Enhancing Mental Health and Social Connection Through an Acute Online Dance Intervention


 Dance is a multidimensional form of physical activity that includes sensory, motor, cognitive, rhythmic, creative, and social elements. Long-term, interventional studies in dance have shown positive effects on both mental and social health; however, little has been done to examine the acute effects and no studies to date have explored the relationship between the mental health and social outcomes of dance. We examined the hypothesis that online dance can acutely improve mental health and social connection during a time of social isolation, namely, the COVID-19 crisis. Healthy adults (age ≥18; n=47) completed a single session of 60 minutes of self-selected online dance, completing a series of validated self-reported questionnaires before and after class. We found that online dance acutely improves mental health as measured by increased positive affect and self-esteem and decreased negative affect and depression. Additionally, online dance acutely enhanced social and community connectedness. Further, we found that those who experienced the largest increases in self-esteem and decreases in negative affect demonstrated the largest gains in social connectivity. Although in-person dance classes may be optimal for dance technique training, online dance instruction has important clinical benefits. We conclude that through online dance, individuals can experience a body-mind-community connection.


Introduction
The positive effects of exercise on mental health are well established, both in terms of acute and longterm effects [1,2]. These effects include but are not limited to, mood enhancement, anxiety prevention, and sleep regulation [3], and a signi cant negative relationship exists between physical activity level and mood disorders, including major depression, panic disorder, and social phobia [4]. Dance differs from other physical activities in that dance is a multimodal artistic practice that incorporates aspects of sensory, motor, cognitive, social, emotional, rhythmic, and creative processes [5]. In this way, we can think of dance as an enriched form of physical activity. Animal studies have shown that the combination of physical exercise with sensory enrichment has the most prominent effect on the survival of new neurons and their integration into brain circuits [6,7]. Though exercise is one of the best ways to promote neurogenesis (i.e., the birth of new neurons), other cognitive enrichment experiences are needed for those new neurons to become an integral part of the brain and contribute to its functional activity. In fact, studies suggest that dance may be superior to aerobic exercise as a way to increase brain plasticity and possibly prevent or delay impairments in mood or cognition in both healthy, aging, and other clinical populations [8,9].
Previous research has shown that dance is bene cial for mental health issues, including depression, anxiety, schizophrenia, and disordered eating [10][11][12]. Qualitative studies have shown that dance reduces self-reported symptoms of depression, with participants providing four primary reasons for the bene cial effects, including dancing for their own health, social acceptance, connection with others, and not wanting to stop due to unexpected bene ts [13]. Additionally, dance has been shown to reduce psychological distress and improve stress management, with effects lasting up to 6 months after the intervention [14]. Importantly, dance has also been shown to improve mental well-being among non-clinical populations, especially amongst children and the elderly [10][11][12]. In the realm of improving social connection, dance has also shown favorable outcomes. Murrock and Graor [13] studied disadvantaged adults who experience depression from social isolation and their response to a 12-week dance intervention. The research revealed feelings of social acceptance, connection with others, and adoption of a group identity among participants. Further, Bognar et al. [15] found that patients with Parkinson's disease developed a sense of social connectedness through dance, improving outlook on life and decreasing feelings of negativity toward disease diagnosis. These studies have focused on long-term interventions of at least 6 weeks in duration; however, a signi cant gap in the literature is the focus on the acute effects of dance on psychological health. Additionally, no studies to date have examined the relationship between mental health and social outcomes of dance.
The COVID-19 pandemic increased rates of social isolation from as low as 10% across the USA, Europe, and China to upwards of 24%, with millennials experiencing loneliness more than any other age group [16,17]. As a result, recent studies detail the mental health impacts of COVID-19, including increased symptoms of anxiety, depression, stress, avoidance behaviors, insomnia, anger, and fear [18,19]. These alarming symptoms call for a need to address an increasing mental health crisis along with the physical health crisis we face. Recent literature shows that in the middle of a pandemic, exercise has been particularly bene cial in boosting mood, especially in combatting the effects of depression, stress, and exhaustion, all of which are often experienced by people in quarantine [20].
Due to stay-at-home orders, social media, video conferencing, and other forms of digital communication have seen an increased use, not only for personal reasons but for conducting necessary work and school functions [21]. Dance classes have made a similar shift; since March 2020, many studios, companies, and teams have transitioned to online collaboration or have canceled in-person movement classes completely. This presents a unique challenge to dance teachers as they learn how to use a twodimensional format of education for a skill that is so three-dimensional in nature. Gingrasso [22] identi es several obstacles in virtual dance instruction, such as an inability to give tactile cues or not being able to see the whole body in movement due to screen restrictions. However, while some may argue that online learning inhibits social connection, Smith-Merry et al. [23] identi es online forums as a useful tool in maintaining social connectivity during times when a face-to-face connection is limited. We hypothesize that online dance instruction may be a way to combat social isolation and mental health issues resulting from the pandemic. Therefore, the present study examined the effects of an online dance class on mental health and social connectivity during the COVID-19 pandemic. We hypothesized that a single 60-minute session of online dance would signi cantly decrease negative affective states and increase positive affective states. Additionally, we hypothesized that dance would acutely increase levels of social connectivity and that the enhancements in the affective state would predict these social outcomes. These ndings have important clinical implications for individuals experiencing social isolation during the pandemic and otherwise.

Methods
This acute experimental study consisted of a repeated-measures design; participants took a self-report, questionnaire-based assessment both before and after an online dance class. Participants were recruited via online communication through social media, advertisements on Dancing Alone Together (dancingalonetogether.org), networking with New River Moving Arts (newrivermovingarts.com), and Virginia Tech's Listserv resources. Interest emails were directed to the Embodied Brain Lab, where participants signed up for a class using their email addresses. Once participants signed up for the class using Sign Up Genius (signupgenius.com), they received an email with a link to the pre-participation Qualtrics survey. Potential participants were eligible to complete both pre-and post-intervention surveys if they were age 18 or older, living within the United States, and English speaking (since the dance classes were taught in English). Participants also needed to pass the Physical Activity Readiness Questionnaire (PAR-Q) [24] before participating to indicate capability of engaging in physical activity. Additional exclusionary criteria included not ambulatory and intellectual impairment that would impact psychological assessment or intervention participation, as our aim was to study primarily healthy populations. Participants were screened for former or current psychiatric diagnoses and medication intake, but these did not serve as exclusionary criteria; rather, these pieces of information provided researchers with a broader picture of the mental and physical health of the individual. All methods were approved by the Virginia Tech Committee on Activities Involving Human Subjects and were performed in accordance with the relevant guidelines and regulations.
During the pre-intervention survey, participants signed an electronic informed consent, then answered a series of questions to assess demographic information, including age, gender, race, ethnicity, employment status, education, income, living environment, height, weight, and body mass index (BMI).

Mental Health Measures
Our primary study endpoints included mental health measures, which were assessed using the following scales: 1) self-esteem as assessed by the Rosenberg Self-Esteem Scale [25]; 2) positive and negative affect as assessed by the Positive and Negative Affect Schedule -Short Form [26]; 3) anxiety as assessed by the Beck Anxiety Inventory [27]; and 4) depression as assessed by the Beck Depression Inventory [28].
Rosenberg Self-Esteem Scale [25]: The Rosenberg Self-Esteem Scale (RSES) is the most widely used selfesteem measure; it consists of a 10-item scale that measures positive and negative feelings towards oneself using a 4-point Likert scale. Total scores range from 0-30, with a score lower than 15 suggesting low self-esteem (Cronbach's alpha=0.72 to 0.87). Beck Anxiety Inventory [27]: The Beck Anxiety Inventory (BAI) is a valid and reliable tool consisting of a 21-item questionnaire, where items are scored using a 4-point Likert scale. The sum of the items yields a total anxiety score ranging from 0 to 63 (Cronbach's alpha=0.83 to 0.92).
Beck Depression Inventory [28]: The Beck Depression Inventory (BDI) is a consistent and reliable tool that utilizes a 21-item self-reported inventory to measure the severity of depressive symptoms (Cronbach's alpha=0.92).

Social Connection Measures
Social measures, another primary endpoint, included the following: 1) social connectedness as assessed by the Social Connectedness Scale [10]; 2) loneliness as assessed by the UCLA Loneliness Scale [29]; and 3) community inclusion as assessed by the Inclusion of Community in Self Scale [30].
Social Connectedness Scale [10]: The Social Connectedness Scale (SCS) is a highly reliable and valid tool used to assess the degree of an individual's connection to others within their social environment. It utilizes a 20-item questionnaire that is scored using a 6-point Likert scale (Cronbach's alpha=0.94).
UCLA Loneliness Scale [29]: The UCLA Loneliness Scale is a validated tool that measures a person's subjective feelings of loneliness and social isolation. It is a 20-item questionnaire that allows participants to self-rate each item as "often," "sometimes," "rarely," or "never" (Cronbach's alpha=0.96).
Inclusion of Community in Self Scale [30]: The Inclusion of Community in Self Scale (ICS) is a reliable and valid single-item measure of community connectedness. Participants self-select one of seven pictorial models that best identify their feelings towards their relationship to their community, represented by circles that converge to varying degrees.
Subjective Exercise Experiences Scale [31]: The Subjective Exercise Experiences Scale (SEES) assesses three categories of subjective response to exercise: positive well-being, psychological distress, and fatigue. It is a questionnaire that is highly consistent across a variety of populations and consists of a 12item scale, with each item rated using a 7-point Likert scale (Cronbach's alpha=0.85 to 0.88).
Study personnel reviewed the pre-intervention assessment to ensure that each participant was eligible for the study. If the participant was deemed ineligible, they were sent an email informing them of their inability to be included in the study. Eligible participants were sent a standard email con rming their participation with the date and time of their scheduled dance class. Twenty-four hours prior to the class, participants were sent a reminder and a Zoom link to the class. At their scheduled time, participants logged into Zoom and engaged in a 60-minute online dance class. Participants were asked to activate their camera (to ensure similar levels of engagement with the instructor and other participants), but had the option to mute their microphones. Additionally, participants were reminded to clear an approximately 8-by 10-foot space to minimize any risks associated with dancing in their home. Furthermore, choreography was taught and group performance/observation was included at the end of class. These classes were taught by New River Moving Arts faculty and included three genres of dance: modern, ballet, and jazz. Participants were able to self-select their preferred dance genre prior to the class. All classes were taught at a beginner level, but movement options were offered for each dance exercise to accommodate dancers of all skill levels. Participants received the dance class at no cost.
Thirty minutes into the dance class, the participants received an emailed link to a post-intervention Qualtrics survey; this was to encourage the greatest adherence in nishing both surveys by having the email already waiting in their inbox when class ended. At the end of class, dance instructors reminded the participants to complete the survey within two hours after the Zoom meeting ended. This interval was selected because the greatest number of acute physiological effects of exercise exist within 120 minutes after cessation of activity [1]. All primary endpoint measures were re-assessed in the post-intervention survey as part of the repeated-measures study design. The SEES was only given at the post-intervention survey as a secondary endpoint measure.

Data Analysis
To determine the sample size to su ciently power our study, an a-priori power analysis was conducted using the UCSF Clinical and Translational Science Institute correlation sample size calculator (https://sample-size.net/). The power analysis was based on a correlation with an alpha ( ) of 0.05, a beta (β) of 0.20, and a correlation coe cient (r) of 0.4, with results indicating a sample size of n=47.
All data were analyzed using SPSS Statistics 26.0 [32]. Change scores were calculated by subtracting the pre-intervention from the post-intervention measures. Paired-samples t-tests were used to determine whether there was a statistically signi cant difference between the pre-intervention and post-intervention outcomes for all affective state and social connectivity measures. Further, we used Pearson's productmoment correlations to examine the relationship between changes in affective state measures as well as the subjective exercise experience. Finally, we used linear regression models to analyze the predictive validity of dance-induced changes in affective state on changes in social connection. For all analyses, statistical signi cance was assessed utilizing an alpha value of 0.05.

Demographics
We collected data from n=59 participants; however, the nal number of participants who completed the dance class and all research procedures was n=47. The age of participants ranged from 18 to 70, with a mean age of 32.47 years (+/-2.32 SEM). Of these participants, 68% had an annual household income in the middle-or high-income categories, 92% had a college or advanced degree, and 38% worked full time (Table 1). Our study contained more female participants (n=43) than male participants (n=4), primarily of Caucasian descent (77%) and non-Hispanic (98%). Urban and rural communities were nearly equally represented at 49% and 51%, respectively. Further, 55% of participants were in the BMI range for normal weight (BMI of 18.5-24.9), compared to the national obesity prevalence of 42.4% in adults [33].
We further compared our ve affective state measures using Pearson's correlation (Figure 2; Table 2). There was a statistically signi cant negative correlation between positive affect and negative affect change scores (Figure 2A; r = -.427, p=.003), as well as a negative correlation between positive affect and BDI change scores ( Figure 2B; r = -.493, p<.0001). Self-esteem was also analyzed using Pearson's correlation and yielded a signi cant negative correlation to positive affect (r= -.317, p=.030). Inversely, the data demonstrates that a negative affective state is positively correlated with both BAI change scores ( Figure 2C; r = .691, p<.0001) and BDI change scores ( Figure 2D; r = .507, p<.0001). The correlation between BAI and BDI change scores is also signi cant (r = .629, p<.0001).
The psychological response to exercise predicts changes in mental health SEES positive well-being was positively associated with the change in positive affect experienced after dance (r= .356, p= .014). SEES positive well-being was also negatively associated with the change in anxiety experienced after dancing (r= -.320, p= .028). Psychological distress was positively associated with the change in both anxiety (r= .567, p<.0001) and depression (r= .328, p= .025). Fatigue levels also demonstrated positive correlations with the change in psychological distress (r= .622, p<.0001) and anxiety (r= .422, p= .003).

Social Connection Outcomes Acute Effects of Dance on Social Connectivity
Online dance signi cantly enhanced social connection (t(46)= -3.069, p= .004) and community connection (t(46)= -2.183, p= .034) (Figure 3). Positive correlations were also seen between social and community connectedness (r = .328, p= .025). There were no statistically signi cant effects associated with the UCLA Loneliness Scale.
The affective state response predicts social connectedness Those who experienced the largest decreases in negative affect also demonstrated the largest gains in social (r = -.317, p=.030) and community connectivity (r= -.295, p=.044). Further, we found that participants with greater self-esteem experienced the largest gains in social (r= -.441, p=.002) and community (r= -.420, p=.003) connectedness. A linear regression model established that self-esteem increases could statistically signi cantly predict enhanced social connection ( Figure 4A; F(1, 45) = 10.863, p = .002), and improvements in self-esteem accounted for 19.4% of the explained variability in social connectivity increases with adjusted R 2 = 17.7%. To assess linearity, a scatterplot of social connectivity against self-esteem change with a superimposed regression line was plotted. Visual inspection of these two plots indicated a linear relationship between the variables. An identical linear regression model established that decreases in negative affect predicted enhanced social connection ( Figure 4B; F(1, 45) = 5.044, p = .030). Negative affect accounted for 10.1% of social connection variability with adjusted R 2 = 8.1%. A visual inspection of the plotted regression line indicates a negative variable relationship.

Discussion
The current study examined the effects of an acute session of online dance on mental health and social connection during the COVID-19 crisis. We found that online dance acutely decreased negative affect and depression and enhanced positive affect. Additionally, we found that the subjective experience of dance signi cantly predicted the change in mental health. Further, we found that online dance acutely enhanced social and community connectedness. Importantly, the change in mental health signi cantly predicted the change in social connection. This suggests that dance, even via an online platform, can be used to improve mental and social health, suggesting a body-mind-community connection. These ndings have important implications for both healthy and clinical populations dealing with social isolation and resulting mental health issues during times of a pandemic or otherwise. We suggest that online platforms can be used effectively to disseminate dance to diverse populations.

Dance Acutely Enhances Mental Health
Regarding affective state, our data shows that a dance intervention increases positive affect and selfesteem while minimizing negative affect and depression. Further, decreases in negative affect and depression were signi cantly associated with both decreases in anxiety and increases in positive affect. While several researchers have examined the effects of dance on depression and anxiety in clinical populations, fewer studies have speci cally analyzed positive and negative affect. Current research demonstrates that Dance Movement Therapy (DMT) focused on elements associated with happiness can signi cantly enhance feelings such as empowerment, pride, and determination, which are part of positive affective states [34]. More closely related to our study, Koch et al. [35] investigated the impacts of a single dance movement session on depression and positive affect in 31 psychiatric patients diagnosed with depression. Comparing dance, listening to music, and riding a stationary ergometer, they found that the dance group pro ted most in terms of decreased depression and more vitality. Additionally, a recent, large randomized controlled trial (RCT) demonstrated that DMT reduced negative affect, depression, and loneliness in older adults with mild dementia whereas exercise alone did not [36]. These ndings are analogous to our study; a similar effect was seen after a single dance session, suggesting that a dance class is at least as effective, if not more effective, in improving affective states when compared to other forms of exercise. While our study was not designed as an RCT and did not explicitly compare dance to other forms of exercise, the repeated measures design shows promising affective improvements.

Dance Acutely Enhances Social Connectivity
Current evolutionary theories posit that dance has evolved as a form of imitation for the purposes of social communication, connection, and learning [37]. The current study demonstrated that online dance signi cantly enhanced acute social connection and community connection with positive correlations between social and community connectedness. This nding coincides with the social alignment theory, where motor, cognitive, and emotional synchrony happens as humans build relationships and activate areas of the brain associated with the action observation network [5,38,39]. As individuals enter a dance practice, they use motor and cognitive areas of the brain to process and execute choreography, but they also exhibit emotional expression through their movements. These three elements then contribute to synchronization and harmony of movement with other dancers, increasing feelings of social connectivity [40]. In our study, as participants danced, social alignment and connection took place without physically being in the same room as the other participants. Previous research has shown that implementing social inclusion strategies through an online-based forum is strongly suggested in the mitigation of negative effects from con nement [41]. Our research is notable in the fact that it is the rst to study the link between dance and social connection in healthy populations.

Subjective Exercise Experience Predicts the Acute Mental Health Response of Dance
The subjective exercise experience also in uenced affective state in that it was a signi cant predictor of affective state changes. This nding is similar to other work showing that individuals who participated in movement with other dancers showed an increase in subjective enjoyment of the dance experience [42].
In our study, we saw that positive well-being experienced from dance enhanced positive affect and decreased feelings of anxiety. This nding is supported by the research of Campion and Levita [43] who compared the effect of dance on affect and cognition to music or exercise in a young, non-clinical population. Their research demonstrated that both dancing and passively listening to music enhanced positive affect, decreased negative affect, and reduced feelings of fatigue. Our results also demonstrated that those who experienced psychological distress and fatigue in response to acute dance had ampli ed anxiety and depression levels. We hypothesize that this nding could be related to perceived class di culty level and corresponding increases in anxiety.

Dance and the Body-Mind-Community Connection
Additionally, we found that the acute effects of dance on mental health signi cantly predicted the change in social connection. Speci cally, those individuals who demonstrated the largest gains in self-esteem and decreases in negative affect showed the largest enhancements in social connectivity. This is the rst time that this relationship has been investigated in the context of an acute dance protocol.
Other work has shown that mental health and social connectivity are inextricably linked. Individuals with low levels of social connectedness show impaired mental and physical health, including increased levels of depression and shorter life expectancies than those with strong social bonds [44,45]. In fact, in the realms of public health and epidemiology, it is well accepted that social connection acts as a protective mechanism against mental illness [46,47]. Conversely, mental health impacts one's ability to engage in social interactions, and there is often a lack of social connection in individuals with depression, anxiety, or substance use disorders due to dysregulated interpersonal processes [48,49].
In related work, one study found that a 4-week physical-activity based youth development program for low-income youth improved social and physical competence as well as physical and global self-worth [50]. Further, they found that the changes in self-competence predicted the changes in mental health measures including self-worth and hope. Additionally, a 3-month Gero t exercise program in older Veterans signi cantly improved posttraumatic stress disorder symptoms, and this improvement was signi cantly associated with the level of social connectedness [51].
Our work shows that dance, even in an online platform, can enhance both mental health and social connection and that these effects are integrally linked. We hypothesize that our video conference software, which allowed participants to see each other during all dance instruction, signi cantly contributed to this effect. Future research should investigate which aspects of the online platform (e.g., camera on versus off; gallery view versus speaker view) support enhancements in mental and social health. Additionally, online physical activity programs that have a social component, such as the one in the current study, have been shown to enhance engagement in physical activity. Importantly, the relationship between app use and physical activity level is mediated by the level of social support experienced [52]. As physical activity, even in acute doses, is known to enhance mental health (e.g., increase positive affect, decrease negative affect) [1], having a social component may be of integral importance to sustain physical activity in service of improving mental health.

Limitations and future directions
We acknowledge several limitations of the current study. First, as we did not include a control group, future RCTs are warranted. Second, as females tend to self-select dance experiences, and our sample was made of primarily females (91.5%), more males will need to be intentionally incorporated in future research studies. Though there was an equal sampling between urban and rural communities, future iterations should include more racially and ethnically diverse populations. Additionally, only three dance styles were represented based on the expertise of the instructors (ballet, jazz and contemporary/modern); we see potential for future expansion of the project into hip hop, tap, non-Western forms, or social dances.
Based on the current ndings, we suggest some potential directions for future research. First, though this study was su ciently powered for its cross-sectional nature, future studies should increase the sample size, perhaps including comparisons between sexes or differences seen across dance styles. Second, future work will need to investigate measures beyond self-report, such as neurocognitive assessments as well as the neural mechanisms underlying the bene cial effects of online dance on mental and social well-being. We also see potential in how this intervention could be applied to other forms of exercise or movement classes (e.g., yoga and other mindfulness practices), as well as its application in clinical populations, such as individuals with autism spectrum disorder, depression, anxiety, and posttraumatic stress disorder.

Conclusions
The present study examined the effects of an acute online dance class during the COVID-19 crisis on mental and social connectedness. We found that even a single session of online dance enhanced mental health by increasing positive affect and self-esteem, decreasing negative affect and depression, and enhancing social and community connectedness. Further, those individuals who experienced the largest improvements in mental health demonstrated the most signi cant gains in social connectivity. Future study should increase breadth and depth of both sample size and population diversity in order to maximize mental health and physical wellness bene ts across a wide array of participants. Ultimately, we anticipate that the utility of online dance interventions can increase access to these mental health and social connection improvements in both healthy and clinical subjects.

Declarations
Ethics approval and consent to participate: All work was approved by the Virginia Tech Committee on Activities Involving Human Subjects (IRB-20-415) and was performed in accordance with the relevant guidelines and regulations. All participants signed an informed consent prior to participation in the study.

Consent for publication: Not applicable
Availability of data and materials: The datasets generated and/or analysed during the current study are not publicly available due to the complicated nature of the data variables but are available from the corresponding author on reasonable request.
Competing interests: The authors have no competing interests as de ned by BMC, or other interests that might be perceived to in uence the results and/or discussion reported in this paper.
Funding: This work is supported by the National Center for Advancing Translational Sciences of the National Institutes of Health under Award Numbers UL1TR003015 and KL2TR003016.
Authors' contributions: J.C.B conceptualized the study. AH helped design the study and carried out all practical aspects of the study. RR helped design the intervention and implement the study, especially as it relates to the dance intervention. AH and JCB conducted data analysis, and AH created all gures and tables. All authors helped to write, edit, and review the manuscript.  Acute online dance signi cantly enhanced community and social connectedness. *Statistically signi cant at p<0.05

Figure 4
A. Dance-induced increases in self-esteem were signi cantly associated with enhancements in social connection. B. Dance-induced decreases in negative affect were signi cantly associated with enhancements in social connection. *Statistically signi cant at p<0.05

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.