Physical activity and social relationships are critical “flashpoints” for health policy [1, 2]. Low levels of physical activity and high levels of loneliness have been independently associated with poorer mental and physical health outcomes and mortality [3]. Despite the established benefits of sufficient physical activity and secure social ties for mental and physical health, levels of physical inactivity are extremely high globally [e.g., 4, 5]. Although evidence on the prevalence of loneliness is less well established, loneliness as the perception of social isolation (even when among other people), is also widely recognized as a “real mental health challenge for the nations” [6] and a growing problem worldwide [7, 8]. In the world’s first government loneliness strategy [9] and in its first Loneliness Annual Report [10], the UK government committed to improving the evidence base on loneliness through the use of standard measures [7, 8].
Medicalist perspectives predominate in assessments of the negative effects of physical inactivity and loneliness (e.g., morbidity-mortality risks, “pandemic” terminology). Nevertheless, these are widely considered to be complex social issues that require collaborative, integrated and holistic public health approaches [11]. Indeed, evidence suggests that one problem compounds the other, with loneliness having been identified as a risk factor for physical inactivity [12, 13] and physical inactivity as a risk factor for psychiatric and psychosocial health problems that are directly or indirectly associated with depression and loneliness [14, 15]. Physical inactivity and loneliness can therefore be approached as interlinked problems that require integrated solutions. Toward this general aim, this paper contributes new theoretical and empirical perspectives on the behavioural and psychological synergies between physical exercise and social relationships.
Although considerable research attention has been directed at understanding how social-environmental factors influence physical activity behaviour [e.g., 16], connections between the affective dimensions of physical activity, particularly exercise, and sociality remain underappreciated. Previous research offers some promising clues. For example, positive affect in exercise – i.e., liking or enjoyment – is a key determinant of adherence [17] and, in general, intense emotional experiences happen more often in the context of interdependent social connection and belonging than in independent situations [18]. This suggests a possibility for social enhancement of positive affect in physical exercise [19–22], with corresponding increases in motivation and adherence. In addition, there are links between collective physical activity in diverse forms, such as play, sport, dance, and exercise, and feelings of social bonding and belonging [e.g., 20, 23], which in turn are associated with positive motivation and adherence [24]. Taken together, these links trace a virtuous circle between affectively rich, meaningful and positive social connections and intrinsically motivated engagement and enjoyment in collective physical activity.
Besides links to social-motivational psychology, a large and growing literature spanning the behavioural and psychological sciences, physiology and neuroscience has also elucidated the effects of perceived social support on the homeostatic regulation of stress, fatigue and pain [e.g., 25, 26, 27]. Applying these insights to the physical activity domain, it can be hypothesised that perceptions of social support buffer, or reduce, unpleasant exercise-induced affect, such as fatigue and pain [19], thereby potentially increasing performance outputs, sense of achievement, and engagement in exercise.
Despite the apparent connections among affective and behavioural dimensions of exercise and sociality, surprisingly little research has directly investigated the effects of either the rewarding or buffering aspects of social bonding and support on exercise experiences. Identifying how social experiences modulate positive affect, feelings of energy and fatigue, and performance in exercise contexts can begin to provide much-needed insight into the public health value of exercise in social settings, including community-based physical activity events and interventions [28]. It can also inspire new research directions on the psychobiological pathways via which human sociality, psychology and biology co-regulate one another in exertive physical activity more generally.
We examine these questions in the context of parkrun, a community-based organisation that convenes free, weekly, timed 5 km runs in public parks and spaces. Since its inception in 2004, parkrun has seen rapid and sustained international growth; at the time of writing, parkrun events occur in over 2,000 locations in 22 countries worldwide. The research adds to a growing literature on the role of community-led collective exercise initiatives in facilitating and promoting positive social and exercise experiences [e.g., 29, 30, 31].
1.1. Exercise in social context
Despite wide recognition and understanding of the impact of social-demographic factors on physical activity uptake and maintenance [28], relatively little research has investigated the psychological and biological pathways by which fundamental elements of cooperative sociality (e.g., social bonding, belonging and support) influence exercise experiences and performance [32]. Traditional social psychological approaches in sports and exercise science, which originate as far back as the early observations of Norman Triplett in the late 19th century, have focused on competitive and evaluative aspects of social presence and their facilitating effects on motivation, effort and performance across different types of task [for a review, see 33]. Early studies on “social facilitation” [34] demonstrated that social presence enhanced simple task performance but impaired complex task performance [35, 36]. Subsequently, following Cottrell’s [37] social evaluation account of facilitation effects (beyond “mere presence”) the bulk of research focused on anxiety-related reactions to potentially evaluative and threatening social presence on performance in technically complex vs. simple tasks (see [38]). In this paradigm, mechanistic explanations have emphasised arousal, evaluation apprehension, and distraction caused by the presence of others. More recent research has shown that personality traits (e.g., neuroticism, extraversion, self-esteem) can moderate orientations toward social presence also via positive appraisals, and that personality effects contribute more to predicting performance than the level of task complexity [e.g., 39]. This hints at the many blind spots that remain in our understanding of how, when, why and on what outcomes social facilitation operates, particularly via positively valanced social perceptions.
In a largely separate line of research, team cohesion has been studied as a predictor of effort and performance in sport. For example, group cohesion in sport settings – defined as “a group dynamic process that is reflected in the tendency of a group to stick together and remain united in the pursuit of instrumental objectives and/or for the satisfaction of member affective states” [40; p. 213] – positively predicts performance success [40], adherence to group exercise programmes [41] and physical exertion in team sports [42]. Related research drawing inspiration from social identity theory has begun to identify the importance of social-group identities for promoting physical activity engagement, adherence, enjoyment and effort. For example, in a recent parkrun study, Stevens et. al [31] found that stronger identification with the parkrun running group positively predicted participation, life satisfaction, exercise-specific satisfaction, and group cohesion. Similarly, Graupensperger et. al. [43] identified links between subjective evaluations of groupness (i.e., the perception of a group as an interdependent unit in which members adopt roles and group norms) and exertion, enjoyment and affective valence in a fitness class context. In a preliminary study with recreational runners (n = 17), Carnes and Mahoney [44] also found that higher perceived task cohesion (defined as individual involvement in and group unity around common goals) and social support and were associated with lower perceived exertion during group running, and higher enjoyment in both interval and group running. These studies offer support for the idea that individuals’ perceptions of the social group as cohesive and supportive, and with which they can strongly identify as group members, can promote positive affective exercise experiences, increase participation in physical activity, and facilitate performance via socially-mediated mechanisms other than arousal, evaluation apprehension and distraction.
Overall, a large and well-developed literature has elucidated the types, causes and consequences of group cohesion in sports. However, the psychological, physiological, and performance effects of bonded sociality outside of team or group-based sport (e.g., in transient collectives, virtual settings, and exercise groups without clearly defined boundaries, interdependent roles and shared goals) remain relatively unexplored, both theoretically and empirically. Research across a wider range of activities, social contexts and outcome measures is needed in order to determine how social behaviours and perceptions of social support, integration and belonging influence exercise-related affect, physiology and performance, and to better understand the public health significance and value of community exercise schemes and events.
In the present study, we examine how social behaviours as well as perceptions and cues of social connectedness and support affect the experiences and performance of exercisers in a naturalistic setting (parkrun). We hypothesised that social engagement and perceptions of social support would increase positive affect (e.g., feelings of enjoyment and energy), decrease negative affect (e.g., feelings of fatigue), and enhance performance outputs (i.e., decrease run times). These hypotheses are theoretically situated in a broad evolutionary approach to humans as thoroughly social creatures, for whom surviving and thriving depends on social connections [45–47]. In this view, the social environment directly influences the brain-body signalling that undergirds adaptive homeostatic maintenance in everyday life, including in physical exercise, via affective states experienced as pleasure-displeasure [48–51]. Moreover, the intrinsic pleasure of social connection that is experienced in the context of close bonds is thought to arise in part via activation of endogenous neurobiological systems, such as the opioidergic and endocannabinoid systems [47, 52, 53], that are also involved in modulating responses to nociceptive stimuli and in sustaining endurance exercise [54–57]. Social bonding in the context of exercise can therefore potentially boost feelings of enjoyment and pleasure in exercise while also buffering pain and fatigue.
Our hypotheses are corroborated by an extensive clinical and neuroscientific literature on the positive effects of social support in stress, pain, immunity and healing responses [26, 58]. Across this literature, perceived social support (i.e., the belief or expectation of available help when needed) has proven to be a more consistent predictor of positive health outcomes than received social support (i.e., help given). There is also preliminary evidence for buffering effects of perceived social connection and support in exercise. For example, experimental studies have found that exercising with others versus alone leads to significantly greater pain thresholds, and cues to social bonding prior to an individual sprint challenge improve performance outputs, without corresponding increases in subjective ratings of fatigue [19, 21, 59].
Altogether, existing research suggests that rewarding and supportive social environments can influence pleasure or enjoyment during physical activity, that social support can improve performance via reduced fatigue and increased energy, and that these socially-derived experiences and effects can increase motivation and engagement in exercise activities. Distinct from social facilitation, group cohesion and social identification, perceived social support refers specifically to the individual’s belief that they are valued and cared about by others who will offer help when needed [60]. As a consistent predictor of physical and mental health, and an effective buffer of stress, it seems unlikely that the hypothesised positive effects of social reward and support on exercise experiences and performance can be wholly or primarily attributed to social facilitation factors, such as evaluation pressure and self-monitoring. Perceptions of social support are anchored in our cooperative sociality, and are thought to be associated – evolutionarily, ontogenetically and cognitively - with perceived safety and resource availability [50, 61]. As such, perceived social support aligns more closely with existing constructs of team cohesion and social identification. Importantly, however, the effects of social reward and support on homeostatic regulation and affective states in exercise can occur outside or alongside team cohesion or group identification processes, and can be sourced not only in co-actors, but also family, friends, fans and even rivals.
1.2. The current study
In the current study, we investigate how perceived social support and behaviours associated with social reward influence feelings of enjoyment, fatigue, and energy as well as objective performance among parkrunners. The aim of parkrun is to “promote physical activity and community spirit, by providing supportive opportunities to exercise” [30], p. 171). Organised by local volunteers and describing itself as a ‘run and not a race’ with no ability or attendance requirements, parkrun offers a welcoming, community-based setting for participants to attend as often as they please. This inclusive ethos allows parkrun to attract participants who do not identify with traditional views on competitive running, thus creating a participating population that includes a wide range of demographic categories and skill levels [29, 30]. Researchers have described parkrun as a “mass community event” and a “home away from home” that offers participants “psychological comfort, warmth, and mutual support” [29], p. 10). While parkrun strives to create an inclusive ethos – it has a relatively even gender split (reported as a binary variable), and all age groups are well-represented across sites – it should be noted that individuals from ethnic minorities and low-socio-economic status (SES) groups are underrepresented at parkrun, including in areas that are ethnically diverse [62].
Parkrun’s community-based ethos makes it an ideal naturalistic setting in which to study how social connectedness and support influence exercise experiences. According to a study conducted at one UK parkrun site, ‘social togetherness’ was the second most important aspect of parkrun among surveyed participants, following “getting exercise”[29]. Current literature on parkrun and our own ethnographic observations in the context of this research suggest that social relationships are built and solidified through interactions among parkrunners before and after runs [29, 30, 63]. Individuals report chatting while running alongside other parkrunners, and often begin casual conversations as they wait in line (ordered by finishing rank) to scan their parkrun ID barcodes (used to record run times). Although participants at a given parkrun location can vary from week to week, there is a continuity of core runners and volunteers that work to create a sense of community, which offers runners an informal network of emotional support where “even loose ties with relative strangers can be highly valued” [29, p. 10; see also Morris & Scott, 2018].
Using survey and publicly available performance data, we assessed the influence of social factors on exercise experiences and performance. Specifically, we were interested in how participating with others, and particularly close others, as well as feelings of being integrated within the parkrun community influenced perceptions of fatigue, feelings of energy, enjoyment, and objective performance (run times). Although parkrun has an overall ethos that is supportive and not generally competitive, participants do care about their run performance, with many citing a desire to improve upon previous run times [30, 63].
Three social predictor variables were included: 1) whether participants attended with friends or family vs. attended alone; 2) whether or not participants interacted with others socially before the event; 3) the degree to which participants felt a) supported by, and b) integrated into the parkrun community. Previous research suggests that social reward and support effects vary according to relationship quality and are strongest when individuals are socially integrated [58, 65]. We hypothesised that higher ratings on these social predictor variables would be associated with lower perceptions of fatigue (H1), higher levels of felt energy (H2) and higher levels of enjoyment (H3). We also hypothesised that higher ratings on social variables would be associated with faster run times (H4). Finally, we hypothesised that subjective feelings of fatigue (H5) and energy (H6) would mediate associations between social variables and run times, with higher energy and lower fatigue predicted by higher ratings on social variables, and these, in turn, predicting faster run times (see Table 1). To assess general social facilitation effects, we also analysed associations between social variables and ratings of perceived effort. If social behaviour and perceived support and integration are associated with higher enjoyment, increases in feelings of energy, decreases in perceptions of fatigue, and increases in performance outputs, this suggests an important but underappreciated role for social reward and support in exercise experiences, motivation and adherence, and in exercise-related health benefits.
Table 1
|
H1. Main effects of social predictors on subjective fatigue
|
1.1–1.3
|
Higher subjective ratings of community support and integration will predict lower fatigue (1.1); coming or meeting up with friends/family will predict lower fatigue (1.2); being social (vs. not being social) before the run will predict lower fatigue (1.3).
|
|
H2. Main effects of social predictors on subjective energy
|
2.1 – 2.3
|
Higher subjective ratings of community support and integration will predict higher energy (2.1); coming or meeting up with friends/family will predict higher energy (2.2); being social (vs. not being social) before the run will predict higher energy (2.3).
|
|
H3. Main effects of social predictors on subjective enjoyment
|
3.1 – 3.3
|
Higher subjective ratings of community support and integration will predict higher enjoyment (3.1); coming or meeting up with friends/family will predict higher enjoyment (3.2); being social (vs. not being social) before the run will predict higher enjoyment (3.3).
|
|
H4. Main effects of social predictors on 5 km run times
|
4.1 – 4.3
|
Higher subjective ratings of community support and integration will predict faster 5 km run times (4.1); coming or meeting up with friends/family will predict faster 5 km run times (4.2); being social (vs. not being social) before the run will predict faster 5 km run times (4.3).
|
|
H5 & H6. Mediators of main effects of social predictors on 5 km run times
|
5.1–5.3
|
Higher scores on the social predictor variables (5.1: community support and integration | 5.2: coming or meeting up with friends/family | 5.3: pre-run sociality) will predict higher perceived energy levels, and higher perceived energy levels will predict faster 5 km run times.
|
6.1–6.3
|
Higher scores on the social predictor variables (6.1: community support and integration | 6.2: coming or meeting up with friends/family | 6.3: pre-run sociality) will predict lower fatigue, and lower fatigue will predict faster 5 km run times.
|