Applying the constraints-led approach to facilitate exploratory learning of the volleyball serve

ABSTRACT

Exploration is an important feature for successfully learning motor skills. However, game rules such as one attempt to serve in volleyball could discourage exploration due to an individual’s fear of making a mistake and forfeiting a point. The constraints-led approach is a coaching methodology that encourages exploration by selectively manipulating task constraints such as rules. Therefore, the aim of this study was to examine whether the addition of the task constraint of a second serve would encourage volleyball players to use their first serve to explore their action capabilities. Forty male high school students competed in two volleyball games; a regulation (single serve) game and a modified (2-serve) game. Participants reported that having a second chance at serving allowed them to feel more confident and relaxed which facilitated the exploration of their serving capability. In the 2-serve game, participants attempted a more powerful (M = 60.3 km/hr), and complex (M = 44.5% jump topspin serves) first serve, compared to the regulation game (M = 55.6 km/hr; M = 25.2% jump topspin serves). Findings suggest that to facilitate learning of motor skills, it is important to manipulate the practice environment using task constraints to address the factors that restrict exploration.

KEYWORDS:

• Constraints-led approach
• exploration
• motor learning
• volleyball
• coaching

Introduction

Evidence from motor learning literature suggests that exploration is an important feature for the successful learning of motor skills (J.-Y. Chow et al., 2006, 2009; Hristovski et al., 2012; Komar et al., 2019). Put simply, exploration is the continuous and active process whereby an individual searches for what works and what doesn’t, which, by definition, involves a certain amount of trial and error (E. J. Gibson, 1988; Handford et al., 1997; Komar et al., 2019). During this process, the individual searches the perceptual-motor workspace for new solutions to the task requirements (Pacheco et al., 2019). Searching is the process of learning to attend to specifying information to support the emergence of functional perception-action couplings (i.e., the reciprocal and direct relationship between perceptual information and actions) (Button et al., 2021; Davids et al., 2012). Individuals who explore during practice have been shown to be more successful in ultimately enhancing their performance compared to individuals who do not explore (J.-Y. Chow et al., 2009; Seifert et al., 2015). As individuals explore, they seek customised solutions to specific performance problems, allowing their own unique functional behaviours to emerge (M. C. Y. Lee et al., 2014). Individuals therefore have the potential to benefit when given the freedom to actively explore, discover, and exploit movement solutions. This freedom enhances their capability to successfully interact with the information in the environment amplified by the task constraints (e.g., rules) (Davids & Araújo, 2010; Davids et al., 2015; E. J. Gibson, 1988; Hacques et al., 2021). The exploratory process elicits more variability in movements and/or outcomes, enabling the emergence of more functional movement solutions (Seifert et al., 2015), whilst the individual is also afforded the opportunity to become more adaptable (M. C. Y. Lee et al., 2014; Renshaw et al., 2009): a hallmark of highly skilled performers (Button et al., 2021; Ranganathan et al., 2020; Seifert et al., 2013).

Despite the importance of exploration in learning, motor skills are often taught using isolated practice drills that are characterised by imitation rather than exploration (Moy et al., 2023). In isolated drills, the performer is provided with highly prescriptive and idealistic movement solutions to mimic, rather than being challenged to actively explore the learning environment to discover their own unique and adaptable movement solutions (Renshaw et al., 2010). An alternative sports coaching methodology that is compatible with the facilitation of learning through exploration is the constraints-led approach (CLA) (Renshaw et al., 2019). The distinguishing feature of the CLA is that its learning design is grounded in the contemporary motor learning theory of ecological dynamics, which recognises the interaction between the environment and the individual in the acquisition of motor skills (J.-Y. Chow et al., 2013). Essentially, ecological dynamics is a meta-theory that combines the sub-theories of ecological psychology (J. J. Gibson, 1986) and dynamic systems (Newell, 1986). It proposes that human behaviours are an emergent property of self-organising, nonlinear dynamical movement systems, which are regulated by the continuous interactions between the physical and psychological capacities of the individual, the physical and social environment, and the specific task demands (Davids et al., 2008).

In a practice setting, a constraints-led learning design involves selectively manipulating constraints (e.g., rules, equipment, playing area) that have the potential to allow individuals the freedom to actively explore their movement capabilities (Renshaw et al., 2019). These constraints channel an individual’s exploration within a narrower area (limited number of movement solutions) of the practice environment towards desired functional movement solutions (J.-Y. Chow, 2013; Davids et al., 2005; Komar et al., 2019). Pivotal to the CLA learning design is the detection of affordances or opportunities for action provided to an individual by the environment, dependent on one’s capabilities to exploit the landscape of affordances (Rietveld & Kiverstein, 2014). For example, a space at the back of the volleyball court is an affordance for an individual to serve into if they have the confidence and capability to serve accurately.

Also pivotal to the CLA learning design is the identification of rate limiters (i.e., “handbrakes” that can restrict an individual’s exploration and learning of motor skills) and the manipulation of constraints targeted at reducing their influence (Brymer & Davids, 2014; Correia et al., 2018). A study by Seifert et al. (2014) highlighted how fear was a rate limiter that restricted the perceptions and exploratory actions of beginner ice climbers. A fear of falling resulted in the beginners adopting an “X” position with their arms and legs, which provided a highly stable position on the surface, but which restricted their capability to explore how to climb the ice more efficiently and effectively (Seifert et al., 2014). Similarly, the rate limiter of anxiety (elicited by climbing high on a wall) restricted exploration by rock climbers, resulting in the execution of shorter reaching distances and slower, more rigid, and less efficient movements (Nieuwenhuys et al., 2008; Pijpers et al., 2006).

A potential constraint that may act as a barrier to enhancing performance in volleyball are the rules around serving, that is, that players are only allowed one serve, in contrast to the two serves tennis players are allowed. Anecdotal evidence from volleyball coaches suggests that many junior volleyball players opt for a conservative (e.g., slower) serve to try to ensure that the ball lands within the opponent’s playing area. In this case, the rules that permit only a single serve may be actively discouraging the performer from exploring their action capabilities and taking risks to search for alternative movement solutions due to fear of making a mistake and forfeiting a point (see Ely et al., 2017). Indeed, some experimental studies in tennis have shown that the task constraint of a second serve may encourage players to push the outer boundaries of their co-ordination patterns by exploring riskier (Ely et al., 2017) and faster (J. Chow et al., 2003) first serve attempts. Thus, a possible solution to encourage players to engage in deeper exploration of their serving capabilities under game conditions could be to play a modified volleyball game with a task constraint (rule) that affords players a second chance at serving (see also, J. Chow et al., 2003; Ely et al., 2017). Despite extensive theoretical evidence demonstrating that the manipulation of learning environments using task constraints facilitates the development of movement skills through exploration, there is limited empirical evidence to support this theory in practice (Hacques et al., 2021; Renshaw & Chow, 2019; Renshaw et al., 2019).

The purpose of the current study was to investigate whether the application of the CLA in a practical setting could actively facilitate exploratory learning in a complex and dynamic movement task with high accuracy demands. Specifically, the aim was to examine whether the addition of a second serve during a volleyball game (2-serve game) would encourage high school children to use their first serve to explore their movement capabilities, relative to a regulation game involving a single serve. The task constraint of a second serve was anticipated to facilitate participants’ exploration by reducing the magnitude of rate limiters such as fear, anxiety, the perception of pressure, and/or a lack of confidence (see Nieuwenhuys et al., 2008; Pijpers et al., 2006). It was hypothesised that this exploratory behaviour would manifest in a faster and more qualitatively difficult first serve in the 2-serve game (e.g., a jump topspin serve), compared to the serves typically performed in the regulation game (e.g., an overarm/float serve; see Wise, 2002). However, it was hypothesised that during their exploration of the boundaries of their performance capabilities, the participants would exhibit significantly reduced accuracy for their first serve in the 2-serve game, compared to the accuracy of their serves in the regulation game. It was also hypothesised that the increase in exploratory activity during the 2-serve game would result in a significant increase in serve speed variability compared to that observed in the regulation game.

Methods

Participants

The study consisted of 40 male high school students from the volleyball excellence program at an inner-city school in Australia. The school was well-resourced and offered a variety of sporting excellence programs aimed at providing students with the opportunity to combine their academic studies with their sporting development. The participants were recruited from the three different age categories of the volleyball excellence program which included a junior age group comprising students in years 7 and 8 (n = 15; M age = 13.4 years, SD age = 0.8), a middle age group comprising students in years 9 and 10 (n = 13; M age = 14.8 years, SD age = 0.6), and a senior age group comprising students in years 11 and 12 (n = 12; M age = 16.2 years, SD age = 0.6). The volleyball playing experience of the participants ranged from 1 to 7 years (junior M = 2.4, SD = 0.8; middle M = 3.0, SD = 1.2; senior M = 4.5, SD = 1.4). The participants were of mixed abilities as illustrated by their reported highest level of junior representation in volleyball which included school (n = 26), district (a collection of local schools; n = 4), regional (a collection of districts; n = 3), state (n = 5), and national levels (n = 2). The study received institutional ethical approval from the university’s Human Research Ethics Committee (approval number 2,000,000,299) and all participants and their parents/guardians provided informed written consent.

Procedure

The study adopted a counterbalanced, crossover, repeated measures experimental design. All participants competed in two different volleyball games including (i) a regulation (single serve) volleyball game and (ii) a modified (2-serve) volleyball game where the server had two attempts at serving, similar to the serving rules of tennis. The players in each age group were allocated into two evenly matched teams based upon the head coach’s perception of the players’ ability. The coach was well placed to make these judgements as he had worked with many of the players over several seasons (Note: the data was collected towards the end of the season and all set scores were close). The two volleyball games were played consecutively during the participants’ regular training times, with the two teams in each age group competing against each other. To overcome the order effect, the regulation and 2-serve volleyball games were counterbalanced. The junior and senior age groups completed the 2-serve game first, immediately followed by the regulation game, whereas the middle age group completed the regulation game first, immediately followed by the 2-serve game.

Each game incorporated player numbers, rules, ball sizes, net heights, and scoring methods consistent with the by-laws of the state Volleyball Schools Competition (net heights: junior = 2.15 m, middle = 2.35 m, senior = 2.43 m). All games were officiated by the players’ coaches, and game scores were displayed to players using scoreboards. Apart from the additional serve provided in the 2-serve game, the only new rule introduced was that if the same server won 3 consecutive points, that server was rotated out of the server position. This rule was introduced to provide opportunities for players to perform at least three serves in each game which was the minimal number of serves required for participants’ inclusion in the data analysis. Each game had an approximate duration of 30 to 45 min. All age groups completed two sets of the 2-serve game, but due to the restricted duration of the training sessions, only the middle age group played two sets of the regulation game. The junior and senior age groups played one set of the regulation game. This resulted in the junior age group averaging 3.7 serves for the regulation game and 4.5 serves for the 2-serve game; the middle age group averaging 6 serves for the regulation serve game and 6.6 serves for the 2-serve game; and the senior age group averaging 7.8 serves for the regulation game and 6.7 serves for the 2-serve game.

Quantitative and qualitative data were used to determine whether the addition of the second serve during the volleyball games would encourage the participants to use their first serve to explore their movement capabilities. The quantitative dependent variables were serve speed, serve accuracy, and serve type.

Serve speed

Serve speed was measured using a handheld radar gun (Pocket Radar Ball Coach, model PR1000-BC) which was positioned behind the baseline at the opposite side of the court, directly in line with the server. The speed of each participant’s first serve in the 2-serve game and their only serve in the regulation game were measured in km/hr. The average serve speed for each individual for each game type was then calculated. The standard deviation of serve speed was also collected and analysed to investigate whether players exhibited changes in the amount of variability in their serve speeds across the two game types.

Serve accuracy

Serve accuracy was measured by recording the total number of successful first serves for each participant in the 2-serve game, and the total number of successful serves for each participant in the regulation game, and then dividing those values by the total number of overall serves in the given game type to create a percentage score. A successful serve was defined as a serve that landed in the field of play (i.e., an ace) or that was intercepted by a receiver.

Serve type

Each game was filmed using a standard digital camera (Sony FDR-AX53; 50 fps; 1920 × 1080pixels) which was positioned side-on and slightly behind the server. The recording was later analysed to classify the type of serve participants attempted on their first serve in the 2-serve game and on their only serve in the regulation game. The serves were classified as either an overarm/float serve or a jump topspin serve (for diagrams and further details of the serves, see Dearing, 2019; Pellett & Lox, 2000). The overarm/float serve was defined as a serve where the player threw the ball into the air and struck the ball with the hand, typically with minimal follow-through of the arm, with no (or only slight) spin imparted upon the ball (see Ciuffarella et al., 2013; Costa et al., 2012; Dearing, 2019; Moras et al., 2008; Pellett & Lox, 2000). In the present study, the overarm/float serve may or may not have involved a jump to strike the ball. A jump topspin serve was defined as a serve where the player threw the ball into the air before then jumping to strike the ball with the hand using a powerful overarm action and full follow-through, attempting to place topspin on the ball (see Ciuffarella et al., 2013; Costa et al., 2012; Dearing, 2019; Wise, 2002). In general, the key differences used to distinguish between the two serve types were that the jump topspin serve involved a higher jump, a higher ball toss, and a more powerful and pronounced swing of the arm compared to the overarm/float serve (see Ciuffarella et al., 2013; Costa et al., 2012; Dearing, 2019; Pellett & Lox, 2000). The jump topspin serve is generally considered to be the most difficult type of volleyball serve to execute (Wise, 2002), with a higher risk of error (Moras et al., 2008). To ensure the accuracy of serve classification, a volleyball coach who had 24 years playing experience and 15 years elite coaching experience trained two members of the research team to correctly recognise the different serve types. Once all serves were classified, the percentage of jump topspin serves was calculated for each individual participant for each game type.

Data analysis (quantitative)

The data for serve speed, serve accuracy, and serve type were analysed using 2-way repeated measures analysis of variance (ANOVA) tests with age group (junior, middle, senior) and game type (regulation game, 2-serve game) entered as the between- and within-group variables, respectively. A 2-way ANOVA was also used to analyse the standard deviations from the serve speed data to determine whether the variability in serving speed was different in the 2-serve game compared to the regulation game. Pairwise comparisons for main effects and interactions were conducted using Sidak adjustments. The data for serve type showed deviations from a regulation distribution, as well as a significant Box’s test (p = .001), but transformation of these data was unsuccessful in rectifying these assumptions and so given the robustness of ANOVA (Mood et al., 2020), the untransformed data were used for all analyses. To calculate the required sample size to achieve a power of .80, the repeated measures ANOVA option in G*Power (version 3.1.9.6) was used with a moderate effect size (f = 0.25) and a moderate correlation (r = 0.6) (Faul et al., 2007). The estimated sample size from G*Power was 36, which was within the total sample size of 40 used for the present study. Alpha was set at p < .05 for all statistical analyses.

Data analysis (qualitative)

Immediately after both games were completed, participants were asked to respond in writing to the following two questions aimed at determining their feelings and intentions when serving their first serve in the 2-serve game: “How did you feel when serving your first serve in the 2-serve game?” “What did you do differently (try) on your first serve when you had a second chance at serving?” The participants’ written responses to the two questions were examined using content analysis. This method of analysis was used to check the patterns and trends of words used, and their frequency and relationships (Grbich, 2007). Due to the specific nature of each question, the responses for each question were analysed separately by the lead researcher. This process involved the following 3 steps: (1) repeated reading of the responses to become familiar with the data, (2) line-by-line open coding of the raw data to identify categories (codes) that described the participants’ feelings and behaviours when serving their first serve in the 2-serve game, (3) collating similar codes together into themes, and (4) review and verification of data codes and themes by the second researcher (Hsieh & Shannon, 2005). A similar method of data analysis was adopted in a study that investigated how the presence of a scoreboard influenced athletes’ affective, cognitive, and behavioural responses (Maloney et al., 2022).

Results

Quantitative data

For the serve speed data, the ANOVA revealed a significant main effect for game, F(1, 37) = 30.20, ƞ_p² = .45, p < .001, showing that the combined serve speed for all age groups was faster in the 2-serve game compared to the serve speed in the regulation game. There was also a significant main effect for age group, F(2, 37) = 3.26, ƞ_p² = .15, p = .0495, with pairwise comparisons indicating that the combined serve speed across the regulation and 2-serve games for the junior group (p = .045; M = 53.9 km/hr; SD = 5.9 km/hr) was slower than the serve speed for the senior group (M = 62.4 km/hr; SD = 11.1 km/hr), but not the middle group (p = .44; M = 58.4 km/hr; SD = 10.5 km/hr). There were no significant differences in serve speed between the middle and senior groups (p = .58). The interaction between game and age group was not significant, F(2, 37) = 2.76, ƞ_p² = .13, p < .077, highlighting that the increase in game speed in the 2-serve game was not isolated to a particular age group. For the analyses of the variability of the serve speed data, the ANOVA revealed no significant main effect for game, F(1, 37) = 0.004, ƞ_p² < .001, p = .95, or age group, F(2, 37) = 1.25, ƞ_p² = .06, p = .30, and there was no significant interaction between game and age group, F(2, 37) = 0.05, ƞ_p² = .003, p = .95.

For the analyses of the serve accuracy data, the ANOVA revealed a significant main effect for game, F(1, 37) = 4.54, ƞ_p² = < .11, p = .04, showing that the combined accuracy for all age groups in the 2-serve game was lower than that observed in the regulation game. The main effect for age group was not significant, F(2, 37) = 0.63, ƞ_p² = .03, p = .95, showing that there were no significant differences in serve accuracy across the three groups. The interaction between game and age group was also not significant, F(2, 37) = 0.78, ƞ_p² = .04, p = .47, suggesting that the decline in serve accuracy observed in the 2-serve game was similar across the three age groups.

For the analyses of serve type, the ANOVA showed a significant main effect for game, F(1, 37) = 19.79, ƞ_p² = .35, p < .001, revealing that the percentage of jump topspin serves used in the 2-serve game was greater than the percentage of jump topspin serves used in the regulation game (Table 1). The main effect for age group was not significant, F(2, 37) = 2.72, ƞ_p² = .13, p = .08, but there was a significant interaction between game and age group, F(2, 37) = 15.22, ƞ_p² = .45, p < .001. Follow-up tests showed a significant increase in the percentage of jump topspin serves for the middle age group in the 2-serve game compared to the regulation game (p < .001), but not for the junior (p = .86) or senior groups (p = .66).

Table 1. Serve data across the two game types for each age group.

Age group	Serve speed (km/hr)		Serve accuracy (%)		Jump topspin serves (%)
	Regulation game	2-serve game	Regulation game	2-serve game	Regulation game	2-serve game
Junior	52.4 (5.0)	55.4 (6.5)	79.9 (21.2)	68.6 (24.1)	16.7 (36.2)	18.0 (31.9)
Middle	54.6 (9.4)	62.2 (10.4)	86.4 (19.1)	68.7 (20.9)	23.1 (39.4)	77.7 (38.3)
Senior	60.6 (10.2)	64.3 (12.2)	72.0 (22.7)	69.7 (20.2)	38.1 (48.5)	41.7 (51.5)
Combined	55.6 (8.8)	60.3 (10.3)	79.7 (21.3)	69.0 (21.4)	25.2 (41.1)	44.5 (47.0)

Values in parentheses show standard deviation. “Combined” is the combined mean of all data across all three age groups. The data includes participants’ first serve in the 2-serve game, and their only serve in the regulation game.

Qualitative data

Overcoming rate limiters

When analysing participants’ responses to the question “How did you feel when serving your first serve in the 2-serve game?”, one prominent theme of overcoming rate limiters emerged. Specifically, the addition of a task constraint in the form of a second serve helped participants to overcome the emotional or cognitive factors that were restricting the exploration of their serving capabilities. This theme was consistent across all three age groups with 32 of the 40 participants (80%) reporting that having a second serve either boosted their confidence (n = 14, 35%) or alleviated their feelings of pressure and anxiety (n = 14, 35%), or both (n = 4, 10%). For example, participants reported feeling more relaxed and confident when they had a second chance at serving: “I felt nervous that I would miss, but after I knew I had a second chance, I was more relaxed and chilled”. (Junior participant). “It felt like there was less pressure and it boosted my confidence knowing that I had a second chance”. (Middle participant).

Intentions to explore

An analysis of participants’ responses to the question “What did you do differently (try) on your first serve when you had a second chance at serving?”, revealed one prominent theme of intentions to explore. This theme was consistent across all three age groups. Specifically, the opportunity to use a second serve in the 2-serve game encouraged 33 of the 40 (82.5%) participants to explore their serving capabilities. Participants reported trying a more aggressive or powerful first serve (n = 15, 37.5%), a more complex/riskier first serve such as a jump topspin serve (n = 3, 7.5%), aiming their serve at a specific target such as the opposing setter or the back corner of their opponent’s court (n = 3, 7.5%), or exploring a combination of the aforementioned changes and other aspects of their serve including ball toss height, stance, and jump (n = 12, 30%). Overall, when participants’ responses to both questions were interpreted in combination, the reduction in the influence of rate limiters elicited by the provision of a second serve in the 2-serve game acted as an invitation to participants to engage in more exploratory behaviours relating to their serve as exemplified by the following quotes: “On my first serve, I tried a jump topspin serve and on the second serve I did a jump float which is my normal serve”. (Middle participant). “I was contacting the ball with more force than normal because I had the reassurance of a second serve.” (Senior participant).

Discussion

The purpose of this study was to determine whether the application of the CLA would actively facilitate exploratory learning in a complex and dynamic movement task with high accuracy demands. Specifically, the research investigated whether the addition of a task constraint in the form of a second serve during a volleyball game would reduce the influence of rate limiters and subsequently encourage high school players to overcome barriers to exploring their movement capabilities, relative to a regulation game involving a single serve. In line with our expectations, compared to their serve in the regulation game, participants across all age groups attempted a more complex (i.e., jump topspin) and faster first serve in the 2-serve game, which was often targeted at the setter or the back corner of the court. However, as predicted, these serves were not as accurate (successful) when compared to participants’ serves in the regulation game. The greater exploration can be explained by the fact that the participants across all age groups reported feeling more confident, calm, and relaxed which changed their intentions (e.g., cognitions in the form of thoughts relating to potential serving strategies or different options to explore), and facilitated the exploration of their serving capability.

The results provide further empirical evidence of the efficacy of the CLA in facilitating the development of movement skills by searching for new and more functional solutions through exploration (e.g., Komar et al., 2019). Critical to an individual’s learning within a CLA is the manipulation of constraints in the practice environment to help reduce the influence of rate limiters on exploration (Correia et al., 2018). Previous studies have shown the extent to which rate limiters such as anxiety can restrict exploration (Nieuwenhuys et al., 2008; Pijpers et al., 2006; see also Seifert et al., 2014; Seifert & Davids, 2012; Seifert et al., 2015). For example, when novice climbers perform under higher levels of anxiety, they tend to narrow their perception of affordances towards the use of more proximal holds, rather than exploring holds at the outer limits of their reach (Nieuwenhuys et al., 2008; Pijpers et al., 2006). The results of the present study demonstrated how rate limiters in a team sport task were attenuated by selectively manipulating key task constraints. Specifically, the addition of a second serve in a volleyball game provided an affordance that led to the exploration of a broader repertoire of serve types and serve strategies. As reported by the players, the addition of a second serve changed their intentions, prompting them to attempt more strategically challenging serves (e.g., serving at a particular target), as well as encouraging them to attempt to hit the ball with greater power and/or perform a more complex jump topspin serve. Collectively, this manifested as an 8.5% increase in serve speed and a 76.6% increase in the number of more complex jump topspin serves that were attempted (although as discussed later, this latter result was largely due to the middle age group). These results align with previous research in tennis where two serve attempts were associated with players executing a riskier (Ely et al., 2017) and faster (J. Chow et al., 2003) first serve.

While exploration has been postulated as facilitating learning, it has also been associated with reduced performance success (e.g., Seifert et al., 2015; see also Handford et al., 1997; Komar et al., 2019) and increased levels of variability (Seifert et al., 2015). The hypothesised increase in serve speed variability did not occur, suggesting that although participants increased the speed of their serves in the 2-serve game, they did not explore variations in their serve speed across the two game conditions. As hypothesised, the serve accuracy of participants in the present study significantly decreased from 79.7% in the regulation game, to 69.0% on their first serve in the 2-serve game. Given that exploration involves the search for new movement solutions in the perceptual-motor workspace (McDonald et al., 1995; Ranganathan et al., 2020), it is likely that both successful and unsuccessful solutions will emerge during the exploratory process (Handford et al., 1997; Komar et al., 2019; Seifert et al., 2015). For example, when intermediate rock climbers were encouraged to engage in increased exploration, they exhibited longer ascent times and had lower levels of climbing fluency (i.e., higher entropy) (Seifert et al., 2015). This result, and those of the present study, further highlights the need for practitioners to accept that learning through exploration will inevitably involve the emergence of unsuccessful solutions, especially during the early stages, but with further practice, exploration can help lead to enhanced performance outcomes (Komar et al., 2019; Seifert et al., 2015).

Comparisons between the junior, middle, and senior age groups for the regulation game and 2-serve game for serve speed, serve speed variability, and serve accuracy revealed no significant differences between the groups. However, for serve type, the middle age group performed significantly more jump topspin serves in the 2-serve game (77.7%) compared to the regulation game (23.1%), whereas the junior (18.0% and 16.7%, respectively) and senior (41.7% and 38.1%, respectively) age groups showed no significant differences in serve type between the two game conditions. It seems that the middle age group was especially sensitive to the task constraints manipulated in this study, and these participants elected to attempt a qualitatively different and functionally more challenging serve. In contrast, while many of the junior participants reported attempting a more complex serve, this younger group may have lacked the skill and/or physical capability to perform a qualitatively different serve type in the form of the jump topspin serve, particularly given that this serve typically requires a powerful strike of the ball and a high level of coordination (see Ciuffarella et al., 2013; Costa et al., 2012; Dearing, 2019; Wise, 2002). This result highlights the individual nature of affordances (see E. J. Gibson, 1988; Withagen et al., 2012), and the need for coaches to consider individual differences in the intentions of each player. That is, while an affordance may be present, the performer must have the necessary action capabilities and intention to utilise that affordance (Rietveld & Kiverstein, 2014; Withagen et al., 2012). The manipulation of constraints to provide affordances for learning must therefore be based upon an understanding of the current action capabilities and motivations of the individual (Renshaw & Moy, 2018; Renshaw et al., 2019).

In contrast to the junior age group, while the senior age group may have had the necessary physical attributes (e.g., limb length, leg power) to perform the jump topspin serve, it is likely that their extensive experience in volleyball had created a relatively stable co-ordination pattern for their serve. This stable state likely encouraged them to maintain their existing serve type, rather than to explore a qualitatively different serve in the 2-serve game (see Button et al., 2021; Kugler et al., 1990; T. D. Lee et al., 1995; Ranganathan et al., 2020). This conclusion is aligned with the concept of behavioural flexibility outlined by Ranganathan et al. (2020) which refers to the capability of the performer to successfully complete a given task by varying their movement pattern (e.g., serving a volleyball at different speeds) and/or by qualitatively changing to a different movement pattern (e.g., switching from a float serve to a jump topspin serve). Ranganathan et al. (2020) proposed that as an individual increases in skill level, they may exhibit less flexibility because they have started to reach the upper limits of their available solution space. They also explained how attempts to change a highly stable movement pattern are typically accompanied by reductions in performance (see also T. D. Lee et al., 1995). These explanations suggest that the senior participants in the present study may have avoided exploring a different serve type due to the challenges and potential risks (e.g., losing a point) involved in attempting to vary and/or qualitatively change an already stable motor solution (Ranganathan et al., 2020; see also Button et al., 2021; T. D. Lee et al., 1995).

This research has important implications for practitioners and sports administrators by highlighting how practice activities can be designed, and junior sport rules modified, to reduce the influence of rate limiters and encourage individuals to take risks and explore their capabilities. The 2-serve game can be used in training and competition to develop players’ confidence and capability to serve more aggressively in a more relaxed, but highly representative environment. Over time, participation in the 2-serve game may accelerate players’ use of a riskier and aggressive serve in a regulation (single serve) game. This would have a significant impact on both an individual’s and a team’s performance, as “serve points” and “serve reception errors” are key variables that best discriminate between winning and losing in competitive volleyball matches (Paulo et al., 2018; Silva et al., 2014). Additionally, modifying game rules such as adding a second serve has the potential to enhance player’s all round skills. For example, allowing players two serves may expose receivers to a greater number and variety of faster serves, giving them the opportunity to improve their reception skills in preparation for competition (Afonso et al., 2009).

It is recommended that future research should examine whether the addition of a second serve in a volleyball practice task encourages similar levels of exploration in more advanced performers when the 2-serve practice task is used over a prolonged period (e.g., several weeks or months). Moreover, while the inclusion of a second serve in the present study encouraged performers to explore their serving capabilities within a single training session, future research should implement this task constraint over several training sessions to examine the extent to which this exploratory behaviour is maintained (Komar et al., 2019; see also Hossner, Kӓch, & Enz 2016). Given that the majority of research in sport and exercise science is conducted using male participants, it is important to conduct similar research with females to determine whether the results of the present study are equally applicable across both sexes (Cowley et al., 2021).

Conclusion

The addition of a second serve in a volleyball game reduced the influence of rate limiters related to emotional or cognitive factors, prompting performers to explore their serving capabilities by attempting a faster and more complex serve. However, this exploration was associated with reductions in serving accuracy, highlighting the need for practitioners to accept that exploration will inevitably lead to the emergence of unsuccessful solutions as performers search the perceptual-motor workspace to find a coordination pattern that achieves the desired outcome (Handford et al., 1997; Komar et al., 2019; Seifert et al., 2015). The results further demonstrate the benefits of the constraints-led approach (e.g., Komar et al., 2019) and highlight how simple changes to practice tasks can encourage increased exploration.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.