The Role of Gaze Behavior in Basketball Shooting Performance: A Systematic Review and Meta-Analysis of Elite and Near-Elite Players

Abstract

(1) Background: This systematic review and meta-analysis aimed to assess the reliability and consistency of the impact of gaze behaviors (e.g., QE duration, FFD, FD) on basketball shooting performance. (2) Methods: Searches were conducted in PubMed, Web of Science (WOS), SPORTDiscus, and Scopus, covering all records from each database’s inception. Overall, 17 studies were included examining the relationship between the gaze behavior of basketball players and shooting performance and the studies The quality of the studies was assessed using the McMaster Critical Review Form for Quantitative Studies. A meta-analysis was performed using R 4.4.2 software. (3) Results: Longer QE duration (r = 0.67, p = 0.04) and FFD (r = 0.55, p = 0.03) were positively correlated with SA, while FD showed no significant association (p = 0.82). Elite players exhibited significantly longer QE durations than near-elite players (p = 0.01), but no significant differences were observed in FFD (p = 0.12) or FD (p = 0.18) between competition levels. Meta-regression analysis confirmed a significant positive relationship between QE duration and SA in different competition levels (p = 0.0143). (4) Conclusions: The results of this study highlight the importance of QE duration in basketball shooting performance, with elite players showing better gaze control.

1. Introduction

Basketball is an intermittent high-intensity sport in which players combine physical, technical, tactical, and psychological attributes for optimal performance. Players are required to execute a wide range of actions, from shooting and passing to defending and rebounding, all while under significant pressure from defenders and time restrictions [1]. Among these actions, shooting is particularly critical given that it directly determines the game’s score and, thus, the likelihood of winning. Shooting a basketball is a complex action that requires the player to process gaze behavior stimuli, engage in cognitive processing, and initiate an appropriate motor response in real time [2]. This process involves integrating environmental information, such as the positions of teammates, opponents, and the basket, while coordinating visual processing and motor skills to execute the shot effectively [2]. Consequently, shooting ability is often a key focus in player development strategies and performance analysis within basketball teams.

Gaze behaviors play a crucial role in the shooting process, covering a series of key indicators such as fixations (e.g., stationary gaze at a target) and saccades (e.g., rapid eye movements from one gaze fixation to another) [3,4]. Specifically, the quiet eye (QE) duration is defined as the time from the onset of fixation stability to the end of the first observable movement in the shooting motion [5,6,7]. QE duration is relevant for both open- and closed-loop control processes, as it aids in action pre-programming before execution in the former and assists in real-time movement adjustments in the latter [5,6,7]. For example, the free throw scenario is associated with improved shooting accuracy due to the role of quiet eyes in the pre-programmed action [6], especially in high-pressure situations [7,8]. As such, studies have established a positive correlation between QE duration and shooting accuracy [4].

Since 2002, research has emphasized that Final Fixation Duration (FFD) also plays a crucial role in the shooting process [9], which refers to the time spent visually fixating on the target immediately before initiating the shooting motion, serving as a critical moment of attentional focus [10]. Compared to QE duration, FFD emphasizes the finalized visual input just before shot execution, including the final gaze adjustments made after the shooting action has already begun [9]. It has been found that the FFD can be significantly affected by shooting style [9,11]. Additionally, recent research has demonstrated that the FFD impacts shooting performance differently across various groups [8].

In contrast, fixation duration (FD) represents the cumulative time that relevant cues encompass overall visual engagement during a task, reflecting visual search strategies and attentional allocation [3,8,12,13,14]. For example, basketball players rely on their vision to predict the trajectory of the ball and identify scoring opportunities, using FD to optimize their anticipatory and decision-making processes [5,10,13]. Similarly, this concept extends to other sports, such as badminton, which involve modulating the dynamic nature of visual attention and fixation strategies under rapid and complex situational changes [15].

Basketball players’ ability to focus on relevant cues and ignore distractions has been shown to improve their shooting accuracy and overall performance [16]. Research has confirmed that cognitive factors, alongside physical attributes, are critical determinants of player performance [17]. Central to this cognitive framework is the visual system, which processes dynamic environmental inputs through three key functions: (1) Tracking moving targets (e.g., predicting ball trajectories) [9], (2) assessing spatial relationships (e.g., defensive positioning) [10], and (3) rapidly scanning visual fields for tactical information [6]. In fast-paced basketball contexts, athletes with enhanced visual–cognitive integration demonstrate superior information processing speeds [18]. Crucially, these visual–cognitive capacities are trainable. This evidence establishes visual processing not merely as sensory input, but as a measurable component of sports intelligence, necessitating its evaluation in comprehensive athlete selection models [19].

In this regard, experienced athletes use advanced tracking and fixation techniques that significantly improve their shooting accuracy [7,11,19]. Research has suggested that elite basketball players—namely, those competing at the professional or international level—demonstrate superior cognitive and gaze behavior abilities compared to near-elite players (e.g., national or university-level athletes) [13]. Players with longer and more stable QE durations are more likely to exhibit superior shooting accuracy, making QE a strong indicator of elite performance [5,20]. Additionally, FD and FFD reveal differences in visual processing efficiency, where elite players exhibit longer, more stable fixations, while near-elite players struggle with fixation inconsistency and visual overload [13]. Given these distinctions, training programs for near-elite players should emphasize fixation control, gaze stability, and anticipatory processing to help them develop more effective visual attention strategies [21]. Furthermore, analyzing FD and FFD patterns can assist coaches in designing targeted interventions to enhance visual processing speed and shooting consistency, ultimately improving overall performance.

Understanding how gaze behavior diverges across competition levels provides critical insights for the development and selection of athletes [20]. Player performance can be improved by identifying clear distinctions between elite and near-elite players, training methodologies, and developing evidence-based interventions to enhance gaze behavior in the context of basketball shooting [8]. Given the increasing reliance on sports science in elite basketball, understanding how gaze behaviors differentiate elite from near-elite players is essential for refining training strategies, performance diagnostics, and player selection criteria [5,10].

In summary, this systematic review and meta-analysis aimed to assess the reliability and consistency of assessments of the impacts of gaze behaviors (i.e., QE duration, FFD, FD) on basketball shooting performance. By synthesizing findings from studies that quantitatively evaluate gaze behaviors during shooting tasks in competitive basketball settings, this review identified how these gaze behavior mechanisms contribute to shooting accuracy. To ensure methodological rigour, only original research studies involving able-bodied elite and near-elite basketball players were included in competitive contexts. Studies involving psychological interventions, wheelchair basketball, noncompetitive players, or broader cognitive factors beyond perceptual gaze behavior were excluded.

2. Materials and Methods

2.1. Search Strategy and Study Selection

This study was developed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines [22]. Multiple databases were searched to retrieve relevant studies, including PubMed, Web of Science (WOS), SPORTDiscus, and Scopus. The following search string was used to retrieve studies: (1) “basketball” OR “basketball players” AND (2) “selection” AND (3) “gaze behavior” OR “quiet eye” OR “QE duration” OR “final fixation duration” AND (4) “shooting performance” OR “shooting accuracy” AND (5) “visual attention” OR “focus” OR “visual control”.

Database searches were initially conducted on 12 November 2022, and an updated search was completed on 13 October 2024. The searches covered all available records from their inception up to the respective search dates, ensuring a comprehensive review of the relevant literature. In addition to database searches, manual reference list checks and searches in sports research portals and organizational websites were performed to identify articles not indexed in academic databases and cited studies related to the research topic.

To streamline the screening process, Rayyan software (Rayyan for Systematic Reviews, Qatar Computing Research Institute, Doha, Qatar, https://www.rayyan.ai, accessed on 27 March 2025) was used to collate search results from all databases into a unified spreadsheet, automatically remove duplicate studies, and facilitate blinded independent screening. The titles and abstracts of the retrieved studies were first screened by H.M., who assessed them against the inclusion and exclusion criteria. Full-text versions of studies that met the initial inclusion criteria were then retrieved and independently reviewed by both H.M. and J.L. to determine their final eligibility. Each researcher flagged studies in the Rayyan software as included, excluded, or possibly included.

Whenever conflicts arose regarding the inclusion decisions, they were resolved through a structured process, first through discussion between H.M. and J.L. If disagreements persisted, a third independent researcher (J.A.) was consulted to make the final decision.

2.2. Inclusion and Exclusion Criteria

2.2.1. Inclusion Criteria

Inclusion criteria were established following the ‘PICOS’ framework [23], including (1) “Population”: Basketball athletes classified as elite (professional/international) or near-elite (regional/university) based on the framework of McKay et al. (2021) [24] (Table 1) (2) “Interventions”: Studies measuring gaze behavior (QE duration, FFD, FD) (Table 2) during shooting tasks; (3) “Comparisons”: studies that compared gaze behavior and shooting performance between groups based on their competition level (elite, near-elite athletes) [24]; (4) “Outcome”: shooting performance, measured through specific indicators such as shooting accuracy and gaze behavior; and (5) “Study design” (S): no constraints on design were imposed [25], including at least the corresponding participants, gaze behavior, and shooting accuracy, without considering intervention changes or post-test experimental data.

Table 1. Basketball player classification framework [24].

Classification	Basketball League Level	Training Frequency
Elite	Competing at the highest level globally in the most exclusive leagues.	6–7 days per week (often multiple sessions per day)
	Competing at the international level, including professional leagues worldwide, reserve players in NBA/WNBA, NCAA Division I players, and national team representatives.	5–7 days per week (structured periodized training)
Near-elite	Competing at the national level, including NCAA Division II and III, US high school varsity teams, and competitive community-based clubs. Participating in recreational basketball leagues, junior varsity teams, and non-professional competitions.	4–6 days per week
Amateur	No strict performance standard, but regular training and intent to compete required.	~3 times per week
Beginner	Playing basketball for fun or fitness without structured competition goals.	1–3 times per week (non-structured)
	Individuals who do not engage in regular basketball activity and do not meet minimum physical activity guidelines.	No regular training

Table 2. Comparison and definitions of gaze behavior metrics in basketball shooting.

Gaze Behavior	Definition	Stage of Action
QE Duration	The time from the onset of gaze stability to the end of the first observable movement in the shooting motion.	Preparation and early execution phase
FFD	The duration of gaze fixation on the target immediately before the shooting motion begins, emphasizing final visual input and adjustments.	Final pre-shot phase
FD	The total time spent fixating on relevant cues throughout the shooting process, reflecting overall visual engagement and attention allocation.	Entire shooting process

2.2.2. Exclusion Criteria

Exclusion criteria were also used when deliberating on the retrieved studies, detailed as follows: (1) Shooting performance assessed during unofficial sporting contexts (e.g., physical education or sport in educational centers), due to their significantly lower intensity, technical complexity, and tactical demands compared to competitive matches [9]; and (2) examination of players who participate in wheelchair basketball or have specific disabilities. The rationale for these criteria was that the performance outcomes (cognitive, perceptual, and physical) differ substantially between disabled and able-bodied athletes due to differences in physiological adaptations and playing styles [26]. For example, wheelchair basketball players may exhibit unique movement patterns, spatial awareness, and decision-making processes [27], as the use of a wheelchair alters the requirements of the sport and, therefore, the performance mechanisms [28]; (3) amateur and beginner athletes‘ data were excluded, as their training frequency (≤3 times per week) [24] and lack of competitive experience do not reflect the gaze behavior characteristics of high-performance athletes [5] (Table 1); (4) outcomes were reported surrounding interventions involving additional influences, such as mental flips, self-efficacy, and time pressure scenarios. These interventions differ from the primary focus of cognitive and perceptual measures in this review by introducing external psychological or situational factors that could impact performance; (5) publications that are editorials, letters to the editor, reviews, conference abstracts, or opinion pieces, rather than original investigations [29]; and (6) studies published in languages other than English.

2.3. Quality of Studies

The study quality was assessed using the McMaster Critical Review of Quantitative Studies Form [30], which is widely used in systematic reviews to assess the methodological rigor and transparency of quantitative studies [31]. This tool provides a comprehensive and structured approach for assessing the key aspects of quantitative research, such as study design, data collection, and statistical analyses, ensuring that each study is critically appraised against standardized criteria. The tool was chosen to ensure consistency and robustness in assessing study quality, as it has previously been validated in systematic reviews within similar fields [25,32].

Two researchers (H.M. and J.L.) evaluated each study according to the 15 following criteria: (1) Objective, (2) Background relevance, (3) Study design, (4) Sample, (5) Sample inclusion, (6) Consent, (7) Outcome measures, (8) Measure validity, (9) Intervention, (10) Result, (11) Analysis, (12) Dropouts, (13) Clinical importance, (14) Conclusion and Practical implications, and (15) Limitations. Each criterion was scored as either ‘1’ (if the requirement was satisfied) or ‘0’ (if the requirement was inadequately satisfied); see Table 1. The total score was used to classify the quality of the study into the following categories: ‘very good’ (scores of 13–15), ‘good’ (scores of 10–12), or ‘poor’ (scores of <10) [25]. In instances where evaluation scores differed between researchers, decisions were finalized via further discussion.

2.4. Data Extraction and Meta-Analysis

Data extraction was performed independently by M.H. using a standardized Excel template and was reviewed by J.L. Key information extracted from the studies included participant characteristics (age, sex, competitive level), task design (consistency of measuring instruments), gaze behavior, shooting performance, and outcome metrics. Competition levels were classified as “elite” or “near-elite”, following the framework of McKay (2021) et al. [24] (Table 2).

Meta-analysis was conducted using R 4.4.2, incorporating the ‘metafor’ and ‘dmetar’ packages. Data were imported from Excel, and standardized mean differences (SMDs) were calculated [25,33]. Additionally, a meta-regression analysis was performed to examine the effect of quiet eye (QE) duration on shooting accuracy across different competition levels. The regression model included an intercept and slope estimate to predict variations in performance. Statistical significance was set at p < 0.05, and effect sizes were interpreted accordingly.

To ensure robust and accurate analyses, data reported with standard errors, no statistical significance, inconsistent units, or without p-values were excluded from the meta-analyses. The data extraction process revealed that gaze behavior metrics (QE duration, FFD, FD) and shooting accuracy were sufficient for conducting meta-analyses. Accordingly, a meta-analysis was conducted where three or more studies reported common associations to ensure the results were sufficiently varied and comparable [34]. Standardized mean differences (SMD) and 95% confidence intervals (CI) were calculated using the mean and standard deviation (SD) reported for variables [34,35].

Heterogeneity across studies within analyses was assessed using the I² statistic and Q test [36]. Heterogeneity was identified as low (I² < 25%), medium (25% ≤ I² < 50%), or high (I² ≥ 50%) [37]. A random-effects model was applied for analyses with I² > 50%, while a fixed-effects model was applied for analyses with I² ≤ 50. Furthermore, the statistical significance for heterogeneity tests was set at p < 0.05 [34].

The I² statistic was calculated as 100% × (Q − df)/Q, where Q is Cochran’s heterogeneity statistic and df denotes the degrees of freedom. Negative I² statistics were interpreted as 0%, which indicates no heterogeneity, with larger values (up to 100%) representing greater heterogeneity in analyses [37].

To further validate the reliability of the data, R software was used to download and clean the dataset through the removal of irrelevant or redundant data points. Additionally, sensitivity and correlation analyses were conducted to ensure that the findings were not driven by a single study, and scatter plots were generated to visually assess data trends. A leave-one-out approach was also applied to evaluate the robustness of QE duration, FFD, and FD impacts on shooting performance [38]. This systematic review and meta-analysis aimed to assess the reliability and consistency of the impacts of gaze behaviors (e.g., QE duration, FFD, FD) on basketball shooting performance. By synthesizing findings from studies that quantitatively assess gaze behaviors during shooting tasks in competitive basketball settings, this review sought to identify how these gaze behaviors contribute to shooting accuracy.

3. Results

3.1. Search, Screening, and Quality Assessment Outcomes

A total of 4487 studies were retrieved across the databases (Web of Science, PubMed, SPORTDiscus, and Scopus; Figure 1). In addition, 49 studies were identified via other methods. Following the removal of duplicate records (n = 1727), a total of 2760 studies remained for screening. In turn, 780 studies were excluded based on their title and abstract, and 1980 full-text articles were sought for retrieval. Of these, 221 records were inaccessible, and 1009 were excluded using Rayyan. A total of 971 studies were excluded based on full-text review. Consequently, 17 studies were included in the systematic review.

Figure 1. Flow diagram showing the search and screening process.

Table 3 provides the quality scores of the 17 included studies, where nine studies were classified as ‘very good’ and eight studies were classified as ‘good’ [25].

Table 3. Quality scores of the included studies.

Study	Criteria															Total	Rating
	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
Czyż et al., 2019 [13]	1	1	1	1	1	1	1	1	1	1	1	1	0	1	1	14	Very good
Harris et al., 2017 [39]	1	1	1	1	1	1	1	1	1	1	1	1	0	0	0	12	Good
Amaro et al., 2023 [40]	1	1	1	1	1	1	1	0	1	1	1	1	0	0	1	12	Good
Klostermann, 2019 [7]	1	1	1	1	1	1	1	1	1	1	1	1	0	1	0	13	Very good
Esteves et al., 2021 [41]	1	1	1	1	1	1	1	1	1	1	1	1	0	0	1	13	Very good
Oudejans et al., 2002 [9]	1	1	1	1	1	1	1	1	1	0	1	1	0	1	0	12	Good
Wilson et al., 2009 [21]	1	1	1	1	1	1	1	1	1	1	1	1	0	0	1	13	Very good
Klostermann et al., 2018 [20]	1	1	1	1	1	1	1	1	1	0	1	1	0	0	1	12	Good
Zhao et al., 2023 [42]	1	1	1	1	1	1	1	1	1	1	1	1	0	0	0	12	Good
Rui et al., 2023 [8]	1	1	1	1	1	1	1	0	1	1	1	1	0	1	1	13	Very good
Vaughan & Laborde, 2021 [43]	1	1	1	1	1	1	1	1	0	1	1	1	0	1	1	13	Very good
Vickers, 1996 [5]	1	1	1	1	1	1	1	1	1	1	1	1	0	0	0	12	Good
Van Maarseveen et al., 2018 [44]	1	1	1	1	1	1	1	1	1	1	1	1	0	1	0	13	Very good
Vickers et al., 2019 [6]	1	1	1	1	1	1	1	1	1	1	1	1	0	1	1	14	Very good
Meyer et al., 2022 [19]	1	1	1	1	1	1	1	1	0	0	1	1	0	1	1	12	Good
Zwierko et al., 2017 [3]	1	1	1	1	1	1	1	1	1	1	1	1	0	0	1	13	Very good
Steciuk & Zwierko, 2015 [45]	1	1	1	1	1	1	1	1	0	0	1	1	0	1	1	12	Good

3.2. General Characteristics of Studies

Seventeen studies were included in the systematic review, with their general characteristics shown in Table 4. Regarding participants, six (35%) studies recruited elite basketball players, four (24%) studies recruited near-elite basketball players, and seven (41%) studies recruited both elite and near-elite basketball players. The sample sizes in the included studies ranged from 6 to 168 basketball players. Regarding participant sex, seven (41%) studies examined only male basketball players, four (24%) studies examined only female basketball players, and six (35%) studies examined both male and female basketball players. Considering the age of the participants, eleven (65%) studies examined adult basketball players (aged >18 years), five (29%) studies examined youth basketball players (i.e., aged between 12 and 17 years), and one (1%) study included both adult and youth basketball players.

Table 4. Overview of studies.

Study	Participants				Equipment	Study Design	Variable		Main Findings
	n	Age (Years)	Sex	Competition Levels			Gaze Behavior	Shooting Performance
Oudejans et al. (2002) [9]	10	27.5 ± 10.5	M	Elite	ETG	Cross-sectional design (CSD)	Final fixation duration (FFD)	Shooting accuracy (SA) (free throw and 3-point)	Full vision improved both FFD and SA (p < 0.001).
Zwierko et al. (2017) [3]	13	20.9 ± 2.4	M	Elite	ETG	Between-subject design	Fixation duration (FD)	Shooting accuracy (SA)	Longer FD leads to higher SA (p < 0.05).
Harris et al. (2017) [39]	18	23.88 ± 7.56	M/F	Elite	ETG	Cross-sectional design (CSD)	QE percentage;	shooting score	QE duration decreased significantly under pressure (p < 0.05).
Amaro et al. (2023) [40]	18	22 ± 3.72	M/F	Elite	ETG	Cross-sectional design (CSD)	Fixation duration (FD) (free-throw and 3-point line)	Shooting score (free throw and 3-point line)	Closer distances led to longer FD and higher SA (p < 0.05).
Esteves et al. (2021) [41]	10	24.8 ± 4.52	M	Elite	ETG	Cross-sectional design (CSD)	Fixation duration (FD)	Shooting accuracy (SA)	Taller players demonstrated better backboard SA (p < 0.05).
Van Maarseveen et al. (2018) [10]	13	16.8 ± 1.8	F	Elite	ETG	Cross-sectional design (CSD)	Final fixation duration (FFD)	Shooting accuracy (SA) (Jump shot)	Defensive pressure reduced FFD and SA (p < 0.05).
Steciuk & Zwierko (2015) [45]	6	22 ± 3	F	Near-elite	ETS	Cross-sectional design (CSD)	Fixation duration (FD), QE duration	Shooting accuracy (SA) (jump shot)	FD were longer for successful shots (p < 0.05).
Wilson et al.(2009) [21]	10	20.3 0.9	M	Near-elite	ETG	Cross-sectional design (CSD)	QE duration	Shooting accuracy (SA)	Pressure significantly reduced QE and SA (p < 0.05).
Klostermann et al. (2019) [7]	16	23.3 ± 6.4	M/F	Near-elite	ETS	Single-case design (SCD)	QE duration	Shooting accuracy (SA)(free-throw)	Longer QE correlated with higher SA (p < 0.05).
Vickers et al. (2019) [6]	12	22.4 ± 2.2	M/F	Near-elite	ETS	Cross-sectional design (CSD)	QE duration	Shooting accuracy (SA)	Longer QE during arm flexion led to higher SA. Defender reduced QE duration and SA.
Vaughan & Laborde (2021) [43]	168	18.9 ± 1.01	M	Elite/Near-elite	Neuropsychological test	Cross-sectional design (CSD)	Rapid visual information, match to sample visual search, spatial span task	Shooting accuracy (SA) (free-throw)	Gaze behavior assessments linked to shooting performance (p < 0.05).
Klostermann et al. (2018) [20]	17	18.8 ± 0.06/17.7 ± 0.2	M	Elite/Near-elite	ETS	Randomized controlled trial (RCT)	QE duration	Shooting accuracy (SA)	Elite had longer QE in defended scenarios, leading to better SA (p < 0.01).
Vickers (1996) [5]	16	22.4 ± 2.2	F	Elite/Near-elite	ETS	Cross-sectional design (CSD)	QE duration	Shooting accuracy (SA)	Elite > near-elite in QE duration, FFD, and SA (p < 0.01).
Zhao et al. (2023) [42]	42	23.6/19.6	F	Elite/Near-elite	ETG	Between-subject design	Fixation duration (FD (Free throw)		Elite > non-elite in FD and SA (p < 0.05).
Czyż et al. (2019) [13]	20	21.8 ± 0.79	M	Elite/Near-elite	ETS, ETG	Randomized controlled trial (RCT)	Fixation Duration (FD);	Shooting accuracy (SA)	Elite > near-elite in FD and SA (p < 0.05). Increase in FD led to higher SA.
Meyer et al. (2022) [19]	31	23 ± 3.13	M/F	Elite/Near-elite	ETG	Cross-sectional design (CSD)	Fixations on the head rather than on the ball, hips, and legs	Shooting accuracy (SA)	Elite had higher SA compared to near-elite (p < 0.05).
Rui et al. (2023) [8]	20	15.2 ± 0.4 27.6 ± 3.7	M/F	Elite/Near-elite	ETG	Cross-sectional design (CSD)	QE duration; final fixation duration (FFD)	Shooting accuracy (SA)	Younger players had shorter QE duration and accuracy than older players (p < 0.05). Youth group showed large differences in gaze behavior and SA and differed from professional players.

Abbreviations: M = male, F = female, n = number, QE = quiet eye, ETG = eye tracking glasses, ETS = eye tracking system 3.3. Description and data analysis of the results.

In terms of tools and measures, the studies used a variety of equipment, including eye tracking systems (ETG and ETS), which were mentioned in 16 (95%), making them the most-used tool, while one (5%) study used neuropsychological tests. In terms of study design, most studies used cross-sectional designs (CSD), which were used in 12 (71%) studies, while randomized controlled trials (RCT) were used in two (12%) studies. Single-case designs (SCD) were used in one (6%) study and a between-subjects design was used in two studies (12%).

The 17 studies (Table 4) were grouped according to their main findings, where three studies reported differences in gaze behavior metrics and shooting performance between groups, 10 studies reported correlations between gaze behavior metrics and shooting performance, and four studies reported each of these types of analysis.

3.2.1. Correlations Between Gaze Behavior and Shooting Performance

Table 5 presents the results of the 14 studies in terms of variables regarding correlations between gaze behavior and shooting performance, with a particular focus on the relationships between key metrics such as QE duration, FFD and FD; and shot scoring and SA. The studies showed that longer QE duration [6,7,20], FD [3,45], and FFD [9,44] were positively correlated with improved shooting accuracy. Three of the articles emphasized the importance of QE duration [6,39] and FFD [44] in terms of shooting performance in defense. A better FD throughout the shooting process can mitigate the effects of differences in shooting distance and physical condition [42,46].

Table 5. Overview of studies reporting the correlations between gaze behaviors and shooting performance.

Study	Participants				Variable				Main Findings
	n	Age (years)	Sex	Competition Level	Gaze Behavior			Shooting Performance
					QE Duration (ms)	FFD (ms)	FD (ms)	SA (%)
Oudejans et al. (2002) [9]	10	27.5 ± 10.5	M	Elite	--	353 ± 34 (no-vision) 357 ± 35 (full-vision)	--	17.5 ± 12.6 (no-vision) 61.5 ± 7.4 (full-vision)	Full vision improved both FFD and SA (p < 0.001).
Zwierko et al. (2017) [3]	13	20.9 ± 2.4	M	Elite	--	--	797.698 ± 508.770 (free throw) 355.547 ± 124.929 (jump shot)	65.38% (free-throw) 48.37% (jump shot)	Free throws had a longer FD and a higher SA (p < 0.05).
Harris et al. (2017) [39]	18	23.88 ± 7.56	M/F	Elite	QE percentage			Shooting score	QE duration decreased significantly under pressure (p < 0.05).
Amaro et al. (2023) [40]	18	22 ± 3.72	M/F	Elite	--	--	707.41 ± 345.45 (free throw) 532.64 ± 322.16 (3-point shot)	--	Closer distances led to longer FD and higher SA (p < 0.05).
Esteves et al. (2021) [41]	10	24.8 ± 4.52	M	Elite	--	--	259.9 ± 153.06 (ball) 273.55 ± 161.96 (rim) 282.95 ± 190.3 (backboard)	84 ± 4.47 (higher) 66.4 ± 6.22 (smaller)	Taller players demonstrated better backboard SA (p < 0.05).
Van Maarseveen et al. (2018) [10]	13	16.8 ± 1.8	F	Elite	--	364 ± 191 (defended) 443 ± 221 (undefended)	--	52.2 ± 8.1 (undefended) 51.3 ± 15 (defended)	Defensive pressure reduced FFD and SA (p < 0.05)
Steciuk & Zwierko (2015) [45]	6	22 ± 3	F	Near-elite	--	--	916 ± 272	--	The FD was longer for successful shots (p < 0.05).
Wilson et al. (2009) [21]	10	20.3 ± 0.9	M	Near-elite	--	--	--	68.6 ± 11.2 (undefended) 50.5 ± 5.07 (defended)	Defended reduced QE and SA (p < 0.05).
Klostermann et al. (2019) [7]	16	23.3 ± 6.4	M/F	Near-elite	764.7 ± 352	--	--	73.1 ± 12.2	Longer QE correlated with higher SA (p < 0.05).
Vickers et al. (2019) [6]	12	22.4 ± 2.2	M/F	Near-elite	Arm flexion: 311.10 ± 13.72 (undefended) 237.02 ± 9.47 (defended) Arm extension 168.58 ± 12.12 (undefended) 136.98 ± 6.73 (defended)	--	--	Undefended: 58% Defended: 50%	Longer QE during arm flexion led to higher SA. Defender reduced QE duration and SA.
Klostermann et al. (2018) [20]	7	17.7 ± 0.2	M	Elite	460.9 ± 45.8 (undefended) 452.4 ± 43.3 (defended)	--	--	67 ± 7 (undefended)	Elite had a longer QE, leading to better SA (p < 0.01).
	10	18.8 ± 0.06		Near-elite	389.5 ± 51.1 (undefended) 431.9 ± 36.8 (defended)	--	--	46.7 ± 8.5 (undefended)
Vickers (1996) [5]	8	21.3 ± 2.5	F	E	972 ± 780	1471 ± 951	--	75 ± 6.3	Elite > near-elite in QE duration, FFD and SA (p < 0.01).
	8	20.8 ± 4.7		NE	357 ± 401	865 ± 595	--	42 ± 28.1
Rui et al. (2023) [8]	N = 20	15.2 ± 0.4	M/F	Near-elite	551 ± 230 (long) 504 ± 238 (middle)	606 ± 195 (long) 564 ± 221 (middle)	--	33.6 ± 13.6 (long) 52 ± 10.5 (middle)	Youth group showed big differences in gaze behavior and SA and differed from elite.
		27.6 ± 3.7		Elite	600 ± 249 (long) 572 ± 233 (middle)	663 ± 243 (Long) 663 ± 245 (Middle)	--	58.8 ± 9.8 (long) 69.2 ± 16.3 (middle)
Czyż et al. (2019) [13]	9	21.8 ± 0.79	M	Elite	--	--	390 ± 50	27 ± 10.85	An increase in FD led to higher SA.
	10	22.7 ± 1.42		Near-elite	--	--	360 ± 60	20 ± 11.3

Abbreviations: M = male, F = female, E = elite player, QE = quiet eye, FFD = final fixation duration, FD = fixation duration, SA = shooting accuracy, Middle = 4.23 m shot distance [8], Long = 6.80 m shot distance [8], Defended = a shooting condition with defensive pressure, Undefended = A shooting condition without defensive pressure, Full-vision = The player has full visual access to the target and surroundings, No-vision = The shooter’s vision is restricted or blocked, Ball = fixation on the basketball during shooting, Rim = fixation on the hoop during shooting, Backboard = fixation on the surface behind the rim during shooting, Higher = shot by taller players, Smaller = shot by shorter players.

Figure 2 and Figure 3 illustrate positive correlations between QE duration and SA, as well as between FFD and SA. The correlation coefficient of 0.67 (p = 0.04 < 0.05; see Table 4) indicates a statistically significant positive relationship between QE duration and SA. Similarly, the correlation coefficient of 0.55 (p = 0.03 < 0.05; Table 6) confirms that FFD is also positively associated with SA.

Figure 2. Scatterplot of correlation between QE duration and shooting accuracy (SA). Note: The yellow regression line represents their linear relationship, with blue dots as data points and the gray shaded area as the confidence interval (CI).

Figure 3. Scatterplot of correlation between fine fixation duration (FFD) and shooting accuracy (SA). Note: The yellow regression line represents their linear relationship, with blue dots as data points and the gray shaded area as the confidence interval (CI).

Table 6. Correlation analysis between different gaze behaviors and SA.

Independent Variable	Dependent Variable	Correlation Coefficient	p-Value
QE Duration	SA	0.67	0.04 *
FFD	SA	0.55	0.03 *
FD	SA	0.67	0.82

Abbreviations: QE = quiet eye, FFD = final fixation duration, FD = fixation, SA = shooting accuracy, * p < 0.05 indicates a statistically significant correlation.

However, Figure 4 demonstrates a non-significant correlation between FD and SA. The plot shows no clear trend, with the fitted line remaining nearly flat, indicating that changes in FD do not substantially impact shooting accuracy. The correlation coefficient of 0.67 (p = 0.82 > 0.05; Table 6) further supports the absence of a statistically significant relationship between FD and SA.

Figure 4. Scatterplot of correlation between fixation duration (FD) and shooting accuracy (SA). Note: The yellow regression line represents their linear relationship, with blue dots as data points and the gray shaded area as the confidence interval (CI).

3.2.2. Differences in Gaze Behavior with Shooting Performance Between Elite and Near-Elite Basketball Players

Table 7 summarizes the seven studies that explored differences in gaze behavior and shooting performance between elite and near-elite basketball players. The findings highlight that elite players tend to have longer QE durations compared to near-elite players, which is positively correlated with improved SA [5,8,20]. Additionally, elite players generally present better FD during shooting [13,42], particularly in free throws [42]. Moreover, the studies indicated that a longer FFD is associated with better shooting performance [5,8]. In terms of shooting accuracy, elite players consistently demonstrated superior performance compared to their near-elite counterparts across the studies [19,43].

Table 7. Summary of studies investigating the relationships between gaze behaviors and shooting performance at different competition levels.

Study	Participants				Variable				Main Findings
	n	Age (Years)	Sex	Competition Level	Gaze Behavior			Shooting Performance
					QE Duration (ms)	FFD (ms)	FD (ms)	SA (%)
Klostermann et al. (2018) [20]	7	17.7 ± 0.2	M	Elite	460.9 ± 45.8 (undefended) 452.4 ± 43.3 (defended)	--	--	67 ± 7 (undefended)	Elite had longer QE, leading to better SA (p < 0.01).
	10	18.8 ± 0.06		Near-elite	389.5 ± 51.1 (undefended) 431.9 ± 36.8 (defended)	--	--	46.7 ± 8.5 (undefended)
Vickers (1996) [5]	8	21.3 ± 2.5	F	Elite	972 ± 780	1471 ± 951	--	75 ± 6.3	Elite > near-elite in QE duration, FFD and SA (p < 0.01).
	8	20.8 ± 4.7		Near-elite	357 ± 401	865 ± 595	--	42 ± 28.1
Czyż et al. (2019) [13]	9	21.8 ± 0.79	M	Elite	---	--	360 ± 60	27 ± 10.85	Elite > near-elite in FD and SA (p < 0.05).
	10	22.7 ± 1.42		Near-elite	---	--	390 ± 50	20 ± 11.3
Zhao et al. (2023) [42]	21	23.6	F	Elite	--	--	629 ± 93 (front area of basket) 563 ± 117 (top area of basket) 319 ± 84 (bottom)	--	Elite > near-elite in FD and SA (p < 0.05).
	21	19.6		Near-elite	--	--	476 ± 56 (front area of basket) 365 ± 50 (top area of basket) 368 ± 57 (bottom)	--
Vaughan & Laborde (2021) [43]	81	18.9 ± 1.01	M	Elite	Rapid Visual Information, Match to Sample Visual Search, Spatial Span Task			53.67 ± 6.37	Gaze behavior assessments linked to shooting performance (p < 0.05).
	87			Near-elite				47.31 ± 6.28
Rui et al. (2023) [8]	20	15.2 ± 0.4	M/F	Near-elite	551 ± 230 (long) 504 ± 238 (middle)	606 ± 195 (long) 564 ± 221 (middle)	--	33.6 ± 13.6 (long) 52 ± 10.5 (middle)	Younger players had shorter gaze behavior and SA than older players (p < 0.05).
		27.6 ± 3.7		Elite	600 ± 249 (long) 572 ± 233 (middle)	663 ± 243 (long) 663 ± 245 (middle)		58.8 ± 9.8 (long) 69.2 ± 16.3 (middle)
Meyer et al. (2022) [19]	15	23 ± 3.13	M/F	Elite	Fixations on the head rather than on the ball, hips, and legs			44.2 ± 13.94	Elite players had higher SA compared to near-elite (p < 0.05).
	16			Near-elite				31.1 ± 14

Abbreviations: M = male, F = female, QE = quiet eye, FFD = final fixation duration, FD = fixation duration, SA = shooting accuracy, Middle = 4.23 m shot distance [8], Long = 6.80 m shot distance [8], Front area of basket = fixation on the front of the hoop, Top area of basket = fixation on the upper part of the hoop, Bottom of basket = fixation on the lower part of the hoop.

The forest plots in Figure 5, Figure 6 and Figure 7 show the QE duration, FFD, and FD at different competitive levels, and the results of the meta-analysis are summarized in Table 8.

Figure 5. Forest plot of the effect sizes (Hedges’ g) for the difference in QE duration between elite and near-elite athletes. Note: Each square represents an individual study’s effect size, with the horizontal lines indicating the 95% confidence interval (CI) [5,8,20].

Figure 6. Forest plot of the effect sizes (Hedges’ g) for the difference in FFD between elite and near-elite athletes. Note: Each square represents an individual study’s effect size, with the horizontal lines indicating the 95% confidence interval (CI) [5,8].

Figure 7. Forest plot of the effect sizes (Hedges’ g) for the difference in FD between elite and near-elite athletes. Note: Each square represents an individual study’s effect size, with the horizontal lines indicating the 95% confidence interval (CI) [13,42].

Table 8. Meta-analysis results comparing elite and near-elite athletes in terms of QE duration, FFD, and FD.

Metric	Number of Studies	Effect Size	z-Value	p-Value	95% CI	I2 (%)	Q-Value	p-Value for Heterogeneity
QE Duration	5	0.58	2.55	0.01 *	0.15, 1.01	0.0%	0.00	0.4471
FFD	3	0.43	1.56	0.12	−0.10, 0.97	0.0%	0.00	0.7860
FD	4	0.98	1.35	0.18	−0.44, 2.41	93.0%	1.9605	<0.0001

Abbreviations: QE duration = quiet eye duration, FFD = final fixation duration, FD = fixation duration. Note: The table presents the number of studies included, effect sizes (Hedges’ g), z-values, p-values (p < 0.05 is significant), 95% confidence intervals (CI), and heterogeneity (I², Q-value, and p-value for heterogeneity), “*” indicates a statistically significant correlation.

For QE duration, the results indicated a statistically significant difference between elite and near-elite athletes (p = 0.01, effect size = 0.58, 95% CI [0.15, 1.01]). The heterogeneity was low (I² = 0%) and the Q-value (0, p = 0.4471) indicated consistency across studies.

In terms of FFD, the effect size was 0.43, with p = 0.12 and 95% CI [−0.10, 0.97], indicating no significant difference between competition levels. The heterogeneity was low (I² = 0%) and the Q-value (0, p = 0.786) confirmed the consistency of the data.

Regarding FD, the effect size was 0.98, but the p = 0.18 and 95% CI [−0.44, 2.41] indicated no statistically significant difference. However, heterogeneity was high (I² = 93%), with a Q-value of 1.9605 (p < 0.0001), suggesting substantial variability among studies’ results.

Additionally, the relationship between QE duration and SA across different competition levels was examined (Table 9). The intercept (−2.0446, p = 0.0244, 95% CI [−3.8249, −0.2644]) indicates that near-elite athletes have a lower baseline SA. The QE duration effect (0.0034, p = 0.0143, 95% CI [0.0007, 0.0060]) suggests an association between longer QE duration and higher SA.

Table 9. Meta-regression results data of QE duration on SA at different competition levels.

Variable	Estimate	Std. Error	z-Value	p-Value	95% CI
Intercept	−2.0446	0.9083	−2.251	0.0244 *	−3.8249, −0.2644
QE duration on SA	0.0034	0.0014	2.4507	0.0143 *	0.0007, 0.0060

Abbreviations: SA = shooting accuracy, QE duration = quiet eye duration. Note: Intercept represents the baseline SA when QE duration = 0 and the reference group is near-elite athletes, * p < 0.05 indicates a statistically significant correlation.

The meta-regression results demonstrated a significant positive association between QE duration and SA (Table 9). Additionally, to facilitate an integrated understanding of the results, a summary of the key correlations and systematic review findings is provided in Table 10.

Table 10. Summary of gaze behavior correlations at different competition levels.

Gaze Behavior	Elite vs. Near-Elite [95% CI], p-Value	Data Analysis Results	Systematic Review Results
QE Duration	2.55 [0.15, 1.01], p = 0.01	Elite athletes exhibited significantly longer QE durations, strongly linked to higher SA (p = 0.0143).	Younger players (<18 years) had greater variability in QE durations. Adult: The same as the data analysis results.
FFD	1.56 [−0.10, 0.97], p = 0.12	Positive SA correlation, but no significant difference between competition levels.	Younger players (<18 years) showed greater variability in FFD. Adult: The same as the data analysis results.
FD	1.35 [−0.44, 2.41], p = 0.18	No significant relationship with SA or competition-level differences.	More experienced players exhibited longer FD, reflecting more advanced SA.

Abbreviations: SA = shooting accuracy, QE duration = quiet eye duration.

4. Discussion

This systematic review and meta-analysis aimed to explore the relationships between gaze behavior—specifically, quiet eye duration (QE), final fixation duration (FFD), and fixation duration (FD)—and basketball shooting performance (shooting accuracy). Gaze behavior was assessed in elite and near-elite athletes in order to compare the differences between them.

4.1. Gaze Behaviors and Shooting Performance

Our study found a significant positive correlation between QE duration and shooting accuracy, as confirmed through the meta-analysis. These findings align with previous research [5,7,8], which highlights the stabilizing effect of extended QE durations on performance, particularly under high-pressure conditions [6,21,39]. For instance, elite penalty takers in soccer have been shown to maintain prolonged QE on target areas (e.g., goalkeeper movements or specific goal corners), enhancing their success rates under pressure (Wood & Wilson, 2011) [47]. The prolonged QE provides athletes with additional time to process critical visual information and optimize the planning of movements [21,39]. This mechanism allows athletes to refine the execution of movements through enhanced cognitive readiness and suppression of distractions [5,7]. Notably, our findings extend prior work by systematically quantifying the magnitude of this relationship across diverse competitive levels—a dimension under-explored in earlier reviews [4].

Similarly, final fixation duration (FFD) showed a positive, albeit weaker, association with shooting accuracy. This finding mirrors research in soccer, where strikers exhibit longer FFDs on goalkeepers or target zones to optimize their shot precision [48]. During the final fixation phase, athletes integrate spatial and kinematic information (e.g., basket distance, body positioning) to calibrate their motor output. Research has demonstrated that athletes’ eye movements during the final phase of shooting involve the rapid integration of spatial and motion-related information, such as the distance and angle of the basket and their body positioning [5,9,12,49]. As such, this ability is comparable to the trajectory prediction demands observed in precision sports like darts, where athletes rely on visual regulation to anticipate the path of the dart [15,50]. Notably, this adaptability is context-dependent: free throws demand stable visual focus to regulate force consistency [9], whereas jump shots require dynamic adjustments to account for vertical displacement and defensive interference [10].

In contrast, fixation duration (FD) demonstrated no significant correlation with shooting accuracy, suggesting that its role is more supportive in facilitating broader visual scanning rather than directly enhancing precision [3,45]. For example, during free throws, FD might allow players to evaluate potential passing or shooting paths, rather than solely focusing on the basket [40]. This aligns with findings in soccer and tennis, where FD is distributed across multiple cues (e.g., opponent positioning, ball spin) to inform tactical decisions, rather than focusing on a single target [51,52]. Meanwhile, the observed high heterogeneity in FD could stem from methodological differences, such as variations in fixation tracking device calibration or the inclusion/exclusion of saccadic fixations in FD definitions [7]. This variability may further be influenced by factors such as shot type (free throw vs. jump shot) and defensive scenarios, complicating direct comparisons [40,42]. For example, FD during free throws often reflects deliberate, uninterrupted visual focus [3], whereas FD during dynamic jump shots requires rapid adaptation to changing conditions and stimuli [13,42]. Future studies should standardize FD measurement protocols and account for contextual variables in order to clarify its role in shooting performance.

4.2. Differences Between Competition Levels

Our study found that elite athletes demonstrated significantly longer QE durations when compared to near-elite athletes. This finding suggests that elite athletes possess superior concentration abilities during shooting, thereby optimizing their shooting performance [5,7]. This aligns with research in soccer players, where professional players were found to sustain longer target-focused QE during penalties when compared to less-experienced athletes [47]. This result is consistent with the studies by Vickers [5] and Klostermann et al. [7], who emphasized that a longer QE duration helps to improve the decision-making ability and cognitive readiness of players [5,7]. Additionally, research has indicated that the ability of elite athletes to suppress distractions (e.g., defensive pressure, crowd noise) and prioritize task-relevant visual cues likely reflects advanced attentional allocation mechanisms [6,39]. Such cognitive control is critical in high-pressure environments, where lapses in focus can compromise performance [53]. These results underscore QE duration as a key discriminator of competitive level, offering a quantifiable metric for talent identification and training interventions [11,20].

Although FFD has been described as showing a slightly longer duration in elite athletes [5,8], this trend was not significant in the current study. Variations in FFD may depend on the specific shooting scenario [8]. For instance, during free throws in an open-court setting, FFD tends to be longer, providing the athlete with ample time to adjust their movement [9]. In contrast, during jump shots under high defensive pressure, FFD may be shorter as athletes are required to execute their movements more rapidly to avoid interference [10,21]. This contextual variability highlights the need for sport-specific analyses of FFD, as its functional role may differ across tasks and environments [10].

Similarly, FD did not show significant differences between elite and near-elite athletes. Research has suggested that the role of FD lies more in facilitating overall visual search strategies than directly influencing shooting accuracy [9,13]. For example, FD involves a combination of fixations on key targets (e.g., the basket) and secondary information (e.g., the position of defensive players) [42]. Elite athletes in soccer and tennis often prioritize rapid scanning over prolonged FD in order to process dynamic environments [54,55]. In basketball, FD may involve alternating between primary targets (e.g., basket) and secondary cues (e.g., defender positioning), with its duration modulated by game tempo, being shorter during fast breaks for rapid decisions [10] and relatively longer to allow for comprehensive scene analysis during half-court offense [50]. These findings emphasize the need to contextualize gaze behavior metrics within the scope of specific tactical demands.

4.3. Practical Implications

Training Interventions: Drills emphasizing gaze stability (e.g., prolonged focus on the rim during free throws) and adaptive visual strategies (e.g., rapid shifts between targets in jump shots) can enhance shooting consistency.

Youth Development: Younger athletes (<18 years) exhibited greater variability in gaze behaviors [8], suggesting early interventions to stabilize fixation patterns could accelerate skill acquisition.

4.4. Limitations

Most of the included studies were cross-sectional, restricting the ability to infer causal relationships between gaze behavior and shooting performance. Future research should prioritize longitudinal studies or experimental interventions to determine whether improvements in gaze behavior lead to enhanced shooting accuracy over time.

Additionally, significant heterogeneity was observed in fixation duration (FD) across the considered studies, suggesting that shot type, defensive pressure, and individual playing styles may influence gaze behavior. Addressing these factors in future research is crucial for improving the reliability of the obtained findings. Standardizing methodologies would help to reduce variability and allow for more precise comparisons across studies.

5. Conclusions

The results reported in this study highlight the importance of gaze behavior metrics such as QE duration for elite basketball players who possess the ability to sustain attention for longer periods, potentially contributing to more accurate shooting. Although FFD and FD did not show strong associations with shooting accuracy, further research is warranted to explore the nuances of these gaze behaviors, particularly in the context of developing training interventions for athletes at different competition levels.