article 20. doi:10.1186/1756-0500-6-20

Badcock

N. A.

Preece

K. A.

de Wit

Glenn

Fieder

Thie

McArthur

(2015). Validation of the Emotiv EPOC EEG system for research quality auditory event-related potentials in children. PeerJ, 3, e907. doi:10.7717/peerj.907

Bailey

A. W.

(2014, November). Physiological impacts of adventure activities: Is it all in our minds? Presented at the Wilderness Risk Management Conference, Atlanta, GA.

Bailey

A. W.

Fernando

(2011). Decoding the voluntourism process: A case study of the Pay It Forward Tour. Journal of Experiential Education, 33, 406-410.

Beames

Brown

(2016). Adventurous learning: A pedagogy for a changing world. New York, NY: Routledge.

Box

G. E. P.

Jenkins

G. M.

Reinsel

G. C.

Ljung

G. M.

(2015). Time series analysis: Forecasting and control. Wiley. Retrieved from https://books.google.com/books?id=lCy9BgAAQBAJ

10.

Bradley

M. M.

Lang

P. J.

(1994). Measuring emotion: The self-assessment manikin and the semantic differential. Journal of Behavior Therapy and Experimental Psychiatry, 25, 49-59.

11.

Brown

(2008). Comfort zone: Model or metaphor? Australian Journal of Outdoor Education, 12(1), 3-12.

12.

Chase

D. L.

(2015). Does challenge by choice increase participation? Journal of Experiential Education, 38(2), 108-128. doi:10.1177/1053825914524057

13.

Choi

Gutierrez-Osuna

(2009, June). Using heart rate monitors to detect mental stress. Presented at the Sixth International Workshop on Wearable and Implantable Body Sensor Networks, Berkeley, CA. Retrieved from http://research.cse.tamu.edu/prism/publications/bsn09_choi.pdf

14.

Clarke

A. R.

Barry

R. J.

Heaven

P. C. L.

McCarthy

Selikowitz

Byrne

M. K.

(2008). EEG in adults with attention-deficit/hyperactivity disorder. International Journal of Psychophysiology, 70, 176-183. doi:10.1016/j.ijpsycho.2008.07.001

15.

Coelli

Sclocco

Barbieri

Reni

Zucca

Bianchi

A. M.

(2015). EEG-based index for engagement level monitoring during sustained attention. Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2014, 1512-1515. doi:10.1109/EMBC.2015.7318658

16.

Csikszentmihalyi

Hunter

(2003). Happiness in everyday life: The uses of experience sampling. Journal of Happiness Studies, 4, 185-199.

17.

Dienstbier

R. A.

(1989). Arousal and physiological toughness: Implications for mental and physical health. Psychological Review, 96, 84-100.

18.

Dweck

(2006). Mindset: The new psychology of success. New York, NY: Ballantine Books.

19.

Ewert

(2015). The use of biomarkers in outdoor education research: Promises, challenges, and the development of evidence. Research in Outdoor Education, 13(1), 1-15.

20.

Ewert

Sibthorp

(2009). Creating outcomes through experiential education: The challenge of confounding variables. Journal of Experiential Education, 31, 376-389. doi:10.1177/105382590803100305

21.

Ewert

Yoshino

(2011). The influence of short-term adventure-based experiences on levels of resilience. Journal of Adventure Education and Outdoor Learning, 11, 35-50.

22.

Faul

Erdfelder

Lang

A. G.

Buchner

(2007). G*Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175-191.

23.

Fulmer

S. M.

Frijters

J. C.

(2009). A review of self-report and alternative approaches in the measurement of student motivation. Educational Psychology Review, 21, 219-246.

24.

Gillis

H. L.

Speelman

(2008). Are challenge (ropes) courses an effective tool? A meta-analysis. Journal of Experiential Education, 31, 111-135. doi:10.1177/105382590803100202

25.

Harmon-Jones

Gable

Peterson

C. K.

(2010). The role of asymmetric frontal cortical activity in emotion-related phenomena: A review and update. Biological Psychology, 84, 451-462.

26.

Henderson

(2004). Got research in experiential education? Theory and evidence. Journal of Experiential Education, 26, 184-189.

27.

Johnstone

Gunkelman

Lunt

(2005). Clinical database development: Characterization of EEG phenotypes. Clinical EEG and Neuroscience, 36, 99-107.

28.

Kahneman

(2011). Thinking, fast and slow. New York, NY: Farrar, Straus and Giroux.

29.

Kolb

(1984). Experiential learning: Experience as the source of learning and development. Englewood Cliffs, NJ: Prentice Hall.

30.

R. T.

Kling

S. R.

Salata

M. J.

Cupp

S. A.

Sheehan

Voos

J. E.

(2016). Wearable performance devices in sports medicine. Sports Health: A Multidisciplinary Approach, 8, 74-78. doi:10.1177/1941738115616917

31.

Lutz

Greischar

L. L.

Rawlings

N. B.

Ricard

Davidson

R. J.

(2004). Long-term meditators self-induce high-amplitude gamma synchrony during mental practice. Proceedings of the National Academy of Sciences of the United States of America, 101, 16369-16373. doi:10.1073/pnas.0407401101

32.

Mackenzie

S. H.

Son

J. S.

Hollenhorst

(2014). Unifying psychology and experiential education toward an integrated understanding of why it works. Journal of Experiential Education, 37, 75-88. doi:10.1177/1053825913518894

33.

Mai

J. K.

Paxinos

(2011). The human nervous system. Waltham, MA: Elsevier.

34.

McGonigal

(2015). The upside of stress. New York, NY: Avery Press.

35.

McKenzie

(2003). Beyond “the Outward Bound process”: Rethinking student learning. Journal of Experiential Education, 26, 8-23.

36.

Mutz

Müller

(2016). Mental health benefits of outdoor adventures: Results from two pilot studies. Journal of Adolescence, 49, 105-114. doi:10.1016/j.adolescence.2016.03.009

37.

Neill

J. T.

(2003). Reviewing and benchmarking adventure therapy outcomes: Applications of meta-analysis. Journal of Experiential Education, 25, 316-321. doi:10.1177/105382590302500305

38.

Oathes

D. J.

Ray

W. J.

Yamasaki

A. S.

Borkovec

T. D.

Castonguay

L. G.

Newman

M. G.

Nitschke

(2008). Worry, generalized anxiety disorder, and emotion: Evidence from the EEG gamma band. Biological Psychology, 79, 165-170.

39.

Paisley

Thurman

Sibthorp

Gookin

(2008). Student learning in outdoor education: A case study from the National Outdoor Leadership School. Journal of Experiential Education, 30, 201-222.

40.

Purves

(2012). Neuroscience. Sinauer Associates. Retrieved from https://books.google.com/books?id=B5YXRAAACAAJ

41.

Puth

M.-T.

Neuhäuser

Ruxton

G. D.

(2014). Effective use of Pearson’s product–moment correlation coefficient. Animal Behaviour, 93, 183-189. doi:10.1016/j.anbehav.2014.05.003

42.

Remington

Legge

(2017). Outdoor education in rural primary schools in New Zealand: A narrative inquiry. Journal of Adventure Education and Outdoor Learning, 17, 55-66. doi:10.1080/14729679.2016.1175362

43.

Roe

J. J.

Aspinall

P. A.

Mavros

Coyne

(2013). Engaging the brain: The impact of natural versus urban scenes using novel EEG methods in an experimental setting. Environmental Sciences, 1, 93-104. doi:10.12988/es.2013.3109

44.

Russell

J. A.

Barrett

L. F.

(1999). Core affect, prototypical emotional episodes, and other things called emotion: Dissecting the elephant. Journal of Personal Social Psychology, 76, 805-819.

45.

Schneider

Askew

C. D.

Diehl

Mierau

Kleinert

Abel

. . .Strüder

H. K.

(2009). EEG activity and mood in health orientated runners after different exercise intensities. Physiology & Behavior, 96, 709-716. doi:10.1016/j.physbeh.2009.01.007

46.

Schumann

Sibthorp

(2014). The development and scaling of the Teaching Outdoor Education Self-Efficacy Scale. Research in Outdoor Education, 12(1), 80-98.

47.

Sibthorp

(2003). An empirical look at Walsh and Golins’ adventure education process model: Relationships between antecedent factors, perceptions of characteristics of an adventure education experience, and changes in self-efficacy. Journal of Leisure Research, 35, 80-106.

48.

van de Mortel

T. F.

(2008). Faking it: Social desirability response bias in self-report research. The Australian Journal of Advanced Nursing, 25(4), 40-48.

49.

Walsh

Golins

(1976). The exploration of the Outward Bound process. Denver: Colorado Outward Bound School.

50.

Woo

Kim

Petruzzello

S. J.

Hatfield

B. D.

(2009). Examining the exercise-affect dose-response relationship: Does duration influence frontal EEG asymmetry? International Journal of Psychophysiology, 72, 166-172. doi:10.1016/j.ijpsycho.2008.12.003