Comparison of functional vestibulo-ocular reflex and cognitive functions in e-athletes and non-athletes

Abstract

Background

Esports has become a global phenomenon, yet its cognitive and vestibular demands are less studied than in traditional sports. We compared visuospatial/short-term memory, reaction time, and functional vestibulo-ocular reflex (VOR) between e-athletes and non-athletes.

Methods

In this comparative cross-sectional study, eighteen e-athletes and eighteen age-matched controls (17–27 years) with normal/corrected vision and no neurological or psychiatric disorders were assessed. Cognitive performance was measured with the Stroop Test and Corsi Block Test (forward/backward spans). Functional VOR was evaluated using the Functional Head Impulse Test (fHIT). Group differences were tested with Mann–Whitney U (two-tailed, α = 0.05).

Results

E-athletes showed faster responses and fewer errors on the Stroop Test (Stroop V time: 15.80 ± 2.01 s vs. 18.42 ± 1.26 s, p < 0.001; Stroop IV errors: 0.27 ± 0.46 vs. 0.94 ± 0.53, p = 0.001) and higher Corsi spans (forward: 7.11 ± 0.83 vs. 5.66 ± 0.68, p < 0.001; backward: 5.83 ± 0.85 vs. 4.00 ± 0.78, p < 0.001). No significant differences between-group were observed on fHIT outcomes (p > 0.05).

Conclusions

E-athletes exhibited superior attentional control and visuospatial memory, with no evidence of differences in functional VOR. Intensive gaming may selectively relate to executive and memory functions without influencing vestibular performance. Larger studies are warranted to confirm these findings and inform training/health applications.

Trial registration

NCT07211191–22 / 09 / 2025.

Background

E-sports, or competitive video gaming, has rapidly grown into a global phenomenon, with over 620 million unique viewers and thousands of professional players participating in structured tournaments across diverse game genres such as Multiplayer Online Battle Arena (MOBA), First-Person Shooter (FPS) [1, 2]. For instance, FPS games demand rapid visuomotor coordination and response inhibition, while MOBA and real-time strategy (RTS) games rely heavily on multitasking, long-term planning, and dynamic information integration. The demographic profile of e-sports participants is predominantly male, with most players aged between 16 and 25 years, and significant regional concentrations in Asia, North America, and Europe [3, 4]. This widespread engagement, combined with the high cognitive and strategic demands of competitive gaming, underscores the importance of understanding the health, behavioral, and performance-related aspects of this population [5–7]. As the e-sports industry grows there is a focus on understanding player performance factors including physical skills as well as cognitive abilities such as attention, perception and visuospatial skills [8, 9]. Different game genres impose varying demands. For instance, FPS games demand high visuomotor coordination and rapid reaction times, while RTS games emphasize strategic planning and working memory. These genre-specific demands may influence cognitive development in distinct ways.

E-sports is associated with numerous cognitive abilities including the prefrontal cortex working memory, executive attention, decision-making fluid intelligence and emotion processing [10–12]. While e-sports do not involve the level of physical exertion required for many other sports it can result in a psychological state of stress or tension leading to changes in the autonomic nervous system [13]. Executive functions which are the cognitive functions that govern our behaviors such as decision-making abilities organization planning and goal setting time management and self-regulation are utilized during gameplay. An e-sports player (e-athletes) performs complex actions while simultaneously analyzing various stimuli to form fluid and coordinated actions attempting to minimize the number of erroneous choices that could harm their goals. Some research has found that individuals who play video games may exhibit faster reaction times although accuracy in some executive function measures may decrease [14].

On the other hand, it should be noted that contradictory reports have emerged in some studies, especially regarding the cognitive effects of e-sports. For example, Argyriou et al. [15] and Kronenberger et al. [16] found that response inhibition may be impaired in gamers contrasting with Bediou et al. [17] who found that gaming improves cognitive skills. E-athletes or professional gamers are subject to intense mental and physical challenges that require quick reflexes, precise hand-eye coordination and acute decision-making skills [18]. To understand these demands more comprehensively, researchers have started using assessment tools from other fields, such as vestibular and neurocognitive testing, to evaluate e-sports performance [19, 20]. One such method is the functional head impulse test (fHIT) which provides valuable insights into vestibular function and could be adapted to assess the sensory-motor integration and reaction times of e-athletes.

fHIT is a newly developed testing method where participants are asked to identify the optotype “C” displayed on a computer screen during passive head thrusts. This test differs from the vHIT. Even slight shifts in the retinal image can reduce dynamic visual acuity. The fHIT assesses reading ability and the ability to maintain clear vision during head movement. Therefore, it can be used to evaluate the functional vestibulo-ocular reflex (VOR) [21].

The rotational VOR stabilizes gaze by generating eye movements that counteract head rotations. Evaluating the VOR provides a reliable assessment of vestibular organ function due to the direct rapid three-neuron pathway linking the semicircular canals to the eye muscles [22]. The head impulse test along with its video-oculography-based version known as the video head impulse test (vHIT) measures VOR responses to head accelerations at frequencies between 1 and 5 Hz which are at the higher end of natural head movement frequencies [23]. The test’s result referred to as “VOR gain” is the ratio of eye movement to head movement (such as eye velocity to head velocity) averaged over a specific time window. This VOR gain objectively measures how much of the head movement is compensated by the eye movement response. Clinically assessing VOR gain with vHIT is crucial to rule out vestibular impairment [24].

E-athletes might be assessed with fHIT as well as the assessment of cognitive skills such as visuospatial memory and short-term memory to meet the cognitive and physical demands of competitive games. Visuospatial memory and reaction time is are critical for e-sports players as it enables them to interpret complex visual information anticipate opponent actions and navigate virtual environments efficiently. Short-term memory supports rapid decision-making and strategy formulation during gameplay. The fHIT is particularly relevant as it evaluates vestibular function and reflexive head-eye coordination which are essential for maintaining focus and accuracy during prolonged gaming sessions. By assessing these cognitive and physical aspects comprehensive insights might be gained into an e-athlete’s capabilities allowing for the development of targeted training programs that enhance performance and address any potential deficits. Despite growing academic interest in e-sports, there is a lack of empirical studies that assess both cognitive performance and functional vestibular responses within the same group of participants. To our knowledge, this is the first study to combine Stroop Test (ST), Corsi Block Test (CBT) and fHIT assessments in a single cohort of e-athletes, offering novel insights into the relationship between cognitive flexibility and gaze stabilization in competitive gaming.

The cognitive and oculomotor demands of competitive gaming may involve the concurrent engagement of executive control processes (e.g., inhibition and visuospatial updating) together with sensory–motor mechanisms that contribute to gaze stabilization during rapid visual scanning [25]. Although previous studies have examined cognitive performance or visual–vestibular function independently, these domains have only occasionally been explored within the same group of participants [17, 18]. Integrating assessments such as the ST, CBT, and fHIT may therefore offer potential contributions toward understanding how central processing and gaze-stability mechanisms could interact in e-athletes, who often rely on rapid stimulus–response integration during gameplay.

The aim of this study was to compare functional VOR responses and visuospatial memory reaction time and short-memory (executive functions) cognition skills using fHIT, CBT, and ST in e-athletes and non-e-athletes.

We hypothesized that e-athletes would outperform controls on executive and visuospatial tasks (Stroop and Corsi), while showing similar functional VOR performance (fHIT).

Methods

Participants

The study was carried out in Kucukcekmece Municipality E-sports Center between June-September 2021. Participants were recruited using a non-probability convenience sampling method. Announcements were distributed through the Esports Center at Kucukcekmece Municipality and relevant university departments, inviting eligible individuals to voluntarily participate in the study. This study was carried out in compliance with the Declaration of Helsinki and received approval from the local ethics committee (Decision no: 855/2021, Istanbul Medipol University, Non-Invasive Clinical Research Ethics Committee). Written informed consent was obtained from all participants prior to their involvement in the study. Furthermore, written permission was obtained from them for the publication of representative photographs of all participants during the fHIT performance and test results as illustrative examples in a scientific article. Our study consisted of two groups: Control group (CG) and study group (SG).

This observational study was designed and reported in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines.

Participants were recruited without discrimination based on gender, ethnicity, or socioeconomic background. Although gender and socioeconomic status were not used as formal matching criteria, the final group distributions were similar due to the demographic characteristics of the recruitment pool. Nevertheless, the potential influence of gender- or SES-related differences on cognitive outcomes represents a limitation, and future studies should consider matched or stratified sampling methods. Both recruitment and analysis were performed in a manner ensuring equal treatment and representation. Although the sample predominantly consisted of young adults and males, inclusion criteria did not exclude participants based on demographic characteristics, and all eligible individuals were offered equal opportunity to participate. Although structured anamnesis is appropriate for initial screening in field-based research, it does not allow precise quantitative assessment of visual fatigue, screen exposure, or physical activity levels. These variables ideally require standardized measurement tools for more accurate characterization.

Study design

GPower 3.1 program was used to determine the study sample size. Type 1 error (alpha) was 0.05, Type 2 error (β) was 0.20 (power = 0.80), and effect size was 0.8. It was determined that there should be a minimum of 26 participants in each group for the study. However, 18 participants were included in each group due to the exclusion of participants who did not participate in some tests due to the duration of the total tests. Based on previous studies reporting large effect sizes in cognitive comparisons between e-sports and non-esports participants, we selected an effect size of 0.8 (Cohen’s d), with α = 0.05 and β = 0.20 (power = 0.80), which yielded a required minimum of 26 participants per group [17, 25]. Due to time constraints and test duration, 18 participants per group were ultimately included.

The inclusion criteria for the CG were as follows: (1) 17–27 years of age group, (2) No cognitive and mental problems, adaptability to the test and (3) not having a disability that restricts the neck. The inclusion criteria for the SG were as follows: (1) 17–27 years of age, (2) have been involved in e-sports for at least 4 years, at least 20 h or more per week, (3) can adapt to the test, (4) do not have a disability that restricts the neck. Subjects who did not meet the inclusion criteria specified in both groups were excluded from the study. We have clarified that cognitive and mental health status was assessed using the Montreal Cognitive Assessment (MoCA) as a standardized screening tool. In addition, participants’ clinical history was reviewed as part of the anamnesis process. CG participants were also required to have no prior competitive involvement in esports or organized sports training to avoid potential confounding effects related to cognitive or vestibular adaptation.

All participants had previously undergone the routine medical clearance required by the e-sports center, including evaluations by neurology, psychology, orthopedics, and otolaryngology specialists, and had documented normal hearing thresholds on screening audiometry (500–4000 Hz). Because data collection occurred in a non-clinical field setting and our VNG system was not a portable device, objective vestibular assessments could not be performed. Therefore, vestibular and auditory status was screened using structured anamnesis in accordance with Bárány Society recommendations; however, this approach does not replace instrument-based diagnostic testing and is acknowledged as a methodological limitation.

The evaluation included questions addressing demographic details, gaming experience (GE) any experience of hearing loss, tinnitus or ear buzzing, episodes of dizziness within the past three months, visual impairments, migraine symptoms or diagnosis, as well as a history or diagnosis of benign paroxysmal positional vertigo (BPPV), head trauma, systemic diseases, and neurological conditions. These inquiries were guided by the diagnostic criteria established by the Bárány Society and the International Headache Society’s vestibular disorder classification committee. These anamnesis-based assessments were used to screen for exclusion criteria such as recent vestibular symptoms, visual impairments, history of neurological disease, or other medical conditions that might affect cognitive or vestibular function.

fHIT

The fHIT assesses rotational VOR in terms of high-speed, high-acceleration functional performance, specifically evaluating the capacity for gaze stabilization. The fHIT system (Beon Solutions, Zero Branco, Italy) was used for this procedure, with participants positioned 1.5 m from the fHIT computer monitor (Figs. 1 and 2). During testing, a gyroscope was attached to each participant’s head to evaluate their VOR and record angular head velocity.

Fig. 1.

Vertical canal stimulation position during the Functional Head Impulse Test (fHIT)

Fig. 2.

Horizontal (lateral) canal stimulation position during the Functional Head Impulse Test (fHIT)

A single clinician administered the test consistently across all participants. Initially, each participant’s static visual acuity (SVA) was assessed. The Landolt C optotype size first appeared at 1.0 LogMAR and decreased in size after every three correct responses. This iterative approach helped identify the minimum threshold value for each participant. The optotype size was tailored individually and remained constant throughout the testing process.

At least 10 passive head impulses were administered, involving acceleration levels between 4,000 and 6,000°/s² for horizontal movements and 3,000 to 6,000°/s² for vertical movements (covering both left anterior-right posterior and right anterior-left posterior directions). During this test, participants were instructed to focus on a Landolt C optotype, which appeared in one of eight possible orientations. The optotype was displayed randomly on a computer screen for a brief duration of 80 milliseconds. As participants’ heads were rotated, they were required to identify the orientation of the optotype and select the correct direction for the Landolt C (Fig. 2). The fHIT system then calculated the test outcome, which was expressed as the percentage of correct answers (%CA) for each side.

The Stroop test

Stroop Test (ST) response inhibition selective attention information processing speed and a widely used tool for assessing cognitive flexibility executive function test. The test was developed by John Ridley Stroop in 1935 and is used in many of the tests in use today [26]. Many forms are available. These forms differ in terms of stimulus number type and task order. However, they all produce a phenomenon described as the Stroop effect. The Stroop effect generally provides the individual with a colour-word when tasks involving incompatibility are given (e.g. the word “red” printed in blue ink reading task or saying the ink colour instead of reading task). During such a task the individual’s more automatic task of colour naming by suppressing the reading behavior. The more automatic behavior suppression effort leads to a prolonged reaction time and this is called the Stroop effect [27].

In Turkey, the ST was standardized through the development of the TBAG Form, created by combining the original ST and the Victoria version within the scope of the BILNOT Battery, and normative values were established [28, 29]. In the TBAG Form Victoria Unlike Stroop there are five tasks instead of three. The person is asked to read the colour names say the colour and finally and the task of saying the colour of colour names printed in different colours [28, 29]. The ST was administered by a psychologist competent in this test.

Corsi block test

CBT measures visual-spatial short-term memory in individuals. The classic version of CBT comprises nine cubical blocks arranged on a board. The tester touches the blocks in a certain order; starting with two blocks the number of blocks touched is increased by one throughout the test. The applications in the test are as follows: (1) Forward Recall. After the tester finishes tapping the person taps the blocks in the same order. (2) Backward Recall. After the tester finishes tapping the person taps the blocks backwards. Scoring is based on the number of correct responses (number of blocks correctly tapped in sequence for both applications). In the CBT, the number of blocks correctly tapped in forward recall reflects the participant’s visuospatial memory span, whereas the backward recall score reflects working memory capacity [30]. The test was administered by a psychologist experienced in its application. The assessments were performed sequentially due to the use of a single testing station, which resulted in natural waiting intervals between tasks. Participants typically waited 20–30 min between the Stroop and Corsi assessments, and at least 30 min before the fHIT session, which helped minimize short-term carry-over effects.

Data analysis

The distribution of all continuous variables was evaluated using the Shapiro-Wilk test. Descriptive statistics, including mean values, standard deviations, frequencies, median, interquartile range (IQR) and percentages, were employed in the analysis. Normality was assessed using the Shapiro–Wilk test for each variable with sufficient variance; variables with zero variance were marked accordingly. As many parameters demonstrated non-normal distributions, nonparametric analyses were applied. For all comparisons, False Discovery Rate (FDR)–adjusted p-values and effect sizes (Cohen’s d) were calculated and reported to improve statistical transparency. Shapiro–Wilk p-values, adjusted p-values, and effect sizes have been added to Tables 1 and 2. The Mann-Whitney U test was utilized to compare paired groups (3,000°/s², 4,000°/s², 5,000°/s², 6,000°/s², ST and CBT parameters). All statistical analyses were conducted using IBM SPSS software, version 22.0. A significance level of p < 0.05 was established for determining statistical significance.

Table 1.

Comparison of SVA and fHIT test results between CG and SG

Parameter	CG Mean ± SD	Median (IQR)	Min–Max	SG Mean ± SD	Median (IQR)	Min–Max	FDR-adjusted p	Shapiro–Wilk p (CG)	Shapiro–Wilk p (SG)	Effect size (d)
Static visual acuity	0.32 ± 0.94	0.30 (0.10)	0.2–0.5	0.27 ± 0.80	0.20 (0.10)	0.2–0.4	0.160	–	–	–
Horizontal Right 4000°/s²	98.44 ± 4.97	100 (0)	80–100	100.00 ± 0.00	100 (0)	100–100	0.151	< 0.001	–	0.44
Horizontal Right 5000°/s²	97.22 ± 5.62	100 (2)	82–100	98.77 ± 4.70	100 (0)	80–100	0.380	< 0.001	< 0.001	0.30
Horizontal Right 6000°/s²	98.44 ± 3.11	100 (1)	90–100	99.55 ± 1.88	100 (0)	92–100	0.163	< 0.001	–	0.42
Horizontal Left 4000°/s²	98.55 ± 4.21	100 (1)	86–100	98.61 ± 5.89	100 (0)	75–100	0.598	< 0.001	–	0.01
Horizontal Left 5000°/s²	98.56 ± 4.54	100 (0)	82–100	99.27 ± 3.06	100 (0)	87–100	0.553	< 0.001	–	0.19
Horizontal Left 6000°/s²	96.22 ± 8.81	100 (3)	67–100	98.88 ± 4.71	100 (0)	80–100	0.171	< 0.001	–	0.38
RALP Right 3000°/s²	98.88 ± 4.71	100 (0)	80–100	100.00 ± 0.00	100 (0)	100–100	0.317	–	–	0.33
RALP Right 4000°/s²	97.66 ± 7.79	100 (3)	67–100	100.00 ± 0.00	100 (0)	100–100	0.075	< 0.001	–	0.42
RALP Right 5000°/s²	95.83 ± 6.01	98 (6)	81–100	98.33 ± 4.56	100 (2)	82–100	0.128	< 0.001	< 0.001	0.46
RALP Right 6000°/s²	94.11 ± 9.41	100 (8)	67–100	96.11 ± 9.65	100 (3)	67–100	0.222	< 0.001	< 0.001	0.20
RALP Left 3000°/s²	98.88 ± 3.30	100 (1)	88–100	100.00 ± 0.00	100 (0)	100–100	0.151	< 0.001	–	0.48
RALP Left 4000°/s²	98.22 ± 5.21	100 (3)	82–100	99.33 ± 2.82	100 (0)	88–100	0.509	< 0.001	–	0.26
RALP Left 5000°/s²	94.66 ± 6.85	98 (8)	80–100	97.47 ± 4.47	100 (3)	84–100	0.281	< 0.001	< 0.001	0.57
RALP Left 6000°/s²	93.11 ± 0.75	98 (6)	78–100	97.38 ± 6.29	100 (4)	78–100	0.281	< 0.001	< 0.001	0.56
LARP Right 3000°/s²	99.00 ± 3.08	100 (0)	88–100	100.00 ± 0.00	100 (0)	100–100	0.486	< 0.001	–	0.46
LARP Right 4000°/s²	98.94 ± 3.43	100 (0)	86–100	99.22 ± 3.29	100 (0)	86–100	0.281	< 0.001	–	0.08
LARP Right 5000°/s²	96.61 ± 6.92	100 (4)	81–100	97.89 ± 6.51	100 (2)	80–100	0.598	< 0.001	< 0.001	0.19
LARP Right 6000°/s²	97.55 ± 5.11	100 (2)	80–100	98.88 ± 4.71	100 (0)	80–100	0.598	< 0.001	–	0.27
LARP Left 3000°/s²	98.44 ± 4.57	100 (0)	84–100	100.00 ± 0.00	100 (0)	100–100	0.281	< 0.001	–	0.48
LARP Left 4000°/s²	94.72 ± 12.32	100 (8)	50–100	99.22 ± 3.29	100 (0)	86–100	0.423	< 0.001	–	0.50
LARP Left 5000°/s²	96.66 ± 7.55	100 (4)	78–100	98.00 ± 5.98	100 (2)	78–100	0.196	< 0.001	< 0.001	0.49
LARP Left 6000°/s²	93.33 ± 7.97	98 (8)	80–100	95.55 ± 7.43	98 (6)	74–100	0.247	< 0.001	< 0.001	0.29

SVA Static Visual Acuity, fHIT Functional Head Impulse Test, CG Control Group, SG, Study Group, SD Standard Deviation, IQR Interquartile Range, FDR False Discovery Rate

Table 2.

Comparison of ST and CBT results between the CG and SG

Parameter	Control Group Mean ± SD	Median (IQR)	Min–Max	Study Group Mean ± SD	Median (IQR)	Min–Max	FDR-adjusted p	Shapiro–Wilk p (CG)	Shapiro–Wilk p (SG)	Effect size (d)
Stroop I duration	8.46 ± 0.61	8.45 (0.68)	7.21–9.64	7.95 ± 0.68	7.95 (1.34)	6.69–9.17	0.566	0.838	0.763	-0.79
Stroop I error	0.00 ± 0.00	0 (0)	0–0	0.00 ± 0.00	0 (0)	0–0	1.000	–	–	0.00
Stroop I correction	0.00 ± 0.00	0 (0)	0–0	0.00 ± 0.00	0 (0)	0–0	1.000	–	–	0.00
Stroop II duration	8.09 ± 0.56	8.18 (0.63)	6.89–9.10	7.59 ± 0.84	7.51 (0.85)	6.19–9.40	0.386	0.918	0.990	-0.35
Stroop II error	0.00 ± 0.00	0 (0)	0–0	0.00 ± 0.00	0 (0)	0–0	1.000	–	–	0.00
Stroop II correction	0.00 ± 0.00	0 (0)	0–0	0.00 ± 0.00	0 (0)	0–0	1.000	–	–	0.00
Stroop III duration	11.31 ± 1.22	11.60 (1.52)	8.18–12.84	10.15 ± 1.56	10.26 (2.36)	7.44–12.90	0.070	0.080	0.826	-0.83
Stroop III error	0.39 ± 0.61	0 (1)	0–2	0.11 ± 0.47	0 (0)	0–2	0.106	< 0.001	–	-0.51
Stroop III correction	0.44 ± 0.62	0 (1)	0–2	0.39 ± 0.50	0 (1)	0–1	0.994	< 0.001	–	-0.10
Stroop IV duration	14.33 ± 1.66	14.80 (1.83)	9.29–16.21	12.61 ± 1.80	12.23 (2.36)	9.27–15.90	0.100	0.005	0.935	-1.00
Stroop IV error	0.94 ± 0.54	1 (0)	0–2	0.28 ± 0.46	0 (1)	0–1	0.006	< 0.001	–	-1.59
Stroop IV correction	0.94 ± 0.54	1 (0)	0–2	0.33 ± 0.49	0 (1)	0–1	0.011	< 0.001	–	-1.19
Stroop V duration	18.42 ± 1.27	18.68 (2.02)	15.90–20.54	15.80 ± 2.02	15.27 (3.94)	13.37–19.20	0.006	0.320	0.053	-1.55
Stroop V error	1.17 ± 0.79	1 (1.25)	0–2	0.83 ± 0.86	1 (2)	0–2	0.386	0.002	0.001	-0.41
Stroop V correction	0.94 ± 0.73	1 (1.25)	0–2	0.83 ± 0.79	1 (1)	0–3	0.681	0.003	0.001	-0.15
Corsi block span (forward)	5.67 ± 0.69	6 (1)	5–7	7.11 ± 0.83	7 (2)	6–8	0.006	0.001	0.001	1.89
Corsi block span (backward)	4.00 ± 0.69	4 (0.5)	3–5	5.83 ± 0.86	6 (1.25)	4–7	0.006	0.002	0.023	2.36

SD Standard Deviation, IQR Interquartile Range, Min–Max Minimum–Maximum, FDR False Discovery Rate ; p < 0.05: statistically significant

Results

This study evaluated the results of SVA, fHIT, ST, and CBT to assess and compare performance differences between the study and CGs in detail.

A total of 36 participants were included in the study, comprising 18 e-athletes and 18 controls. The majority of participants were male (72.2%), with a slightly higher proportion of females in the CG. Most participants were university students (50%), followed by university graduates and master’s graduates. The mean age of e-athletes was 20.28 ± 3.08 years, while that of the CG was 21.11 ± 2.17 years. The age range was similar across groups (17–27 for e-athletes; 18–27 for controls). Median ages were 19.5 (IQR = 6.0) and 20.5 (IQR = 3.25), respectively. GE data were reported only in the e-athlete group, with a mean of 7.00 ± 0.51 years. The median was 6.00 years (IQR = 3.25), and the range was 5 to 12 years Within the e-athlete cohort, the distribution of primary game genres was FPS (n = 8, 44.4%), MOBA (n = 6, 33.3%), and Battle Royale (n = 4, 22.2%); genre information was not applicable to controls (Table 3). Given the limited per-genre sample sizes and resultant low statistical power, genre-level analyses were restricted to descriptive statistics and no between-genre inferential tests were performed to avoid unstable estimates and inflated Type I error. Genre information was collected solely to describe the characteristics of the e-athlete group and was not intended for inferential comparison; therefore, only descriptive values were reported.

Table 3.

Participant demographics and descriptive statistics by group

Variable	Category	E-athletes	Control	Total
Gender, n (%)	Male	14	12	26
	Female	4	6	10
Education, n	Univ. Student	10	8	18
	Univ. Graduate	4	7	11
	Master’s Graduate	4	3	7
Age, Mean ± SD		20.28 ± 3.08	21.11 ± 2.17	–
Age, Median (IQR)		19.5 (6.0)	20.5 (3.25)	–
Age Range		17–27	18–27	–
Gaming Experience (GE)		(years)
GE – Mean ± SD		7.00 ± 0.51	–	–
GE – Median (IQR)		6.00 (3.25)	–	–
GE – Range		5–12	–	–
Game genre (E-athletes only), n (%)	MOBA	6 (33.3)	–	–
	FPS	8 (44.4)	–	–
	Battle Royale	4 (22.2)	–	–

SD Standard Deviation, IQR Interquartile Range, GE Gaming Experience, Univ. University. The symbol “–” indicates “not applicable. Game genre data were recorded only for e-athletes (n = 18); percentages are within the e-athlete cohort

SVA analysis revealed that the mean visual acuity in the CG was 0.32 ± 0.94, whereas in the SG, it was 0.27 ± 0.80. The difference between the groups was not statistically significant (p = 0.160), indicating similar levels of performance between the two groups in terms of SVA, with no significant alterations observed (Table 1).

fHIT results were used to evaluate the vestibular function by measuring accuracy rates corresponding to head movements at various planes and velocities. In the horizontal plane, tests conducted at 4000°/s² to the right demonstrated an accuracy rate of 98.44 ± 4.97 in the CG, compared to 100% accuracy in the SG. The difference was not statistically significant (p = 0.151). At 5000°/s², the CG had an accuracy rate of 97.22 ± 5.62 while the SG recorded 98.77 ± 4.70, showing no significant difference (p = 0.380). At 6000°/s², accuracy rates were 98.44 ± 3.11 (range: 90–100) in the CG and 99.55 ± 1.88 in the SG, with no statistically significant difference observed (p = 0.163) (Table 1).

In the leftward direction within the horizontal plane, tests at 4000°/s² yielded accuracy rates of 98.55 ± 4.21 in the CG and 98.61 ± 5.89 in the SG, with no significant difference detected (p = 0.598). At 5000°/s², the CG’s accuracy was 98.56 ± 4.54 compared to 99.27 ± 3.06 (range: 87–100) in the SG (p = 0.553). At 6000°/s², the CG achieved an accuracy rate of 96.22 ± 8.81 (range: 67–100), while the SG showed 98.88 ± 4.71, without statistically significant differences (p = 0.171).

RALP (Right Anterior Left Posterior) fHIT results showed that at 3000°/s² to the right, the CG’s accuracy was 98.88 ± 4.71 compared to 100% in the SG; this difference was not statistically significant (p = 0.317). At 4000°/s², accuracy rates were 97.66 ± 7.79 in the CG and 100% in the SG, with no significant difference (p = 0.075). At 5000°/s², the CG demonstrated an accuracy of 95.83 ± 6.01 compared to 98.33 ± 4.56 in the SG, with no significant difference (p = 0.128). At 6000°/s², accuracy rates were 94.11 ± 9.41 in the CG and 96.11 ± 9.65 in the SG, and this difference was not statistically significant (p = 0.222).

In the left RALP plane, at 3000°/s², the CG had an accuracy of 98.88 ± 3.30, compared to 100% in the SG, with no significant difference observed (p = 0.151). At 4000°/s², the CG demonstrated an accuracy of 98.22 ± 5.21, while the SG showed 99.33 ± 2.82, without a statistically significant difference (p = 0.509). At 5000°/s², accuracy rates were 94.66 ± 6.85 in the CG and 97.47 ± 4.47 in the SG (p = 0.281). At 6000°/s², accuracy was 93.11 ± 0.75 in the CG and 97.38 ± 6.29 in the SG, and the difference was not statistically significant (p = 0.281).

LARP (Left Anterior Right Posterior) fHIT assessments revealed that at 3000°/s² to the right, the CG had an accuracy of 99.00 ± 3.08, compared to 100% in the SG (p = 0.486). At 4000°/s², the accuracy was 98.94 ± 3.43 in the CG and 99.22 ± 3.29 in the SG (p = 0.281). At 5000°/s², the CG achieved an accuracy of 96.61 ± 6.92 compared to 97.89 ± 6.51 (range: 80–100) in the SG (p = 0.598). At 6000°/s², accuracy rates were 97.55 ± 5.11 in the CG and 98.88 ± 4.71 in the SG (p = 0.598). In the left direction, at 3000°/s², the CG recorded 98.44 ± 4.57 and the SG 100% (p = 0.281). At 4000°/s², the CG had an accuracy of 94.72 ± 12.32 while the SG achieved 99.22 ± 3.29, and the difference remained statistically non-significant (p = 0.423). At 5000°/s², accuracy was 96.66 ± 7.55 in the CG and 98.00 ± 5.98 in the SG fiugre(p = 0.196). Finally, at 6000°/s², the CG recorded 93.33 ± 7.97 and the SG 95.55 ± 7.43, with no statistically significant difference (p = 0.247). Overall, fHIT results indicated no statistically significant differences in vestibular function between the SG and CG, suggesting comparable vestibular system performance under the test conditions.

ST results assessed cognitive processing speed, attentional control, and inhibition. Stroop III duration was shorter in the SG (10.15 ± 1.56 s) compared to the CG (11.31 ± 1.22 s), but this difference did not reach statistical significance (p = 0.070). Stroop IV duration was also reduced in the SG (12.61 ± 1.80 s) compared to the CG (14.33 ± 1.66 s), though the difference was not statistically significant (p = 0.100). Stroop V duration showed a statistically significant improvement in the SG, with a mean time of 15.80 ± 2.02 s versus 18.42 ± 1.27 s in the CG (p = 0.006).

The error count in Stroop IV was significantly lower in the SG (0.28 ± 0.46) compared to the CG (0.94 ± 0.54; p = 0.006). Furthermore, the number of corrections made in Stroop IV was also significantly reduced in the SG (0.33 ± 0.49) compared to the CG (0.94 ± 0.54; p = 0.011) (Table 2).

CBT results evaluated visual-spatial memory capacity. The forward block span score was significantly higher in the SG (7.11 ± 0.83) compared to the CG (5.67 ± 0.69; p = 0.006). Similarly, the backward block span score was also significantly greater in the SG (5.83 ± 0.86; range: 4–7) than in the CG (4.00 ± 0.69; p = 0.006). These findings indicate that the SG exhibited superior visual-spatial memory capacity relative to the CG (Table 2). In summary, the SG demonstrated statistically significant improvements in cognitive performance, attentional control, and memory capacity compared to the CG. However, fHIT assessments revealed no significant differences in vestibular function between the groups.

Discussion

The Discussion section has been reorganized to provide (1) a concise summary of the main findings, followed by (2) relevant comparisons with previous literature, (3) physiologic and mechanistic interpretations of the results, and (4) practical implications for training and future research.

The present study compared functional VOR responses, visuospatial memory reaction time and short-term memory (executive functions) between e-athletes and non-e-athletes using the fHIT SVA assessments CBT and ST. Although the results indicate some cognitive advantages among e-athletes, particularly in response inhibition and visuospatial memory, these findings were limited in scope and should be interpreted cautiously. No significant group differences were observed in vestibular function or static visual acuity, suggesting a possible equivalence in those domains. To better understand these outcomes, it is essential to compare these findings with existing literature and explore the underlying reasons for these cognitive and sensory-motor effects.

Argyriou et al. [15] and Kronenberger et al. [16] reported that extensive gaming might impair response inhibition and other aspects of executive function. In contrast, Bediou et al. [17] suggested that gamers often excel in tasks requiring rapid decision-making, attentional control, and response inhibition, implying that gaming could serve as cognitive training for these skills. More recently, Gao et al. [18] highlighted improvements in cognitive flexibility and attentional skills in e-athletes under pressure. Likewise, Green and Bavelier [25] demonstrated that action video game players develop enhanced attentional control due to the complex visual and cognitive demands of gaming environments. These discrepancies could stem from differences in game genres, the intensity of gameplay, and individual cognitive baselines. For instance, action games, which require players to manage multiple tasks, suppress automatic responses, and adapt to rapidly changing scenarios, may enhance executive function over time. Conversely, games that do not challenge these skills or involve repetitive, less cognitively demanding tasks might not yield the same benefits and could even lead to negative cognitive effects. Further analysis of the ST results in our study confirmed that e-athletes exhibited superior attention control, cognitive flexibility, and information processing speed compared to the CG. In particular, during the Stroop IV and V tasks, e-athletes completed the tests more quickly and with fewer errors, highlighting their enhanced ability to make rapid decisions and manage distractions effectively. These results are consistent with those of Cyma-Wejchenig et al. [31], who also observed cognitive advantages in e-athletes despite their relatively low levels of physical activity. Comparisons with physically active athletes were included only to contextualize mechanistic interpretations from prior literature. Because physical activity level was not assessed or used as an inclusion criterion in our sample, these comparisons should not be interpreted as indicating group equivalence.

The results of the CBT revealed that e-athletes had greater visuospatial memory capacity compared to the CG. Their ability to quickly adapt to changing visual information during gameplay may underlie this advantage, as such environments demand continuous updating and flexible allocation of attentional resources. Furley and Memmert [32] reported that team sports athletes, who are constantly required to process dynamic and unpredictable stimuli, exhibit superior visuospatial memory compared to athletes engaged in less variable contexts. This finding parallels the enhanced performance observed in e-athletes in our study. Similarly, Russo et al. [33] emphasized that sports requiring open skills foster greater development in cognitive flexibility and visual processing compared to closed-skill sports. Given that professional gaming environments share characteristics with open-skill sports namely unpredictability, rapid decision-making, and the need to simultaneously monitor multiple cues our findings may reflect the transfer of these demands into enhanced visuospatial memory capacity. Together, these studies suggest that both traditional team sports and competitive gaming place athletes in cognitively demanding conditions that promote the development of advanced visuospatial working memory skills.

No significant differences were observed between e-athletes and the CG in the fHIT, which evaluates the ability to stabilize vision during head movements. Romano et al. [24] demonstrated that vestibular functions vary by sport, with correct response rates decreasing as head movement speed increases. Similarly, Kızılay and Cengiz [34] found differences in fHIT performance between combat sports athletes and ball game players, with the latter group demonstrating higher performance. However, this difference was not observed in our study. Our findings are consistent with these results, as no significant differences in vestibular function were observed between the groups.

Our study found no significant differences in functional VOR responses between e-athletes and non-e-athletes, suggesting that prolonged gaming does not negatively impact vestibular function. This aligns with Romano et al. [24], who demonstrated that professional gamers maintain normal VOR performance despite high cognitive and visual demands. The resilience of the VOR observed in e-athletes indicates that the neural mechanisms responsible for gaze stabilization are robust even in conditions where intense visual processing dominates but physical movements are minimal.

The absence of between-group differences in fHIT performance may reflect the stability of cerebellar-mediated VOR gain, which is primarily shaped by vestibular rather than purely visual stimulation. Cerebellar plasticity underlying VOR adaptation occurs mainly within the flocculus–paraflocculus circuitry and requires error-driven modulation through repeated high-velocity head movements [35, 36]. As summarized by Tabata and Kano [37], these synaptic mechanisms are not robustly engaged in conditions where the vestibular system is minimally challenged. Since e-sports rely predominantly on visually guided, seated, and low-acceleration tasks, the vestibular–cerebellar pathways receive insufficient stimulus to induce measurable adaptive changes, which may explain the lack of group differences observed in our fHIT results.

In contrast, studies on athletes in physically demanding sports such as football and handball indicate that frequent multidirectional head movements promote vestibular adaptation, leading to improved VOR gain and gaze stabilization [19, 20]. Romano et al. [24] also highlighted that athletes involved in such sports exhibit superior VOR performance due to the dynamic vestibular stimulation inherent in their activities. This suggests that VOR performance can be enhanced when head and body movements require constant vestibular engagement, a feature that is absent in e-sports. Because empirical studies on vestibular and visuomotor function in e-athletes remain limited, literature from physically active athletes is referenced solely for mechanistic context rather than inferential comparison.

In the assessment of SVA, no statistically significant differences were found between e-athletes and the CG (p = 0.160). These results are broadly consistent with prior evidence indicating that between-group differences are more likely to emerge on dynamic rather than chart-based static measures of visual acuity [38]. By contrast, Rosenfield [39] highlighted that prolonged screen use is associated with visual fatigue and reduced acuity, a condition often described as Computer Vision Syndrome, thereby pointing to potential risks in contexts of excessive or poorly managed visual load. More recently, Kim et al. [40] found no significant differences in eye–hand coordination and SVA between professional e-athletes and amateurs, while noting that experts may still show advantages in other perceptual domains such as peripheral vision. Taken together, these findings indicate that static visual acuity itself may remain unaffected by intensive gaming, whereas individual variability, exposure patterns, and the broader spectrum of visual skills determine whether gaming exerts neutral, beneficial, or detrimental effects. The discrepancy across studies could be explained by adaptive mechanisms of the visual system, whereby continuous engagement with dynamic and unpredictable visual scenes conditions players to process rapidly changing stimuli without necessarily altering baseline acuity. Such adaptations may also counteract the negative outcomes associated with screen fatigue, supporting the view that the visual impact of gaming is more nuanced than commonly assumed.

Although e-sports require exceptional visuomotor coordination and cognitive skills, the limited physical stimulation likely explains why our study did not observe significant differences in VOR performance between the groups. The lack of vestibular demand in gaming settings means that e-athletes are not exposed to the same vestibular challenges that traditional athletes face, thereby preventing any notable enhancement in VOR function. However, as suggested by Ong et al. [41] in their study on NFL officials, integrating vestibular training into e-sports routines could be beneficial. VOR is critical for maintaining visual stability during rapid head movements, and targeted vestibular exercises may improve both reflexive head-eye coordination and overall gaming performance. Given the high cognitive demands of e-sports, such training might provide an edge in maintaining focus and precision during gameplay, helping e-athletes to better manage rapid visual and motor responses. This opens up a potential area for future research where the effects of vestibular training on e-sports performance could be explored more thoroughly. Overall, the limited number of significant findings suggests that cognitive and vestibular profiles of e-athletes and controls may be more similar than initially anticipated. The observed differences in cognitive flexibility and visuospatial memory warrant further investigation but cannot be generalized without caution. These findings highlight the importance of refining task sensitivity and ensuring robust group comparability in future studies.

Limitations

This study provides preliminary evidence that e-athletes may exhibit modest advantages in certain cognitive domains such as response inhibition and visuospatial working memory. However, no significant group differences were observed in vestibular function or static visual acuity. These findings should be interpreted with caution due to the limited sample size and potential confounding factors. Future studies with larger, demographically matched cohorts and longitudinal designs are warranted to further clarify the cognitive and sensorimotor profiles associated with competitive gaming.

This study has several limitations that should be acknowledged. The overall sample size was relatively small, and recruitment was constrained by the training schedule and availability of participants at the municipal e-sports center. These practical restrictions also limited the ability to obtain balanced representation across different gaming genres, and therefore genre-specific analyses could not be performed. The cross-sectional nature of the study further prevents causal inference regarding the long-term effects of competitive gaming on cognitive or vestibular function.

The use of a non-probability convenience sample and the absence of formal group matching may have introduced differences in unmeasured variables such as screen time, habitual physical activity, or gaming genre specialization. Additionally, because physical activity level was not assessed, comparisons drawn from literature on physically active athletes are presented only as mechanistic context rather than direct group equivalence.

Another important limitation is that visual fatigue, screen exposure, and physical activity levels were evaluated through structured anamnesis rather than validated quantitative instruments (e.g., IPAQ, Visual Fatigue Scale). Although this approach was necessitated by the field-based setting and lack of portable vestibular equipment, it may not fully capture the variability of these factors. Likewise, objective vestibular assessments such as VNG could not be conducted because the available system was not portable.

Moreover, empirical studies examining vestibular and visuomotor function specifically in e-athletes are scarce, which limits the ability to contextualize our findings within existing literature.

Despite these limitations, the study provides preliminary evidence that e-athletes may exhibit modest advantages in certain cognitive domains, whereas vestibular function appears comparable to non-athletes. Future research with larger, demographically matched cohorts, objective assessments, and longitudinal designs is needed to clarify the interaction between cognitive performance and vestibular function in competitive gaming.

Authors’ contributions

Gül ÖLÇEK: Data curation, Project administration, Supervision Writing – original draft. Visualization. Yuşa BAŞOĞLU: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft, Writing – review & editing.İlayda Çelik BAŞOĞLU: Data curation, Formal analysis, Investigation, Methodology, Visualization, Writing – original draft. Sude TOMAÇ: Data curation, Visualization All authors read and approved of the final manuscript.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Non-Invasive Clinical Research Ethics Committee of İstanbul Medipol University (Decision No: 855/2021) and conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants before participation.

Consent for publication

Written informed consent for the publication of identifying images and data was obtained from all participants. The signed consent forms have been securely stored and uploaded to the journal submission system in accordance with ethical publication standards.

Competing interests

The authors declare no competing interests.