Networked Flow in musical bands

Abstract

This study aimed at using the Networked Flow (NF) model to investigate group collaboration in the context of musical bands. We analyzed the relationship between flow, social presence, structural dynamics and performance as they related to 15 bands in a rehearsal room. Flow was measured using the Flow State Scale; social presence was assessed with the Networked Minds Social Presence scale; and interpersonal communication structure (exchange of gazes and verbal orders) was assessed by means of Social Network Analysis (SNA). In addition, we considered: (a) a subjective measure of performance, rated by each member on an ad-hoc questionnaire; and (b) an expert rating of performance, based on the evaluation of audio-video recordings of each group. Findings showed the multifaceted nature of the relationship between social presence and flow. Group flow score was a significant predictor of self-reported performance, but not of expert-evaluated performance. Moreover, several correlations were found between flow, social presence and patterns of interpersonal coordination (both implicit and explicit). Specifically, SNA reveals that flow was positively related to exchanges of gazes and negatively associated with exchanges of orders. Overall, this study contributes to further elucidating the complex interplay between group flow and intersubjective dynamics in music collaboration.

Keywords

flow music networked flow performance social presence

Flow was originally described by Csikszentmihalyi (1975/2000, 1990) as a positive and rewarding psychological state characterized by the following key dimensions: perceived balance between challenges and skills, complete absorption in the task at hand, sense of control, merging of action and awareness, clear goals, unambiguous feedback, loss of self-consciousness, distortion of temporal experience and intrinsic motivation. The experience of flow has been studied in several domains, including sports (Jackson & Kimiecik, 2008), work (Csikszentmihalyi & Csikszentmihalyi, 1992), leisure activities (Csikszentmihalyi, 1975/2000) and musical settings (Croom, 2012, 2015; Fullagar, Knight, & Sovern, 2013; Greasley & Lamont, 2011; Lamont, 2012; O’Neill, 1999; Sinnamon, Moran, & O’Connell, 2012; Wrigley & Emmerson, 2013).

In the music domain, the focus of research has been mainly on the investigation of flow in individual performers. Recently, however, there has been a growing interest in the study of flow in musical groups, including choirs (Freer, 2009), students engaged in musical composition tasks (Bakker, 2005; Byrne, MacDonald, & Carlton, 2003; MacDonald, Byrne, & Carlton, 2006) and improvisational bands (Sawyer, 2003, 2006, 2008; Hart & Di Blasi, 2015).

Freer (2009) performed a qualitative examination of flow experience among young choral singers. Data collected through individual interviews indicated that during flow experiences, singers perceived their performance as inseparable from that of the ensemble and individuals were able to monitor and adjust their singing to the feedback received from the surrounding ensemble. Bakker (2005) investigated to what extent flow may cross over from teachers to their students, using emotional contagion theory (Hatfield & Cacioppo, 1994). Findings showed that the more flow experiences music teachers reported, the higher the frequency of comparable experiences among their students, supporting the hypothesis that flow experience may cross over from one person to another.

Byrne et al. (2003) examined the relationship between flow and creative output of student musical compositions. Results revealed a positive correlation between quality of group compositions (as measured by creativity ratings) and individual flow experience. Interestingly, in a follow-up analysis, MacDonald et al. (2006) found that group levels of flow were more closely related to higher compositional creativity than individual levels. According to these authors, this may suggest that flow reflects not the individual, but the group dynamic that takes place during the compositional process.

Sawyer (a former student of Csikszentmihalyi) has been the first scholar to theorize the existence of group flow in music collaborations, defining it as a “collective state of mind” (Sawyer, 2007, p. 43). To investigate this construct, Sawyer (2003) focused on jazz improvisation, using a technique called “interaction analysis”, which consists of an in-depth observation and classification of participants’ conversations, gestures and body language. By examining data collected over 10 years of observations of several performing groups, Sawyer concluded that group flow is an optimal collective experience that occurs when members develop a feeling of mutual trust and empathy, in which individual intentions harmonize with those of the group. Jazz music players often refer to this state as achieving a “group mind” – a state characterized by a profound emotional resonance that allows artists to be fully coordinated within the improvisational flow. According to Sawyer, group flow cannot be broken down into the work of individuals; rather, this phenomenon emerges from the interactions occurring within a group and is able to positively influence overall performance.

Hart and Di Blasi (2015) used qualitative analysis to investigate group flow (which they termed “combined flow”) in musical jam sessions. These authors observed that combined flow shares several key dimensions with the individual flow experience. They also posit that combined flow is a process articulated in several stages (finding a niche; breaking through; finding the group groove; bridging sound to silence; sharing highs and lows) and characterized by a deep feeling of empathy among musicians.

In other words, according to both Sawyer (2003, 2007) and Hart and Di Blasi (2015), flow in groups resulted in an emergent process that is more than the simple sum of individuals’ flow.

Along this line of research, the goal of the present study was to further investigate group flow in musical bands by relating subjective experience of performers to structural features of musical collaboration. While Hart and Di Blasi (2015) focused on examining first-person experience of combined flow, we extended the analysis to (extramusical) communication variables, which were assessed retrospectively by participants. Specifically, we sought to determine whether group flow was associated with specific patterns of interpersonal coordination among band members.

To this end, we applied the “Networked Flow” (NF) model, which has been described by Gaggioli and colleagues (Gaggioli, Mazzoni, Milani, & Riva, 2015; Gaggioli, Riva, Milani, & Mazzoni, 2013). According to this model, the key to achieving an optimal group experience is the establishment of a “collaborative zone of proximal development”, in which the actions of the individuals and those of the collective are in balance and a sense of social presence is established. In the NF model, social presence is defined as the feeling of “being with other Selves” in a real or virtual space, resulting from the ability to intuitively recognize the intentions of Others in one’s surroundings (Biocca & Harms, 2003, 2011; Riva, 2008; Riva, Waterworth, Waterworth, & Mantovani, 2011). This process is characterized by three levels:

- Proto-social presence (or imitative), which is essentially based on the recognition of the motor intentions of the Other (Other versus the Self);

- Interactive social presence, which allows for the recognition of the intentions of the Other that are oriented towards the Self (Other toward the Self);

- Shared social presence (or empathic), which allows the Self to enter into resonance (like a tuning fork) with the intentions of the Other (Other like the Self).

Since these levels are hierarchically organized, the activation of the maximum level of social presence (empathy) requires the activation of the lower levels (namely, interactive and imitative social presence). In other words, shared social presence is associated with the highest level of empathy between the Self and the Other, allowing individuals to reach a state of “we-intentionality” where the objective of the Self and the objective of the Other become unified; this is well exemplified by the notion of “group mind” often described by jazz musicians (Sawyer, 2006, 2008). Empathic social presence allows for the emergence of an inter-subjective state, whose evolution is unpredictable but nonetheless coherent with the objective of each participant. From this perspective, group flow (Sawyer, 2006, 2008) can be regarded as the experiential correlate of such we-intentionality, associated with the highest level of social presence (Gaggioli et al., 2013). The NF model suggests that when high social presence has been achieved, participants can enjoy an optimal state that maximizes the creative potential of the group (networked flow). Moreover, the adjective “networked” has been used to stress the conceptualization of NF as a systemic emergence, resulting from the micro-interactions between the components of the group. Therefore, the NF model focuses on group collaboration not only as a process but also in structural terms. In particular, a key prediction of the model is that a group enjoying NF is characterized by specific network properties, which can be correlated with features of the group’s quality of experience and performance. To test the NF model in the context of music collaboration, we investigated the experiential and structural features of group music collaboration. More specifically, we were interested in modeling group socio-cognitive interactions by integrating both experiential and structural perspectives towards group collaboration.

Experiential data were collected in order to examine the quality of experience of group participants (intra-personal level: flow and social presence). Structural data were gathered to identify social network structures. For this purpose, we applied the technique of Social Network Analysis (SNA) on specific interpersonal communication variables, which were retrospectively assessed by participants (i.e., Which band member do you look at/give orders to most frequently while performing?). In line with the theoretical model of NF, the specific aims of the research were to test the following hypotheses:

H1: There is a positive relationship between flow and social presence in musical bands engaged in a live performance.

H2: There is a positive relationship between flow, social presence and team performance, as measured by both self-assessment and expert evaluation in the context of musical band.

H3: Bands that experience high levels of optimal experience (i.e., high level of flow and social presence) also have a unique group structure as assessed with SNA.

Ethical approval

The research protocol described in the following section was assessed and approved by the Ethical Committee of the Catholic University of the Sacred Heart. All participants signed an informed consent form before taking part in the experiment. Participants also signed a separate consent form to allow the researcher to videotape their live performance.

Methods

Sample and procedure

The study sample comprised 15 amateur musical bands composed of musicians and singers (75 adults: 64 men and 11 women) living in the Piedmont and Lombardy regions of Italy. Participants had a minimum of one year of experience of regularly playing music in a group setting. Their mean age was 30.81 (SD = 11.62). Band size varied from three to seven members. The mean longevity of each group was 36.20 months (SD = 32.26). Mean group expertise was 3.73 (SD = 0.44) on a 5-point scale.

Each band had to meet four inclusion criteria, drawn from the literature on group flow and music (Hart & Di Blasi, 2015; Sawyer, 2006, 2008), in order to participate in the research:

- Expertise: In order to prevent anxiety during videotaping, groups with members who had a low expertise level were not included (Hart & Di Blasi, 2015).

- Familiarity: Following this requirement, the study included only groups whose members had played together for at least six months. According to Sawyer, group flow is more likely to emerge in groups whose participants are well acquainted with each other (Sawyer, 2003). Furthermore, when participants have been playing together for a sufficient amount of time, individual members are more likely to provide a unique contribution to enable the group to reach high levels of performance (Hart & Di Blasi, 2015).

- Homogeneity: Following this requirement, the study included only groups that were homogeneous in terms of expertise, because each member should feel at ease with others in order to reach high levels of group performance (Hart & Di Blasi, 2015).

- The final requirement was that the group members were all over 18 years of age.

Each band was videotaped for two hours while playing in a rehearsal room. Data concerning flow, social presence, group performance and group structure were gathered through self-reported questionnaires completed individually by group members immediately after the performance in the rehearsal room. Finally, two expert raters (a singing teacher and a piano teacher) assessed the groups’ performance from the videotapes.

Measures and instruments

Except for expert-evaluated team performance, data were gathered through questionnaires completed by team members at the end of the performance. These questionnaires dealt with (a) individual flow level; (b) individual second-order social presence level; (c) self-reported-group performance; and (d) group structure.

Level of expertise

An ad-hoc questionnaire administered to the main contact person of each group was created with the following objectives: (a) to measure the individual and group level of expertise using a 4-item and 5-point answer scale in which each item refers explicitly in its formulation to the team expertise; (b) to gather group demographic data; and (c) to select bands meeting the inclusion criteria (i.e., expertise, familiarity, homogeneity, age > 18).

Flow

Individual flow was measured using the 36-item Flow State Scale (FSS; Jackson & Marsh, 1996; Jackson & Ecklund, 2004) in its Italian translation (Diana, Villani, Muzio, & Riva, 2012). Each item of this scale refers to one of the nine components of flow (Csikszentmihalyi, 1990; Jackson & Csikszentmihalyi, 1999), namely: challenge-skills balance (the balance between perceived challenges in the task at hand and perceived personal resources in facing those challenges); action-awareness merging (the degree to which an activity becomes automatic); autotelic experience (the extent to which the activity is self-rewarding); clear goals; concentration; loss of self-consciousness (the level of absorption in the activity); paradox of control (feeling in control without having to think about it); transformation of time (the subjective alteration of time perception); unambiguous feedback. The internal consistency of the FSS was acceptable, showing a mean Cronbach’s alpha for the 9 subscales of 0.83 (Jackson & Marsh, 1996). Scores for each subscale can be added to obtain a total flow score. The total score for each subscale ranged between 4 and 20, and the overall score ranged from 36 to 180. We chose this scale because it has been empirically validated in a musical context (Wrigley & Emmerson, 2013). In addition, the FSS gives a measure of flow level concerning a specific experience (i.e., playing in a rehearsal room). Global group flow level was computed by calculating the average flow score for each individual member of the band.

Social presence

For the purpose of the present study, the analysis focused on the second-order construct of social presence (interactive social presence), measured by adapting the Networked Minds Social Presence Inventory (NMSPI), developed by Biocca and Harms (2003, 2011). The 26 items included in NMSPI measure the following facets of social presence: (a) perceived attentional engagement (mutual perceived attention among group members); (b) perceived emotional contagion (the transfer of emotional states from a group member to the other, e.g., “I was sometimes influenced by my partner’s moods”); (c) perceived comprehension (the degree of transparency and shared meaning); (d) perceived behavioral interdependence (the degree to which one perceives that his/her behavior is to some degree dependent on the actions of another). The NMSPI demonstrated good internal consistency, showing a mean Cronbach’s alpha of 0.83 for the four subscales (ranging between 0.81 and 0.87). Moreover, the NMSPI was able to distinguish between levels of social presence experienced in unmediated face-to-face social interactions and mediated social interactions (criterion validity; Harms & Biocca, 2004). Global level of group social presence was measured by averaging individual social presence scores for each band.

Team performance: Self-report evaluation

Team performance was first self-assessed by each band member using a self-report 4-item scale created ad hoc for this research. A 5-point Likert scale (1 = very poor to 5 = excellent) was used. The evaluation concerned four key dimensions of performance: (a) Interpretation: the perceived ability to interpret an excerpt of music; (b) Technique: the perceived ability to play each song accurately without making mistakes; (c) Global performance: perceived satisfaction with the whole group performance; (d) Perceived competence: perceived ability to manage the whole group performance.

Team performance: Raters

The level of team performance was also evaluated by two raters on a 4-item, 5-point Likert scale ranging from 1 to 5 (1 = not at all to 5 = highly) created ad hoc for this study. Items refer to the four dimensions of performance. (a) Interpretation: the ability to interpret an excerpt of music; (b) Technique: the ability to play each song accurately without making mistakes; (c) Stage presence: the capacity to command spectators using an impressive style or manner; (d) Harmony: the atmosphere of compliance that can occur among group members. The raters were two music teachers of the Nonsolomusica art and music school. They were asked to participate in this research because of their long-standing expertise and high competence in the music field.

Group structure

Interpersonal coordination was measured using two indicators of information flow: (a) gazes among group members (i.e., which band member did you look at most frequently while you were playing?); and (b) verbal exchanges among group members (i.e., which member did you give the most orders to during the performance?). The first indicator referred to verbal exchanges and was linked to an explicit level of communication, while the second referred specifically to the musical context and was linked to an implicit level of communication during the performance, since communication during a musical performance is mostly implicit. Information flow data retrospectively gathered from participants were then analysed using Social Network Analysis (SNA) technique. SNA is a quantitative method of analysis concerning interactions, used in both real (Zohar & Tenne-Gazit, 2008) and virtual environments (De Laat, Lally, Lipponen, & Simons, 2007; Palonen & Hakkarainen, 2000). This method allows for the analysis of exchanges among group members (regarding, e.g., money, friendship, information) to provide a measure of group structure, so as to consider each member as interdependent and focus on their relationships instead of only on single actors. According to Wasserman and Faust (1994), SNA indexes (particularly centrality and centralization indexes) allow for the analysis of social dynamics such as influence (the more a person activates relations or exchanges, the more he/she is influential) and status (the more a person is the recipient of relations or exchanges, the more he/she could be seen to have high status). The former allows for understanding whether one or more band members influence the others by means of orders, while the latter helps to identify members to whom the others refer during the performance to direct their behavior. In this study, we assessed influence by means of verbal exchanges (which we called “interactions of dominance”) and status by means of the exchange of gazes (which we called “interactions of dependence”). Members who made the majority of orders during the performance can be considered as the most dominant, while the members who received the most gazes from the others can be seen as the leaders other members follow during the performance. Accordingly, we computed two types of graphs (one for relationships of dominance and the other for relationships of dependence) and six indexes for each group:

a) Neighborhood: it indicates the average number of individuals a group member looked at or interacted with;

b) Density: the relationship (ranging from 0 to 1) between the number of exchanges and the maximal number (Scott, 2000; Wasserman & Faust, 1994); here, it indicates the extent of interpersonal coordination.

c) In-Degree Centralization: it measures degree of centrality among members. High in-degree centralization indicates that the group is focused inward on a few core members. It describes if there are members with a higher status than others.

d) Out-Degree Centralization: it indicates the degree of variance of the total network of outgoing relationships identifying influential individuals in the network. It describes whether there are members with a higher influence than others.

e) Betweenness Centralization: it quantifies how other actors control or mediate the relations between connected nodes;

f) Dyad: it measures the group’s tendency to create a couple who interact reciprocally.

In SNA, relations can be described as dichotomous (relation is present/not present) or take on a range of values that indicate the strength, intensity or frequency of a relationship. In this research, we calculated dichotomous relations to perform correlation analyses and comparisons, and valued relations to analyze structural features of musical bands.

Results

Since the level of analysis of the variables is the group level, individual scores of flow, social presence and self-reported performance were aggregated to the group level by calculating the average and standard deviation of individual scores for each band. In the following tables, only group scores are presented. Further, correlation analyses were carried out to explore potential relationships among these variables. A Shapiro-Wilk test of normality showed that most measurement variables were not normally distributed. Consequently, Spearman’s rank correlation coefficient (Kendall & Smith, 1939) was used to test the correlations. We first present results concerning the relationships between social presence, flow and performance. Next, we examine relationships between these variables and SNA indexes. Only correlations with a significance level of p ⩿ .05 were considered.

Correlations between flow, social presence and performance

Results showed strong correlations between several subscales of flow and social presence. Paradox of control (ρ_s = 0.610; p < .05) and unambiguous feedback (ρ_s = 0.555; p < .05) were positively related to perceived comprehension. Paradox of control (ρ_s = -0.537; p < .05) and concentration (ρs = -0.606; p < .05) were negatively correlated with perceived attentional engagement. Loss of self-consciousness was positively correlated with mutual perceived attention (ρ_s = 0.629; p < .05). Autotelic experience was found positively correlated with perceived emotional contagion (ρ_s = 0.672; p < .01). Four simple linear regressions were conducted in blocks to predict aggregated self-evaluated performance variables (global performance, interpretation, technique and perceived competence) based on group flow scores. Results showed that flow score was a significant predictor of both global performance (R²= 0.321, F = 7.621; p < .05) and perceived competence (R² = 0.335; F = 11.145; p < .01). In contrast, group flow scores did not predict expert-evaluated performance variables. Regarding the relationship between social presence and self-reported performance, groups who felt more technically competent also perceived a high sense of shared comprehension (ρ_s = 0.571; p < .05) and experienced less mutual perceived attention among colleagues (ρ_s = −0.624; p < .05). Group age correlated negatively with flow scores. Results showed that older groups perceived that they required greater effort in performing the task (action-awareness merging: ρ_s = −0.589; p < .05). Further, bands characterized by higher longevity were also more self-critical about their performance (global performance: ρ_s = −0.538; p < .05; interpretation: ρ_s = −0.626; p < .05).

Correlations between SNA and flow

Table 1 shows that several flow dimensions correlated with SNA indexes. Almost all the correlations between dominance relationship indexes and flow dimensions were strongly negative. In particular, in groups characterized by a high sense of control, clear goals and a sense of time distortion, members were not inclined to receive or give orders. Furthermore, bands in which members felt highly concentrated on their performance tended to look more to each other.

Table 1.

Correlations between SNA indexes and flow.

Indexes	Variables				Time transformation
Indexes	Autotelic experience	Clear goals	Concentration	Paradox of control	Time transformation
Mean Neighborhood interaction of dependence	0.530^*		0.545^*
Density interaction of dependence			0.517^*
Betweenness Centralization interaction of dependence			0.576^*
Mean Neighborhood interaction of dominance		−0.740^**		−0.557^*
Density interaction of dominance		−0.647^**		−0.553^*
Out-Degree Centralization interaction of dominance					−0.527^*
Betweenness Centralization interaction of dominance		−0.653^**		−0.632^*
Dyad interaction of dominance		−0.549^*		−0.842^**

Note. N = 15 teams.

p < .05, two-tailed; **p < .01.

Correlations between SNA and social presence

Group members perceived more reciprocal comprehension when there were few interactions of dominance among them (Mean of Neighborhood interaction of dominance: ρ_s = −0.571; p < .05; Betweenness Centralization interaction of dominance: ρ_s = −0.688; p < .01; Dyad interaction of dominance: ρ_s = −0.762; p < .01). They experienced a sense of deep emotional contagion when many exchanges of gazes were activated (Mean of Neighborhood: ρ_s = 0.566; p < .05; Density: ρ_s = 0.546; p < .05). Finally, group members paid more attention to each other when they were inclined to give orders to each other (Dyad interaction of dominance: ρ_s = 0.561; p < .05).

Correlations between SNA and performance

Self-evaluated performance

Group members felt more technically competent when fewer exchanges of dominance occurred (Betweenness centralization interaction of dominance: ρ_s = −0.545; p < .05; Dyad interaction of dominance: ρ_s = −0,771; p < .01) and when group communicative structure was decentralized concerning exchanges of dependence (In-Degree Centralization interaction of dependence: ρ_s = −0.642; p < .01). Finally, group members were less satisfied regarding their performance when they were inclined to reciprocally give orders to each other (Dyad interaction of dominance: ρ_s = −0.564; p < .05).

Expert-evaluated performance

Raters evaluated groups as more in harmony when exchanges of dominance were directed mostly towards a specific member (In-Degree Centralization interaction of dominance: ρ_s = 0.565; p < .05). When the group communication structure was centralized concerning exchanges of dependence (i.e., some members were inclined to look more to others), raters evaluated bands as more technically competent (Density interaction of dependence: ρ_s = 0.569; p < .05; Betweenness Centralization interaction of dependence: ρ_s = 0.547; p < .05).

SNA: Descriptive comparison of high-flow and low-flow bands

As a means of illustrating the Networked Flow model, we analyze the interactional structure of four bands characterized by the highest (groups 5 and 14) and lowest (groups 6 and 12) flow levels. We first present descriptive statistics of flow and presence scores for each group (Table 2); then, we describe the SNA profile of selected bands.

Table 2.

Group levels of flow and social presence in selected bands.

Flow			Social presence
Groups	Mean	SD	Mean	SD
5	142.25	13.11	95.75	10.50
14	139.25	5.90	90.00	13.29
6	116.00	19.06	78.20	15.30
12	113.67	19.69	90.83	6.99

For each band selected, we present two types of directed graphs (Figure 1). Graphs concerning interactions of dominance (exchange of verbal orders) are shown on the right, while graphs regarding interactions of dependence (exchange of gazes) are presented on the left. In each graph, group members are represented as points and exchanges are represented as lines. The direction of the vector codes the direction of exchange. Finally, “weights” of exchanges are indicated above each line. In the Social Network Analysis “weight” refers to the frequency of exchanges. For example, in the left graph of group 5 (interactions of dependence), Guitarist looked three times at Bassist, Drummer and Vocalist. Further, in the right graph of group 5 (interactions of dominance), Drummer gave orders to Bassist six times. Vocalist, as well as Guitarist gave orders to Bassist three times. Moreover, since exchanges could also be reciprocal, a single vector could show two directions coding the reciprocity of the exchanges between two members. For example, in the left graph of group 5 (interactions of dependence), Guitarist and Vocalist interacted reciprocally.

Figure 1.

Two types of directed graphs. Left column shows sociograms of groups with the highest flow levels (group 5 and 14). Right column shows sociograms of the lowest flow groups (group 6 and 12). For each group, interactions of dependence are represented on the left side, while interactions of dominance are represented on the right side.

The comparison of SNA graphs and data of groups with highest flow (5 and 14) and lowest flow (6 and 12) showed distinguishable interaction patterns, both in terms of gazes and exchange of verbal orders. Specifically, the analysis revealed that three SNA indexes seemed to discriminate between groups with high and low flow levels, namely, In-Degree Centralization, Out-Degree Centralization and Dyadic Reciprocity. In particular, for interactions of dependence, high-flow groups showed lower In-Degree Centralization (Group 5 = 22.22; Group 14 = 24.49) and higher Out-Degree Centralization (Group 5 = 66; Group 14 = 8.16) than low-flow groups (In-Degree Centralization: Group 6 = 37.5; Group 12 = 76; Out-Degree Centralization: Group 6 = 6.25; Group 12 = 4). In other words, groups experiencing higher levels of flow were not characterized by an explicit leadership, and few or no members checked the others while playing. In contrast, in bands characterized by low flow (group 6 and group 12), some member received more gazes than others. For example, graphs of group 12 showed that most gazes were directed to a single member (i.e., the guitarist). Moreover, in group 6 most gazes were oriented towards the drummer and the keyboardist. By looking at dominance relations, the most evident difference between high-flow and low-flow groups is that the former presented the lowest out-degree centrality (Group 5 Out-Degree Centralization = 11.11; Group 14 Out-Degree Centralization = 24.49), indicating that no one in these groups was particularly dominant over the others. On the other hand, reciprocal interactions of dominance emerged as key features in groups characterized by a low-flow level (Group 6 Dyad rec. = 0.42; Group 12 Dyad rec. = 0.22).

Discussion

The findings of this study are discussed in light of the hypotheses formulated earlier.

H1: There is a positive relationship between flow and (second-order) social presence in musical bands engaged in live performance.

In line with this hypothesis, strong correlations were observed between flow and social presence dimensions. Specifically, band members perceived higher intrinsic motivation (“autotelic experience”) when they felt more emotionally connected (“emotional contagion”). Furthermore, players felt more in control and enjoyed clearer feedback when they perceived higher mutual comprehension. These findings support the idea that emotive dimensions of social presence are key to the emergence of group flow (Engeser, 2012; Gaggioli et al., 2013). Further, members’ intrinsic motivation was found to be related to increases in perceived emotional contagion. This may indicate a “flow contagion” among players (Bakker, 2005; Hatfield & Cacioppo, 1994). Indeed, as observed by Koelsch, Offermanns, and Franzke (2010) and Koelsch (2013, 2014), a feeling of co-pathy (a collective state in which inter-individual emotional states become more homogeneous) can emerge in groups playing music, and promote social cohesion. It is noteworthy, however, that there was an inverse correlation between flow dimension of concentration and control with social presence dimension of mutual perceived attention. This paradoxical finding suggests that the relationship between flow and social presence is more complex than originally assumed by the NF model. A possible explanation of this apparently contradictory finding is that group members needed to concentrate deeply on their own task in order to perceive enough control over it to enter into flow (Nakamura & Csikszentmihalyi, 2009). After they had reached this optimal experience, they also needed to make an effort to stay in this state. Therefore, they were less able to pay attention to their colleagues at the same time. This could be the cause of (a) the negative relationship between attentive dimensions of flow and social presence, and (b) the concurrent emergence of the phenomenon of “shared mutual comprehension” among members. In other words, flow would help group members to feel invisibly connected with colleagues both at an emotive and cognitive level, even though it limited their ability to direct their attention to each other. In this regard, an interesting negative relationship was identified between flow and group age; specifically, the older members experienced a great deal of effort when engaging in the task. Perhaps age acted as a hindrance, making it difficult to manage adequately the initial effort to enter into flow.

H2: There is a positive relationship between flow, social presence and team performance, as measured by both self-assessment and expert evaluation.

Results showed that group flow scores predicted self-reported performance, but not expert-evaluated performance. This finding supports De Manzano, Theorell, Harmat, and Ullén’s (2010) hypothesis that flow perceived by musicians during their performance could function as an internal reward, also influencing their self-evaluation. In other words, if members were in flow, they felt more satisfied regarding their performance, maybe because of the sense of self-rewarding that flow generates, which may also sustain individuals’ well-being (Freer, 2009). On the other hand, the absence of a causal relationship between group and expert-evaluated performance does not align with previous findings (Aubé, Brunelle, & Rousseau, 2014; Byrne et al., 2003; MacDonald et al., 2006). As De Manzano et al. (2010) and Chirico, Serino, Cipresso, Gaggioli, and Riva (2015) suggested, a balance between emotive and cognitive factors of flow is key for an optimal musical performance. Accordingly, it could be that a disequilibrium between these components was perceived by raters but not by executers, who were more focused on the intense and engaging experience of playing together. Indeed, when members were more focused on their own performance, they also (a) perceived the invisible link mentioned in the discussion of H1 (i.e., the sense of “Perceived mutual comprehension”) and (b) felt more competent. On the other hand, when members paid more attention to the group than to their own task, they felt less technically competent. It seemed that when members collected information concerning the group performance not only from themselves but also from other members, they also felt less technically able. Perhaps the increasing of cognitive load, which could restore the balance between cognitive and emotive components of flow, provided them with a more objective perspective towards the whole performance.

Finally, results showed band longevity could be another key factor influencing satisfaction with group performance. Indeed, bands characterized by higher “longevity” are also more self-critical regarding their own work. Therefore, it is possible that as members of long-lasting bands become increasingly well acquainted with each other, they also become more aware of others’ potential and thus more demanding of the group. Even though people who have known each other for a long time collaborate more efficiently (MacDonald, Miell, & Mitchell, 2002; Sawyer, 2003, 2007), they may also be more demanding and expect higher levels of performance.

H3: Bands that experience high levels of optimal experience (i.e., high level of flow and social presence) also have a unique group structure as assessed with SNA.

Data concerning flow and group structure suggested that in groups characterized by a deep sense of perceived control, members were not inclined to give reciprocal orders but preferred communicating through the exchange of gazes. Moreover, concerning social presence, group members were not inclined to give orders to each other when there was a low level of mutual attention perceived, or if members felt a high level of mutual comprehension. Finally, the performance data suggested that when groups felt technically competent, they did not communicate using words but rather through gazes. In detail, there was a high level of perceived technical competence among group members when there were few interactions of dominance (i.e., few verbal exchanges), and when no specific members acted as mediators of gazes. In fact, raters also evaluated bands with a decentralized structure (i.e., gazes were directed to all members rather than to a specific member) as more technically competent. Nevertheless, an indicator of group harmony from an expert perspective appeared to be the centralization of exchanges of dominance. In other words, groups in which members gave orders to one or more specific members were perceived as more in harmony by the raters.

Finally, it is possible that in a musical context the internal communication of a band experiencing a high level of technical competence, concentration and control is thus characterized by “decentralized” gazes instead of words.

Conclusion

This research aimed at applying the Networked Flow model to group collaboration in the context of musical bands. Consistent with the main hypotheses of the NF model, findings showed that the emergence of optimal group experience was associated with specific social presence dimensions. In particular, the analysis of these correlations suggests the existence of a “flow contagion” among group members, characterized by high intrinsic motivation, shared positive emotions, and mutual comprehension. Moreover, group flow predicted self-reported performance, confirming previous findings linking flow and performance in the music domain (Hart & Di Blasi, 2015; Marin & Bhattacharya, 2013; Sinnamon et al., 2012). These observations support the benefits of bands’ use of self-monitoring instruments, such as video recording, in order to learn more about their own performance. Indeed, musical bands are usually self-managed, therefore, having an external viewpoint (i.e., a video-recording of group performance in rehearsals) is the best way for them to learn about themselves and improve their group performance.

Group structure seemed to play a key role in the emergence of group flow. In particular, high-flow groups were characterized by specific interpersonal coordination patterns. Specifically, group flow was positively associated with exchanges of gazes and negatively associated with exchanges of orders. This suggests that in high-flow groups, members tended to rely more on nonverbal communication based on the visual channel.

A main limitation of the present study was the small size of the sample investigated and the retrospective nature of the assessment instruments used. Therefore, it is necessary to analyze in greater depth the relationship between flow performance and social presence in a musical setting, for example by testing the hypothesis of the existence of different dynamics of flow, in order to clearly understand their implications for group performance, or, alternatively, by investigating how this model works in a longitudinal perspective.

Footnotes

Funding

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

References

Aubé

Brunelle

Rousseau

(2014). Flow experience and team performance: The role of team goal commitment and information exchange. Motivation and Emotion, 38(1), 120–130.

Bakker

A. B.

(2005). Flow among music teachers and their students: The crossover of peak experiences. Journal of Vocational Behavior, 66(1), 26–44.

Biocca

Harms

(2003). Guide to the Networked Minds Social Presence Inventory v. 1.2: Measures of co-presence, social presence, subjective symmetry, and intersubjective symmetry. East Lansing: Michigan State University. Retrieved from http://cogprints.org/6743/

Biocca

Harms

(2011). Networked Minds Social Presence Inventory (scales only, version 1.2). East Lansing: MIND Labs, Michigan State University. Retrieved from http://cogprints.org/6742/

Byrne

MacDonald

Carlton

(2003). Assessing creativity in musical compositions: Flow as an assessment tool. British Journal of Music Education, 20(3), 277–290.

Chirico

Serino

Cipresso

Gaggioli

Riva

(2015). When music “flows”. State and trait in musical performance, composition and listening: A systematic review. Frontiers in Psychology, 6, 906. doi:10.3389/fpsyg.2015.00906

Croom

A. M.

(2012). Music, neuroscience, and the psychology of well-being: A précis. Frontiers in Psychology, 2, 393.

Croom

A. M.

(2015). Music practice and participation for psychological well-being: A review of how music influences positive emotion, engagement, relationships, meaning, and accomplishment. Musicae Scientiae, 19, 44–64.

Csikszentmihalyi

(1990). Flow: The psychology of optimal experience. New York, NY: Harper & Row.

10.

Csikszentmihalyi

(2000). Beyond boredom and anxiety. San Francisco, CA: Jossey-Bass. (Original work published 1975)

11.

Csikszentmihalyi

I. S.

(1992). Optimal experience: Psychological studies of flow in consciousness. Cambridge, UK: Cambridge University Press.

12.

De Laat

Lally

Lipponen

Simons

R.-J.

(2007). Investigating patterns of interaction in networked learning and computer-supported collaborative learning: A role for Social Network Analysis. International Journal of Computer-Supported Collaborative Learning, 2(1), 87–103.

13.

De Manzano

Ö.

Theorell

Harmat

Ullén

(2010). The psychophysiology of flow during piano playing. Emotion, 10(3), 301–311. doi:10.1037/a0018432

14.

Diana

Villani

Muzio

Riva

(2012). La validazione italiana della Flow State Scale – FSS. In Muzio

Riva

Argenton

(Eds.), Flow, benessere e prestazione eccellente: Dai modelli teorici alle applicazioni nello sport e in azienda (pp. 123–142). Milano, Italy: FrancoAngeli.

15.

Engeser

(2012). Advances in flow research. New York, NY: Springer.

16.

Freer

P. K.

(2009). Boys’ descriptions of their experiences in choral music. Research Studies in Music Education, 31(2), 142–160.

17.

Fullagar

C. J.

Knight

P. A.

Sovern

H. S.

(2013). Challenge/skill balance, flow, and performance anxiety. Applied Psychology, 62(2), 236–259.

18.

Gaggioli

Mazzoni

Milani

Riva

(2015). The creative link: Investigating the relationship between social network indices, creative performance and flow in blended teams. Computers in Human Behavior, 42, 157–166.

19.

Gaggioli

Riva

Milani

Mazzoni

(2013). Networked flow: Towards an understanding of creative networks. Dordrecht, the Netherlands: Springer.

20.

Greasley

A. E.

Lamont

(2011). Exploring engagement with music in everyday life using experience sampling methodology. Musicae Scientiae, 15(1), 45–71.

21.

Harms

Biocca

(2004). Internal consistency and reliability of the Networked Minds Social Presence measure. In Alcaniz

Rey

(Eds.), Seventh Annual International Workshop: Presence 2004. Valencia, Spain: Universidad Politecnica de Valencia. Retrieved from http://cogprints.org/7026/1/Harms_04_reliability_validity_social_presence_(Biocca).pdf

22.

Hart

Di Blasi

(2015). Combined flow in musical jam sessions: A pilot qualitative study. Psychology of Music, 43(2), 275–290. doi:10.1177/0305735613502374

23.

Hatfield

Cacioppo

J. T.

(1994). Emotional contagion. Cambridge, UK: Cambridge University Press.

24.

Jackson

S. A.

Csikszentmihalyi

(1999). Flow in sports. Champaign, IL: Human Kinetics.

25.

Jackson

S. A.

Eklund

(2004). The flow scales manual. Morgantown, WV: Fitness Education Technology.

26.

Jackson

S. A.

Marsh

H. W.

(1996). Development and validation of a scale to measure optimal experience: The Flow State Scale. Journal of Sport and Exercise Psychology, 18, 17–35.

27.

Jackson

S. A.

Kimiecik

J. C.

(2008). The flow perspective of optimal experience in sport and physical activity. In Horn

T. S.

(Ed.), Advances in sport and exercise psychology (3rd ed., pp. 377–399). Champaign, IL: Human Kinetics.

28.

Kendall

M. G.

Smith

B. B.

(1939). The problem of m rankings. The Annals of Mathematical Statistics, 10(3), 275–287.

29.

Koelsch

(2013). From social contact to social cohesion – the 7cs. Music and Medicine, 5, 204–209.

30.

Koelsch

(2014). Brain correlates of music-evoked emotions. Nature Reviews Neuroscience, 15(3), 170–180.

31.

Koelsch

Offermanns

Franzke

(2010). Music in the treatment of affective disorders: An exploratory investigation of a new method for music-therapeutic research. Music Perception, 27(4), 307–316.

32.

Lamont

(2012). Emotion, engagement and meaning in strong experiences of music performance. Psychology of Music, 40(5), 574–594. doi:10.1177/0305735612448510

33.

MacDonald

Byrne

Carlton

(2006). Creativity and flow in musical composition: An empirical investigation. Psychology of Music, 34(3), 292–306. doi:10.1177/0305735606064838

34.

MacDonald

R. A.

Miell

Mitchell

(2002). An investigation of children’s musical collaborations: The effect of friendship and age. Psychology of Music, 30(2), 148–163. doi:10.1177/0305735602302002

35.

Marin

M. M.

Bhattacharya

(2013). Getting into the musical zone: Trait emotional intelligence and amount of practice predict flow in pianists. Frontiers in Psychology, 4. doi: 10.3389/fpsyg.2013.00853

36.

Nakamura

Csikszentmihalyi

(2009). Flow theory and research. In Lopez

S. J.

Snyder

C. R.

(Eds.), The Oxford handbook of positive psychology (2nd ed., pp. 195–206). Oxford, UK: Oxford University Press.

37.

O’Neill

(1999). Flow theory and the development of musical performance skills. Bulletin of the Council for Research in Music Education, 141, 129–134.

38.

Palonen

Hakkarainen

(2000). Patterns of interaction in computer-supported learning: A social network analysis. In Fishman

O’Connor-Divelbiss

(Eds.), Fourth International Conference of the Learning Sciences (pp. 334–339). Mahwah, NJ: Lawrence Erlbaum. Retrieved from http://www.umich.edu/~icls/proceedings/pdf/Palonen.pdf

39.

Riva

(2008). Psicologia dei nuovi media. Bologna, Italy: Il Mulino.

40.

Riva

Waterworth

J. A.

Waterworth

E. L.

Mantovani

(2011). From intention to action: The role of presence. New Ideas in Psychology, 29(1), 24–37.

41.

Sawyer

R. K.

(2003). Group creativity: Music, theater, collaboration. Mahwah, NJ: Lawrence Erlbaum.

42.

Sawyer

R. K.

(2006). Group creativity: Musical performance and collaboration. Psychology of Music, 34(2), 148–165. doi:10.1177/0305735606061850

43.

Sawyer

R. K.

(2007). Group genius: The creative power of collaboration. New York, NY: Basic Books.

44.

Sawyer

R. K.

(2008). Group genius: The creative power of collaboration: New York, NY: Basic Books.

45.

Scott

(2000). Social network analysis: A handbook. London, UK: Sage.

46.

Sinnamon

Moran

O’Connell

(2012). Flow among musicians: Measuring peak experiences of student performers. Journal of Research in Music Education, 60(1), 6–25.

47.

Wasserman

Faust

(1994). Social network analysis: Methods and applications. Cambridge, UK: Cambridge University Press.

48.

Wrigley

W. J.

Emmerson

S. B.

(2013). The experience of the flow state in live music performance. Psychology of Music, 41(3), 292–305. doi:10.1177/0305735611425903

49.

Zohar

Tenne-Gazit

(2008). Transformational leadership and group interaction as climate antecedents: A social network analysis. Journal of Applied Psychology, 93(4), 744–757. doi: 10.1037/0021-9010.93.4.744