Skip to main content
NCBI home page
European Journal of Sport Science logoLink to European Journal of Sport Science
. 2024 Jun 14;24(8):1056–1066. doi: 10.1002/ejsc.12153

Validity of a motor–cognitive dual‐task agility test in elite youth football players

David Friebe 1,✉, Johanna Sieland 2, Hendrik Both 1, Florian Giesche 1, Christian Haser 2, Thorben Hülsdünker 3,4, Florian Pfab 2, Lutz Vogt 5, Winfried Banzer 1,2
PMCID: PMC11295090  PMID: 38874591

Abstract

Agility, as the ability to react rapidly to unforeseen events, is an essential component of football performance. However, existing agility diagnostics often do not reflect the complex motor–cognitive interaction required on the field. Therefore, this study evaluates the criterion and ecological validity of a newly developed motor–cognitive dual‐task agility approach in elite youth football players and compare it to a traditional reactive agility test. Twenty‐one male youth elite football players (age:17.4 ±0 .6; BMI:23.2 ± 1.8) performed two agility tests (reactive agility, reactive agility with integrated multiple‐object‐tracking (Dual‐Task Agility)) on the SKILLCOURT system. Performance was correlated to motor (sprint, jump), cognitive (executive functions, attention, reaction speed) and football specific tests (Loughborough soccer passing test (LSPT)) as well as indirect game metrics (coaches' rating, playing time). Reactive agility performance showed moderate correlations to attention and choice reaction times (r = 0.48−0.63), as well as to the LSPT (r = 0.51). The dual‐task agility test revealed moderate relationships with attention and reaction speed (r = 0.47−0.58), executive functions (r = 0.45−0.63), as well as the game metrics (r = 0.51−0.61). Finally, the dual‐task agility test significantly differentiated players based on their coaches' rating and playing time using a median split (p < 0.05; d = 0.8–1.28). Motor–cognitive agility performance in elite youth football players seems to be primarily determined by cognitive functions. The integration of multiple object tracking into reactive agility testing seems to be an ecologically valid approach for performance diagnostics in youth football.

Keywords: athlete, cognition, multi‐tasking, soccer, talent

Highlights

  • The study introduces a novel motor–cognitive dual‐task agility approach (incorporation of multiple‐object‐tracking in agility testing), evaluating its criterion and ecological validity in elite youth football players compared to a standard agility test.

  • The standard agility test was shown to have moderate correlations with attention and choice reaction times, while the dual‐task agility approach additionally incorporates executive functions

  • While the agility test correlates to football‐specific test performance, the dual‐task agility test significantly discriminates players based on their potential ratings and in‐season playing time, highlighting its potential as a valuable tool for assessing performance in youth football.

  • The findings suggest that agility performance in elite youth football is primarily determined by cognitive functions

  • Incorporating more complex cognitive elements such as multiple‐object‐tracking in agility testing may improve ecological validity and therefore the predictive value of the testing procedure.

1. INTRODUCTION

In football, athletes move in a highly variable and constantly changing environment. In order to play at a high/professional level, it is crucial for athletes to be able to perceive, filter, and process a variety of external stimuli (e.g., tracking teammates, opponents, ball) (1–3). The ability to adjust their movement plans and actions rapidly to these unexpected changes on the field is defined as reactive agility (Young & Willey, 2010). To successfully cope with these complex demands and to achieve a high playing performance level, an optimal interplay between motor (e.g., cutting, sprinting) (Haugen et al., 2014) and cognitive abilities, such as reaction speed as well as executive functions (working memory, cognitive flexibility, response inhibition (Scharfen & Memmert, 2019; Vestberg et al., 2012; Vestberg et al., 2017); is required.

Contrarily, previous research studies and practical application often assessed motor and cognitive components separately, using sprint, jump or change of direction tests (without a reactive decision making component) for motor performance (Fiorilli et al., 2017; Haugen et al., 2014) and computer‐based assessments such as reaction speed, executive function tests or multiple object tracking for cognitive performance (Ehmann et al., 2021; Faubert & Sidebottom, 2012; Vestberg et al., 2017). Although executive functions have been shown to correlate to sport performance and success in youth football (Huijgen et al., 2015; Vestberg et al., 2012, 2017), results are heterogeneous and the ecological validity of isolated cognitive and motor assessments have repeatedly been questioned by several research groups (Pojskic et al., 2018; Vater et al., 2021; Young et al., 2015). In this context, ecological validity, that is, the generalizability of a test result to the real‐world setting (Schmuckler, 2001), is primarily based on stimulus correspondence and task correspondence between the test and game demands. Common motor tests lack stimulus correspondence as sensory stimuli are either absent (jump, change of direction) or present light stimuli (agility), not considering the high demands on motion perception especially in ball and team sports. Conversely, cognitive assessments performed in front of a computer screen have low task correspondence (Ehmann et al., 2021; Scharfen & Memmert, 2021a; Vater et al., 2021) as motor tasks are often limited to a button press which does not reflect dynamic full‐body movements in a match. Therefore, the combination of sports‐relevant cognitive and motor tasks in order to increase both stimulus and task correspondence has been proposed by many researchers to improve the predictive quality of performance assessments for on‐court performance (Broadbent et al., 2015; Ehmann et al., 2021; Romeas et al., 2019; Scharfen & Memmert, 2019) described this as perception–action coupling.

Previous research studies addressed perception–action coupling by using reactive agility tests that require rapid movements in response to a visual light stimulus (Trajković et al., 2020; Young & Willey, 2010). Indeed, initial evidence suggests that approaches involving athletic movements with time‐constrained decision‐making are more effective in discriminating between levels of performance than traditional change of direction tests (Trajković et al., 2020). However, the contrast character (on/off) of the visual stimulus only provides low stimulus correspondence as visual stimuli in ball sports are not stationary but moving, and there is typically more than one stimulus to be tracked (teammates, opponents, ball). Accordingly, the motor–cognitive demands of complex and dynamic game situations in ball sports require players to constantly track, process, and respond rapidly to multidimensional visual stimuli under dual‐tasking conditions (e.g., offensive dribbling with the ball, defensive actions against several players, looking for free teammates). The importance of combining multiple object tracking and agility has previously been highlighted by Scharfen & Memmert, 2021a but is not reflected in current performance tests.

The aim of the present study was to develop a novel motor–cognitive dual‐task agility approach that integrates higher cognitive demands (multiple object tracking) into reactive agility testing. We assume that by combining reactive agility with tracking stimuli, the test better reflects football specific requirements (perception–action coupling) and might therefore outperform current reactive agility tests in identifying players with higher performance potential. Criterion‐related and ecological validity were evaluated among youth elite football players and compared to traditional reactive agility test.

To determine the criterion validity, we related the individual agility performances (reactive agility, reactive agility with multiple object tracking (Dual‐Task Agility, DT‐Agility)) to the potentially underlying isolated motor (sprint, jump) and cognitive functions (attention, choice reaction speed and executive functions). To determine the ecological validity, the test performances were correlated with an established football test (Loughborough soccer passing test (LSPT) (Le Moal et al., 2014);) as well as indirect match performance measures (i.e., seasonal playing time, expert assessment of the players' potential to perform on professional level in future) (Kelly et al., 2021).

Firstly, we hypothesized that the conventional reactive agility test would relate to basic motor tests as well as reaction speed and attentional processes. Secondly, we assumed that the DT‐agility test, due to its higher cognitive demands and test complexity, does not relate to motor functions but additionally requires executive functions such as working memory and cognitive flexibility. Finally, we assumed that the DT‐agility test would provide the best prediction of football‐specific test performance and game performance measures (playing time and experts player’ potential ratings) due to its likely more comprehensive and ecologically valid characteristics.

2. METHODS

2.1. Participants and ethics

We recruited male youth football players competing at the highest national level from the U19‐team of an elite football academy in Germany. Required sample size was calculated via G°Power (Version 3.1.9.2; Germany). Based on the effect size of ŋ 2 = 0.45 reported by Trajković et al., 2020, for the discriminative value of reactive agility testing, with an alpha error probability of 0.05 and a power of 0.95, a required minimal sample size of 16 participants was determined.

Participants were included when being field players with a year of birth between 2004 and 2006 and having played at least one match in the highest German youth national league within the current season. Players were excluded from the study when suffering from any acute or chronic performance‐impairing disease or musculoskeletal injury. Further exclusion criteria were surgery of the lower extremities in the last 3 years or head injuries (e.g., concussion) in the last 6 months. All participants were instructed to omit intensive sports activities within the last 24 h before the test day. Before the test day, the players were informed about the experimental protocol and a written consent form (signed by legal guarding if under 18 years old) was obtained. The study was approved by the local Ethics Committee of (reference number: 2021–60). We registered the study in the Clinical Trial Registry (ID: DRKS00027157). The investigation was conducted according to the ethical standards set by the Declaration of Helsinki (World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects, 2013).

2.2. Experimental setup

The motor–cognitive agility tests were performed on the SKILLCOURT system (5 × 5 m; movement concepts GmbH, Schweinfurt, Germany), which provides a highly reliable assessment of agility performances (Friebe et al., 2023). A detailed description of the SKILLCOURT can be found in Friebe et al. (2023). Agility performance was correlated to established isolated motor (sprint, jump) and computer‐based cognition tests (choice reaction, visual search, working memory, cognitive flexibility), as well as a football‐specific test (Loughborough Soccer Passing Test (LSPT (Le Moal et al., 2014)), coaches' player rating, and on‐field playing times. The data collection took place between June 28 and July 4, 2022. The observation period covered the U‐19 German Bundesliga season 2022/2023.

The test order was determined via block randomization across all participants. In order to avoid learning effects within the SKILLCOURT tests (Friebe et al., 2023), a familiarization session (instructions and three training trials for each test) was preceded 1 hour before the regular testing. Before the familiarization and the testing sessions, players performed a standardized warm‐up consisting of 3 min of running and 3 min of progressive speed runs with change of direction components. Test sessions were conducted at the same time of the day for all players to counteract possible effects on motor and cognitive performance.

2.3. Motor–cognitive agility test battery

2.3.1. Reactive agility

Reactive agility performance was assessed with the Random Star Run (Friebe et al., 2023) (Figure 1A). Players started in an active stance in the center field of the court. A representation of the SKILLCOURT court with eight outer target fields was displayed on the screen. After a 3 s countdown, one of the eight target fields was highlighted in yellow color. The participant had to run to the corresponding field and return to the center before the next target field was indicated. Trials consisted of eight runs (each target field once), in a randomized sequence. Thus, the runs and cuttings were performed in reaction to a visual stimulus and could not be preplanned. Participants were instructed to finish the trials as fast as possible. Players performed one test attempt, followed by three rated trials. The performance was operationalized by the total time needed (seconds; [s]), while the best trial was used for statistical analysis. A seated rest period of 90 s was held between trials.

FIGURE 1.

FIGURE 1

Open in a new tab

Exemplary illustration of the test contents of the motor–cognitive agility test battery (A): Reactive agility test; (B–D): Procedure of the dual‐task agility test).

2.3.2. Dual‐task agility

The DT‐agility test incorporates multiple object tracking (Faubert & Sidebottom, 2012), demanding several football‐relevant cognitive functions (e.g., processing speed, attention, working memory (Ehmann et al., 2021; Harris et al., 2020); into a reactive Random Star Run agility test (Friebe et al., 2023). This approach follows the proposal of Scharfen and Memmert (2021a) suggesting to integrate multiple object tracking into a sport‐specific task in a dual‐tasking setting.

The procedure of the DT‐agility test is presented in Figure 1B–D. Participants started in the center field while six white balls (foreground) and a representation of the courts fields (background) were displayed on the screen. At the start of each round 2 of the balls were highlighted in blue for 3 s (Figure 1B). Afterwards, the two balls turned white and all of them started moving in a three‐dimensional space, according to the 3D‐multiple object tracking test by Faubert and Sidebottom (2012). Simultaneously, one of five target fields on the court (3 top and 2 middle fields) was highlighted in yellow color, indicating the player to run to the corresponding field (Figure 1C). After reaching the field, the next field lit up (randomized order). After 10 s, the balls stopped moving and each ball and field on the court was assigned a number (Figure 1D). To select the balls that were initially displayed in blue, the player had to activate the corresponding field. If both balls were identified correctly, the movement speed of the balls increased by 20% relative to the base speed. In case of a failed trial, the speed was decreased by the same factor. The test consists of ten successive 10‐s rounds. Participants were instructed to reach as many fields as possible (minimum 3 fields) without losing the balls. Performance was operationalized by maximum movement speed, number of reached fields and highest dual‐task score, calculated based on Equation (1) (EQ1):

Pointsore=(1+numberofidentifiedballs)∗trackingspeed+reachedfields. (1)

2.3.3. Cognitive performance

The open‐access online database PsyToolkit was used for computer‐based cognitive assessment and data collection (Stoet, 2010). Tests were presented on a 17‐inch screen and responses were made by keyboard and mouse click. The cognitive test battery included a choice reaction task with four choices (choice reaction speed), a visual search task (visual attention), a corsi block‐tapping test (working memory), as well as a digit/number switching task (cognitive flexibility). The tests were selected based on their relevance to football‐specific performance and their correspondence to the cognitive components of the reactive agility tests (Harris et al., 2020; Scharfen & Memmert, 2021b; Vestberg et al., 2012, 2017; Young et al., 2015).

Within the choice reaction task (Deary et al., 2011) the participants had to react as fast as possible to a “X” symbol appearing in one of four white boxes. Each of the boxes was assigned to a key (“Y”; ”X”; ”comma”; ”full stop”). After ten training trials, the test consisted of 40 reactions, from which the mean reaction time (milliseconds; [ms]) was calculated.

In the visual search attention task (Treisman & Gelade, 1980), participants had to identify a target stimulus (orange “T”) out of several distractor stimuli (white “T”, white upside‐down “T”, orange upside‐down “T”) as fast as possible by pressing the “space” key. Forty successive trials were presented with the stimuli sets ranging between 5 and 20. Half of the sets included no target stimuli to control for false alarms. Mean reaction time (ms) was calculated.

The corsi block‐tapping task (Kessels et al., 2000) was conducted to assess working memory. Within the test nine boxes were displayed on the screen. The boxes lit up in a randomized sequence (starting with 2). After the presentation of the sequence, the participants had to recall the boxes in the same order. If correct, the sequence length increased by one. The test ended when a participant failed two consecutive trials. Performance was operationalized by the maximum sequence length (absolute number) correctly selected.

In order to assess cognitive flexibility, a letter/number switching task was conducted (Vandierendonck et al., 2010). A series of 40 letter/number combination (e.g., E5) was presented and players had to react by either pressing “b” for both consonants and odd numbers or “n” for vowels and even numbers. The target stimulus (letter or number) switched every two trials. Mean reaction time (ms) for repetitive (e.g., number, number) and switching (e.g., number, letter) sequences were calculated.

2.3.4. Motor performance

The motor test battery included 5 and 10 m sprint times (s) as well as CMJ‐height (in centimeter; [cm]). Sprinting times (standing starts) and CMJ flight time (jump height calculated accordingly) were assessed via photoelectric cell system (Optojump, Microgate, Bolzano, Italy). The players performed three trials of each test with a 2‐min break in‐between. The best result of each test was used for further analysis.

2.3.5. Football‐specific performance

Football‐specific skills were assessed using the Loughborough soccer passing test (LSPT (Le Moal et al., 2014)) as well as a potential rating by two team coaches (Kelly et al., 2021) and the playing time of the current season.

The LSPT is a valid test to assess different football‐related abilities, including passing, agility, ball control, and decision‐making under time pressure (Le Moal et al., 2014). Here, participants react in response to an unanticipated auditory stimulus (color of the target) by passing the ball to one of four colored rebound boards. As soon as the ball bounced back to the player, the next color was called out, forcing the player to align themselves in one of the four directions in order to pass the ball to the corresponding target as fast as possible. In total, every trial consisted of 16 successive passes. Primary outcomes were test completion time and the penalty time for errors (e.g., imprecise passing, touching cones). For further explanation of the test procedure, please refer to Le Moal et al. (Le Moal et al., 2014). The test was performed three times (plus one training trial), with the best trial used for further analyses.

In order to operationalize game‐related performance of the players, a coaches rating and the playing time of the current season were used. Two coaches (head coach, video analyst) independently rated the player's potential to play prospectively on a professional level. This approach seems to be valid and reliable in predicting the chances of players to play in a professional league in the following 5 years (Jokuschies et al., 2017; Sieghartsleitner et al., 2019). Potential was rated on a scale one to six (1 = very low to 6 = very high potential). The rating score of both coaches correlated significantly (r s  = 0.61: p = 0.003). The mean rating score of both coaches was used for further analysis. The playing time of the current season (time played/total match times, [%]; U19 German Bundesliga Season 2022/23) was extracted from the online data‐platform of the DFB (Deutscher Fußball Bund; www.fußball.de), including all league and cup games. Players with downtimes of more than 28 days due to illness or serve injuries (Waldén et al., 2023) were not included in the correlation analysis.

2.4. Statistical analysis

Following an initial range data plausibility check, Shapiro–Wilk test for normal distribution and Levene test for variance homogeneity were conducted. To determine the criterion validity of the agility tests, performance was correlated to the cognitive and motor assessments using Pearson product–moment correlation coefficient (r p ) or Spearman's rank correlation coefficient (r s). To check for ecological validity, correlations were calculated between the motor–cognitive agility tests and the football specific test metrics. t‐test for independent samples or Mann–Whitney U tests were applied to identify if the agility test battery discriminates between groups divided based on the median value of coaches' potential rating and playing time. To determine sensitivity and specificity, a ROC‐analysis was conducted for tests that significantly discriminate between performance levels.

The alpha‐error was set at p < 0.05. Correlation were considered negligible (>0.1), weak (0.10−0.39), moderate (0.40−0.69), strong (0.70−0.89), and very strong (<0.90) according to Schober et al. (2018). Effect sizes were defined as small (d ≤ 0.4), medium (0.4 < d > 0.8) and large (d ≥ 0.8) (Cohen, 1988). All statistical analyses were performed using SPSS 28 (SPSS Inc., IBM).

3. RESULTS

3.1. Descriptive data

Twenty‐three youth elite players participating in the highest German national youth league were recruited to participate in the study. Two of the players were excluded as they left the club within the study period. Therefore, 21 participants (age: 17.4 ± 0.59 years; height: 178 ± 6.9 cm; weight: 73.8 ± 9.3 kg; body mass index: 23.2 ± 1.8 kg/m2; training load: 10 h/week) were included in the further analysis. Table 1 provides an overview of the performed tests and the descriptive data (Mean; 95% CI) of their outcomes.

TABLE 1.

Overview of the conducted tests, tested abilities and the descriptive data.

Test (measure) Addressed abilities Descriptive data (mean; 95% CI)
Agility
Random star run (s) Reactive agility 15.9 (15.5–16.2)
Dual‐task‐agility (log/abs. Number/score) Agility with complex cognitive demands Tracking speed: 1.3 (1.1–1.6)
Fields: 38.6 (35.3–41.8)
Score: 8.4 (7.6–9.2)
Cognition
Choice reaction (ms) Choice reaction time 412 (384–441)
Visual search (ms) Attentional processes 910 (864 −956)
Corsi task (abs. Number) Working memory 5.4 (4.9–5.8)
Switching task (ms) Cognitive flexibility Repeat: 1000 (878 −1130)
Switching: 1406 (1275–1538)
Motor
Sprint (s) Speed, explosive strength 5 m: 1.0 (0.99–1.1)
10 m: 1.8 (1.7–1.8)
CMJ (cm) Explosive strength 39 (37.1–40.9)
Football related
Loughborough soccer passing test (s) Decision making, passing, ball control Time: 36.2 (35.1–37.3)
Penalties: 5.6 (3.7–7.4)
Playing time (%) In‐season performance 42.7 (30.1–55.2)
Potential rating Success potential 3.1 (2.8–3.5)
Open in a new tab

Abbreviations: CI, Confidence Interval; cm, centimeter; ms, milliseconds; s, seconds.

3.2. Criterion validity

We found moderate correlations between the Random Star Run (reactive agility) and choice reaction time (r s  = 0.52; p = 0.017), visual search speed (r p  = 0.48; p = 0.027), and repetitive reactions in the switching task (r s  = 0.63; p = 0.002). No significant correlations could be found between the reactive agility test and switching speed, visual span and any motor performance (p > 0.05).

The maximal multiple object tracking speed in the DT‐agility test was moderately correlated with visual search speed (r p  = −0.48; p = 0.041) and corsi task performance (r s  = 0.45; p = 0.041). In addition, the number of fields players reached within the 10‐s trials during the DT‐agility test was significantly correlated with choice reaction time (r s  = −0.58; p = 0.006), visual search speed (r p  = −0.47; p = 0.032), as well as repetitive (r s  = −0.53; p = 0.003) and switching reactions (r s  = −0.63; p = 0.002) of the switching task. The highest dual‐task score showed significant correlations with choice reaction time (r s  = −0.51; p = 0.019), visual search speed (r s  = −0.582; p = 0.006), and the corsi task performance (r s  = 0.5; p = 0.021). No significant relationships between the number of fields and any motor test were identified (p > 0.05). Finally, we found no significant correlations between the performances within the two agility tests (p > 0.05).

3.3. Ecological validity

Test time in the LSPT revealed moderate correlations with the Random Star Run (r s  = 0.51; p = 0.02). Penalty time within the LSPT showed no correlations with the agility tests (p > 0.05).

Measures associated with playing performance were significantly related to the performance in the DT‐agility test: the coaches' potential ratings were moderately related with the maximal tracking speed (r s  = 0.51; p = 0.02) as well as the maximal dual‐task score (r s  = 0.54; p = 0.01). Likewise, the playing time showed significant relationships with the maximal tracking speed (r p  = 0.61; p < 0.01) and the maximal dual‐task score (r s  = 0.61; p < 0.01). Scatterplots of the correlations between the DT‐agility test and game metrics are displayed in Figure 2.

FIGURE 2.

FIGURE 2

Open in a new tab

Scatterplots of the correlations between the Dual Task agility performance and indirect game metrics.

The high potential group (median split value = 3) had a significantly higher maximal tracking speed (mean difference: 0.59 (0.16–1.0); p < 0.01; T = 2.89; d = 1.26) and maximal dual‐task score (mean difference: 1.95 (0.59–3.3); p = 0.01; U = 20.0; d = 1.28) than the low potential group. Likewise, players with more playing time (median split value = 40%) had a significantly higher maximal tracking speed (mean difference: 0.54 (0.15–0.94); p = 0.01; T = 2.88; d = 0.8) and test scores in the DT‐agility test (mean difference: 1.3 (−0.05–2.73); p = 0.026; U = 17.5; d = 1.18).

The results of the ROC‐analysis can be seen in Table 2.

TABLE 2.

ROC‐analysis of the tracking speed and score within the dual‐task agility test; classification between players regarding potential rating and playing times.

Tests AUC (95%‐CI) Cut‐off value Sensitivity in % Specificity in %
Potential rating
Tracking speed 0.81 (0.61–1) 1.3 80 72.7
Score 0.82 (0.62–1) 9.9 70 100
Playing time
Tracking speed 0.83 (0.64–1) 1.1 80 66.7
Score 0.83 (0.65–1) 7.1 80 66.7
Open in a new tab

Abbreviations: AUC, Area under the curve; CI, Confidence interval.

Both groups did not differ significantly in terms of reactive agility performance or any other cognitive measures (p > 0.05).

4. DISCUSSION

This study evaluated the criterion and ecological validity of two motor–cognitive agility tests on the SKILLCOURT system in youth elite football players. The traditional reactive agility test was significantly associated to attention and choice reaction speed but not with motor performance. Thus our first hypothesis was partially confirmed. In accordance with our second hypothesis, the performance within the more complex DT‐agility test additionally incorporates executive functions and was not related to motor functions. Ecological validity analyses suggest that the reactive agility test is associated with football test performance, while the DT‐agility test was correlated with players' potential rating and playing times. This partially confirms our third hypothesis.

4.1. Criterion validity

Analysis on criterion validity suggest that in elite youth football players, performance within the conventional reactive agility test and reactive agility test with incorporated multiple object tracking are moderately related to cognitive functions but not to motor performance. Thus, criterion validity of these two agility tests seems to be limited and confined to cognitive functions.

To the best of the authors' knowledge no other study has examined the relationship between reactive agility speed and cognitive performance. However, our results are in line with the conclusion made by Scanlan et al. (2014) as well as Young and Willey (Young & Willey, 2010) who found premotor time (processing time between stimulus and movement initiation) within a reactive agility task to be the sole predictor for reactive agility performance in high level basketball and football athletes. This is supported by the review of Young et al. (2015), which indicates only a very heterogeneous explanation of the variance in reactive agility by motor capacities (r 2 = 0.1–0.49) depending on the applied test procedure and study collective. Compared to other reactive agility tests with higher motor component (e.g., Y‐Agility), the Random Star Run has considerably shorter straight running distances and higher number of stimuli to be processed, which again seems to increase the contribution of the cognitive component. Contrarily, a previous study conducted on the SKILLCOURT showed moderate to high correlations between the reactive agility test and sprinting as well as jumping performance (Hülsdünker et al., 2023). However, the study population of university students was more heterogeneous and on a lower level in their motor capacities. As the included football players in the present study play at the highest level, the motor abilities to cope with the test are more homogeneous across participants. This is supported by the very low variance in performance within the 5 and 10 m sprinting times. These findings may suggest that reactive agility performance in elite athletes is primarily related to the ability to make quick decisions rather than motor functions.

Within the DT‐agility test, the integration of the multiple object tracking task leads to an additional requirement of executive functions such as working memory and cognitive flexibility. The maximal tracking speed being associated with attentional as well as working memory capacities is well in line with previous studies on multiple object tracking in single‐task conditions (Ehmann et al., 2021; Harris et al., 2020). The tracking task forces the athletes to constantly memorize and update the position and movement of the balls. The combination with the reactive agility test in a dual‐task approach seems to force the athletes to switch their focus and priority between the tracking and running task, which additionally required cognitive flexibility skills. The lack of correlation between the number of reached fields in the DT‐agility test and motor functions aligns with previous research studies, indicating reduced motor performance within dual‐task situations (Büchel et al., 2022; Moreira et al., 2021). For example, Büchel et al. (2022) found movement speed and therefore the tests motor component decreases with increasing cognitive complexity of an agility task. Therefore, it can be assumed that the capacity to handle the cognitive tasks while being in motion seem to be the determining factor in reaching as many fields as possible.

The analysis on criterion validity indicates that incorporating cognitive tasks in agility testing results in a task‐dependent enhancement of the importance of cognitive functions while reducing the relevance of motor performance. However, considering the moderate correlation between agility performance and cognitive functions, as well as the absence of associations with motor performance, it is reasonable to conclude that this form of agility testing represents a distinct test construct influenced by motor–cognitive dual‐tasking effects. The lack of correlation between the two tests of the agility battery confirm that, depending on the situational demands, agility seems to be highly context specific (Jeffreys, 2011). Therefore, agility in football cannot be a fixed construct but rather a dynamic and fluent interaction between cognitive, motor, and technical abilities (Büchel et al., 2022; Fiorilli et al., 2017). This underlines the importance of examining agility under different motor–cognitive weightings.

4.2. Ecological validity

The reactive agility is moderately associated with football‐specific test performance, while the DT‐agility test correlates with the playing time and coaches rating. These findings support the ecological validity of both motor–cognitive agility tests and underscore the importance of integrating higher cognitive demands to enhance their predictive value for comprehensive performance diagnostics.

The relevance of agility performance for the LSPT is supported by the study of Benounis et al. (2013), who found test time in the Illinois agility test to be one of the primary predictors for the test time. Therefore, the ability to rapidly change direction in response to the auditory stimulus, even before receiving the ball, seemed to be crucial for fast passing in high perceptual‐cognitive demanding situations. The lack of correlation of reactive agility test performance with players' potential as well as playing time could be explained by the comparably low ecological validity of the test for general on‐field performance in football. The ability to respond as quickly as possible to external stimuli might be highly relevant in certain unpredictable game situations, such as those recreated in the LSPT but do not reflect the high motor–cognitive demands of the on‐field environment (Altmann et al., 2023). This is supported by previous research studies, which primarily emphasizes the executive functions of the athletes for the on‐field performance (Huijgen et al., 2015; Vestberg et al., 2012, 2017). However, as the criterion validity analysis confirms, these are not required in the reactive agility test.

In contrast, the DT‐agility test was not significantly associated with LSPT performance, but correlated moderately with potential rating as well as in‐season playing time. Furthermore, it showed an excellent discrimination quality between players on the basis of these indirect measures of playing performance (Hosmer et al., 2013). The correlations between the measures of match‐related performance measures with the DT‐agility test even tend to be higher than with isolated cognitive assessments used in previous studies (Sabarit et al., 2020; Scharfen & Memmert, 2021b; Vestberg et al., 2012). In youth football players, Sabarit et al. (2020) found weak correlations between attention and processing speed (r = 0.22–0.3) with the performance in a small‐sided game. In addition, Scharfen and Memmert (2021b) found that attention and executive functions (working memory, cognitive flexibility) were weakly correlated to rated game intelligence (r = 0.28–0.3) as well as game time (r = 0.22–0.34) in elite football players. However, contrary to our results, they found no relationship between multiple object tracking performance and any game metrics. This supports the authors' assumption (Scharfen & Memmert, 2021a) that embedding the tracking task in a sport‐specific dual‐task setting might increase ecological validity and thus the predictive value for sport performance.

This is supported by the DT‐agility test's ability to differentiate between players based on playing time and potential rating. While previous studies also indicated the discriminative value of cognitive performance (Huijgen et al., 2015; Vestberg et al., 2012), in these studies elite players were compared with sub‐elite or amateur players, suggesting a higher heterogeneity in the performance level. In addition, in the study of Scharfen and Memmert (Ehmann et al., 2021) multiple object tracking alone was unable to discriminate between athletes of different sport types. Hence, our results suggest that isolated tests of cognitive functions are not suitable for diagnostic purposes within elite athletes. Therefore, integrating these relevant cognitive functions (e.g., executive functions, attentional processes, reaction speed) into sport‐specific tasks, as in the DT‐agility test approach, seems to be crucial to increase sensitivity as well as specificity. This confirms the importance of considering both task and stimulus correspondence in the development of ecologically valid assessments for elite sports.

4.3. Limitations and future research directions

One limitation of the study is the comparatively small sample size, which can be explained by the difficulty in recruiting participants on this performance level. Similarly, this is the reason why further regression analysis as well as position‐specific recruitment/analysis could not be performed. Future studies could take this into account to further specify the ecological validity and relevance of the tests for certain playing positions. The final assessment of the criterion validity of our test battery requires further studies assessing additional relevant motor function (e.g., agility test without reactive component, drop jumps). The additional consideration of isolated multiple object tracking would also allow for the evaluation of the extent to which motor–cognitive interferences possibly caused by the dual‐task testing increases the prognostic value of the DT‐agility test with regard to playing performance. The potential rating of the coaches showed only moderate agreement. This could be the case because a team‐coach and video‐analyst evaluate players based on different criteria due to their respective expertise. Nevertheless, a valid estimation of the players' potential to play on professional level should have been achieved by averaging the scores. While valid and reliable in predicting the chance of players to reach the professional level (Jokuschies et al., 2017; Sieghartsleitner et al., 2019), the coaches rating and playing time are no direct game performance metrics. Additional game‐specific parameters such as pass rates, duel statistics, running characteristics, or goal participation could be integrated in future studies to enhance the explanatory power. Although it was found that the DT‐agility test showed a higher discriminative value than the traditional agility test, future studies should investigate the extent to which the ecological validity can be further increased by integrating additional football‐specific components, such as technical or tactical tasks. Finally, future studies could investigate the potential benefits of combining cognitive tasks and reactive agility in a dual‐task training and examine whether this approach increases the transfer to football‐specific performance.

5. CONCLUSION

In elite youth football players, performance in both motor–cognitive agility tests appears to be primarily determined by cognitive rather than motor performance. While the reactive agility test is associated with football‐specific skills (e.g., changing direction with ball handling), the DT‐agility test demonstrates correlations with measures related to game performance (coaches rating, playing time). By considering task and stimulus correspondence, this dual‐task approach appears to have high ecological validity further underscored by an excellent discriminative value in distinguishing between players based on their potential and playing times. Therefore, the DT‐agility test could serve as standardized tool for diagnostics of large collectives with high performance homogeneity, for example, in scouting of elite youth football players.

Thus the developed DT‐agility test may provide a more ecological valid assessment when compared to currently existing agility as well as isolated motor and cognitive tests that can provide great value to athletes and coaches by improving performance diagnostics, training, and scouting.

CONFLICT OF INTEREST STATEMENT

Movement concepts GmbH supported the study by providing the SKILLCOURT technology and funding. TH provides scientific consultancy to movement concepts GmbH. The company was not involved in any aspect of the study including study design, data acquisition, data analysis, result interpretation, and writing the manuscript.

ACKNOWLEDGMENTS

None.

Open Access funding enabled and organized by Projekt DEAL.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available upon request from the corresponding author (DF).

REFERENCES

  1. Altmann, S. , Ruf L., Härtel S., and Woll A.. 2023. “Agilität im Mannschaftssport ‐ Bedeutung, Erfassung, Training.” Trainingswissenschaft und ‐Lehre(1): 23–28. [Google Scholar]
  2. BenOunis, Omar , BenAbderrahman Abderaouf, Chamari Karim, Ajmol Ali, BenBrahim Mehdi, Hammouda Amor, Hammami M.‐Ali, and Zouhal Hassane. 2013. “Association of Short‐Passing Ability with Athletic Performances in Youth Soccer Players.” Asian Journal of Sports Medicine 4(1): 41–48. 10.5812/asjsm.34529. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Broadbent, David P. , Causer Joe, Williams A. Mark, and Ford Paul R.. 2015. “Perceptual‐cognitive Skill Training and its Transfer to Expert Performance in the Field: Future Research Directions.” European Journal of Sport Science 15(4): 322–331. 10.1080/17461391.2014.957727. [DOI] [PubMed] [Google Scholar]
  4. Büchel, Daniel , Gokeler Alli, Heuvelmans Pieter, and Baumeister Jochen. 2022. “Increased Cognitive Demands Affect Agility Performance in Female Athletes ‐ Implications for Testing and Training of Agility in Team Ball Sports.” Perceptual and Motor Skills 129(4): 1074–1088. 10.1177/00315125221108698. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cohen, J. 1988. Statistical Power Analysis for the Behavioral Sciences. 2nd ed. Hillsdale, NJ: Erlbaum. [Google Scholar]
  6. Deary, Ian J. , Liewald David, and Nissan Jack. 2011. “A Free, Easy‐To‐Use, Computer‐Based Simple and Four‐Choice Reaction Time Programme: the Deary‐Liewald Reaction Time Task.” Behavior Research Methods 43(1): 258–268. 10.3758/s13428-010-0024-1. [DOI] [PubMed] [Google Scholar]
  7. Ehmann, Paul , Beavan Adam, Spielmann Jan, Ruf Ludwig, Mayer Jan, Rohrmann Sonja, Nuß Christian, and Englert Chris. 2021. “360°‐multiple Object Tracking in Team Sport Athletes: Reliability and Relationship to Visuospatial Cognitive Functions.” Psychology of Sport and Exercise 55: 101952. 10.1016/j.psychsport.2021.101952. [DOI] [Google Scholar]
  8. Faubert, Jocelyn , and Sidebottom Lee. 2012. “Perceptual‐Cognitive Training of Athletes.” Journal of Clinical Sport Psychology 6(1): 85–102. 10.1123/jcsp.6.1.85. [DOI] [Google Scholar]
  9. Fiorilli, G. , Iuliano E., Mitrotasios M., Pistone E. M., Aquino G., Calcagno G., et al. 2017. “Are Change of Direction Speed and Reactive Agility Useful for Determining the Optimal Field Position for Young Soccer Players?” Journal of Sports Science and Medicine 16(2): 247–253. [PMC free article] [PubMed] [Google Scholar]
  10. Friebe, David , Hülsdünker Thorben, Giesche Florian, Banzer Winfried, Pfab Florian, Haser Christian, and Vogt Lutz. 2023. “Reliability and Usefulness of the Skillcourt as a Computerized Agility and Motor‐Cognitive Testing Tool.” Medicine & Science in Sports & Exercise 55(7): 1265–1273. 10.1249/MSS.0000000000003153. [DOI] [PubMed] [Google Scholar]
  11. Harris, David J. , Wilson Mark R., Crowe Emily M., and Vine Samuel J.. 2020. “Examining the Roles of Working Memory and Visual Attention in Multiple Object Tracking Expertise.” Cognitive Processing 21(2): 209–222. 10.1007/s10339-020-00954-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Haugen, Thomas A. , Tønnessen Espen, Hisdal Jonny, and Seiler Stephen. 2014. “The Role and Development of Sprinting Speed in Soccer.” International Journal of Sports Physiology and Performance 9(3): 432–441. 10.1123/ijspp.2013-0121. [DOI] [PubMed] [Google Scholar]
  13. Hosmer, D. W. , Lemeshow S., and Sturdivant R. X.. 2013. Applied Logistic Regression. Wiley. [Google Scholar]
  14. Huijgen, Barbara C. H. , Leemhuis Sander, Kok Niels M., Verburgh Lot, Oosterlaan Jaap, Elferink‐Gemser Marije T., and Visscher Chris. 2015. “Cognitive Functions in Elite and Sub‐elite Youth Soccer Players Aged 13 to 17 Years.” PLoS One 10(12): e0144580. 10.1371/journal.pone.0144580. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hülsdünker, Thorben , Friebe David, Giesche Florian, Vogt Lutz, Pfab Florian, Haser Christian, and Banzer Winfried. 2023. “Validity of the SKILLCOURT® Technology for Agility and Cognitive Performance Assessment in Healthy Active Adults.” Journal of Exercise Science & Fitness 21(3): 260–267. 10.1016/j.jesf.2023.04.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Jeffreys, Ian . 2011. “A Task‐Based Approach to Developing Context‐specific Agility.” Strength and Conditioning Journal 33(4): 52–59. 10.1519/SSC.0b013e318222932a. [DOI] [Google Scholar]
  17. Jokuschies, Nina , Gut Vanessa, and Conzelmann Achim. 2017. “Systematizing Coaches’ ‘eye for Talent': Player Assessments Based on Expert Coaches’ Subjective Talent Criteria in Top‐Level Youth Soccer.” International Journal of Sports Science and Coaching 12(5): 565–576. 10.1177/1747954117727646. [DOI] [Google Scholar]
  18. Kelly, Adam L. , Wilson Mark R., Jackson Daniel T., Turnnidge Jennifer, and Williams Craig A.. 2021. “Speed of Thought and Speed of Feet: Examining Perceptual‐Cognitive Expertise and Physical Performance in an English Football Academy.” Journal of Science in Sport and Exercise 3(1): 88–97. 10.1007/s42978-020-00081-2. [DOI] [Google Scholar]
  19. Kessels, Roy P. C. , van Zandvoort Martine J. E., Postma Albert, Kappelle L. Jaap, and de Haan Edward H. F.. 2000. “The Corsi Block‐Tapping Task: Standardization and Normative Data.” Applied Neuropsychology 7(4): 252–258. 10.1207/S15324826AN0704_8. [DOI] [PubMed] [Google Scholar]
  20. Le Moal, Emmeran , Rué Olivier, Ajmol Ali, Abderrahman Abderaouf B., Hammami Mohammed A., Ounis Omar B., Kebsi Wiem, and Zouhal Hassane. 2014. “Validation of the Loughborough Soccer Passing Test in Young Soccer Players.” The Journal of Strength & Conditioning Research 28(5): 1418–1426. 10.1519/JSC.0000000000000296. [DOI] [PubMed] [Google Scholar]
  21. Moreira, Pedro Emílio Drumond , Dieguez Gabriel Teles de Oliveira, Bredt Sarah da Glória Teles, and Praça Gibson Moreira. 2021. “The Acute and Chronic Effects of Dual‐Task on the Motor and Cognitive Performances in Athletes: A Systematic Review.” International Journal of Environmental Research and Public Health 18(4): 1732. 10.3390/ijerph18041732. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Pojskic, Haris , Åslin Erik, Krolo Ante, Jukic Ivan, Uljevic Ognjen, Spasic Miodrag, and Sekulic Damir. 2018. “Importance of Reactive Agility and Change of Direction Speed in Differentiating Performance Levels in Junior Soccer Players: Reliability and Validity of Newly Developed Soccer‐specific Tests.” Frontiers in Physiology 9: 506. 10.3389/fphys.2018.00506. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Romeas Thomas, Chaumillon Romain, Labbé David, & Faubert Jocelyn (2019). Combining 3D‐MOT with Sport Decision‐Making for Perceptual‐Cognitive Training in Virtual Reality. Perceptual and Motor Skills, 126(5), 922–948. 10.1177/0031512519860286 [DOI] [PubMed] [Google Scholar]
  24. Sabarit, Alejandro , Reigal Rafael E., Morillo‐Baro Juan P., Juárez‐Ruiz de Mier Rocío, Franquelo Auxiliadora, Hernández‐Mendo Antonio, Falcó Coral, and Morales‐Sánchez Verónica. 2020. “Cognitive Functioning, Physical Fitness, and Game Performance in a Sample of Adolescent Soccer Players.” Sustainability 12(13): 5245. 10.3390/su12135245. [DOI] [Google Scholar]
  25. Scanlan, Aaron , Humphries Brendan, Tucker Patrick S., and Dalbo Vincent. 2014. “The Influence of Physical and Cognitive Factors on Reactive Agility Performance in Men Basketball Players.” Journal of Sports Sciences 32(4): 367–374. 10.1080/02640414.2013.825730. [DOI] [PubMed] [Google Scholar]
  26. Scharfen, H.‐Erik , and Memmert Daniel. 2019. “The Relationship between Cognitive Functions and Sport‐specific Motor Skills in Elite Youth Soccer Players.” Frontiers in Psychology 10: 817. 10.3389/fpsyg.2019.00817. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Scharfen, H.‐Erik , and Memmert Daniel. 2021a. “Cognitive Training in Elite Soccer Players: Evidence of Narrow, but Not Broad Transfer to Visual and Executive Function.” German Journal of Exercise and Sport Research 51(2): 135–145. 10.1007/s12662-020-00699-y. [DOI] [Google Scholar]
  28. Scharfen, H.‐Erik , and Memmert Daniel. 2021b. “Fundamental Relationships of Executive Functions and Physiological Abilities with Game Intelligence, Game Time and Injuries in Elite Soccer Players.” Applied Cognitive Psychology 35(6): 1535–1546. 10.1002/acp.3886. [DOI] [Google Scholar]
  29. Schmuckler, Mark A. 2001. “What Is Ecological Validity? A Dimensional Analysis.” Infancy: The Official Journal of the International Society on Infant Studies 2(4): 419–436. 10.1207/S15327078IN0204_02. [DOI] [PubMed] [Google Scholar]
  30. Schober, Patrick , Boer Christa, and Schwarte Lothar A.. 2018. “Correlation Coefficients: Appropriate Use and Interpretation.” Anesthesia & Analgesia 126(5): 1763–1768. 10.1213/ANE.0000000000002864. [DOI] [PubMed] [Google Scholar]
  31. Sieghartsleitner, R. , Zuber C., Zibung M., and Conzelmann A.. 2019. “Science or Coaches’ Eye? – Both! Beneficial Collaboration of Multidimensional Measurements and Coach Assessments for Efficient Talent Selection in Elite Youth Football.” Journal of Sports Science and Medicine 18(1): 32–43. [PMC free article] [PubMed] [Google Scholar]
  32. Stoet, Gijsbert . 2010. “PsyToolkit: a Software Package for Programming Psychological Experiments Using Linux.” Behavior Research Methods 42(4): 1096–1104. 10.3758/BRM.42.4.1096. [DOI] [PubMed] [Google Scholar]
  33. Trajković, Nebojša , Sporiš Goran, Krističević Tomislav, Madić Dejan M., and Bogataj Špela. 2020. “The Importance of Reactive Agility Tests in Differentiating Adolescent Soccer Players.” International Journal of Environmental Research and Public Health 17(11): 3839. 10.3390/ijerph17113839. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Treisman, Anne M. , and Gelade Garry. 1980. “A Feature‐Integration Theory of Attention.” Cognitive Psychology 12(1): 97–136. 10.1016/0010-0285(80)90005-5. [DOI] [PubMed] [Google Scholar]
  35. Vandierendonck, André , Liefooghe Baptist, and Verbruggen Frederick. 2010. “Task Switching: Interplay of Reconfiguration and Interference Control.” Psychological Bulletin 136(4): 601–626. 10.1037/a0019791. [DOI] [PubMed] [Google Scholar]
  36. Vater, Christian , Gray Rob, and Holcombe Alex O.. 2021. “A Critical Systematic Review of the Neurotracker Perceptual‐Cognitive Training Tool.” Psychonomic Bulletin and Review 28(5): 1458–1483. 10.3758/s13423-021-01892-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Vestberg, Torbjörn , Gustafson Roland, Maurex Liselotte, Ingvar Martin, and Petrovic Predrag. 2012. “Executive Functions Predict the Success of Top‐Soccer Players.” PLoS One 7(4): e34731. 10.1371/journal.pone.0034731. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Vestberg, Torbjörn , Reinebo Gustaf, Maurex Liselotte, Ingvar Martin, and Petrovic Predrag. 2017. “Core Executive Functions Are Associated with Success in Young Elite Soccer Players.” PLoS One 12(2): e0170845. 10.1371/journal.pone.0170845. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Waldén, Markus , Mountjoy Margo, McCall Alan, Serner Andreas, Massey Andrew, Tol Johannes L., Bahr Roald, et al. 2023. “Football‐specific Extension of the IOC Consensus Statement: Methods for Recording and Reporting of Epidemiological Data on Injury and Illness in Sport 2020.” British Journal of Sports Medicine 57(21): 1341–1350. 10.1136/bjsports-2022-106405. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. World Medical Association . 2013. “World Medical Association Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects.” JAMA 310(20): 2191–2194. 10.1001/jama.2013.281053. [DOI] [PubMed] [Google Scholar]
  41. Young, Warren B. , Dawson Brian, and Henry Greg J.. 2015. “Agility and Change‐Of‐Direction Speed Are Independent Skills: Implications for Training for Agility in Invasion Sports.” International Journal of Sports Science and Coaching 10(1): 159–169. 10.1260/1747-9541.10.1.159. [DOI] [Google Scholar]
  42. Young, Warren B. , and Willey Ben. 2010. “Analysis of a Reactive Agility Field Test.” Journal of Science and Medicine in Sport 13(3): 376–378. 10.1016/j.jsams.2009.05.006. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The data that support the findings of this study are available upon request from the corresponding author (DF).

Articles from European Journal of Sport Science are provided here courtesy of Wiley

RESOURCES