Visual Fixations Duration as an Indicator of Skill Level in eSports

Boris B. Velichkovsky, Nikita Khromov, Alexander Korotin, Evgeny Burnaev, Andrey Somov

Abstract

Using highly interactive systems like computer games requires a lot of visual activity and eye movements. Eye movements are best characterized by visual fixation - periods of time when the eyes stay relatively still over an object. We analyzed the distributions of fixation duration of professional athletes, amateur and newbie players. We show that the analysis of fixation durations can be used to deduce the skill level in computer game players. Highly skilled gaming performance is characterized by more variability in fixation durations and by bimodal fixation duration distributions suggesting the presence of two fixation types in high skill gamers. These fixation types were identified as ambient (automatic spatial processing) and focal (conscious visual processing). The analysis of computer gamers' skill level via the analysis of fixation durations may be used in developing adaptive interfaces and in interface design.

Introduction

In the high-stakes world of eSports, success depends on a complex mix of cognitive, motor, and visual skills. While professional performance is clearly superior, quantifying the underlying mechanisms from a Human-Computer Interaction (HCI) perspective remains challenging. This research investigates whether eye-tracking data—specifically the duration of fixations—can serve as an objective metric for expertise. Understanding these patterns is critical for scouting talent and designing adaptive interfaces that respond to a user’s skill level or cognitive load in real-time.

Methodology

The study utilized an in-house IoT platform to synchronize data from multiple sources, including a Tobii EyeX eye-tracker (30 Hz) and in-game loggers for Counter-Strike: Global Offensive (CS:GO). Fixations were extracted using the Mould algorithm, which adapts detection thresholds to individual physiological differences.

The analysis involved three layers:

  1. Descriptive Statistics: Comparing means, medians, and standard deviations across skill groups.
  2. Vincentile Analysis: Partitioning duration distributions into five bins to observe tail behavior.
  3. Kernel Density Estimation (KDE): Visualizing the probability density function to identify unimodal versus bimodal distributions.

Participants

The dataset included 21 participants categorized by CS:GO experience:

  • Low Skill (N=10): < 700 hours of playtime.
  • High Skill (N=7): > 700 hours.
  • Pro (N=4): Professional athletes (Team "Monolith") with > 10,000 hours of experience.

Results and Discussion

The Bimodal Shift in Visual Processing

A primary finding is the fundamental shift in the distribution of fixation durations as skill increases. While novices exhibit a uniform, unimodal distribution, professional players develop a "bimodal" pattern. This suggests the emergence of two distinct visual processing systems:

  • Ambient Fixations (~100ms): Rapid, automatic spatial processing (the "Where" system) used for situational awareness.
  • Focal Fixations (>300-500ms): Conscious, deliberate identification of objects (the "What" system).
Kernel Density Estimation of fixation durations across skill levels, showing the emergence of bimodality in professional players.

Statistical Indicators of Expertise

Statistical analysis confirmed that higher skill levels correlate with increased variability in visual strategy. As shown in the table below, the standard deviation (SD) of fixation durations increased significantly with skill ($p < 0.001$). Professionals also demonstrated a wider range of fixation durations, with significantly lower minimums and higher maximums than novices.

Descriptive statistics of fixation durations (ms) by skill level.

Novices appear "stuck in the middle," lacking both the ultra-fast ambient scans and the deep focal fixations characteristic of experts. This lack of specialization results in a more uniform, less adaptive visual strategy.

Conclusion and Future Work

This study demonstrates that fixation duration distributions are a reliable indicator of gaming expertise. The transition from unimodal to bimodal distributions reflects the development of specialized ambient and focal processing.

Future work should address the small sample size of the professional group and explore generalizability across other game genres. These insights pave the way for adaptive user interfaces that can detect a player's cognitive state and adjust visual saliency to optimize performance or training.

Summary updated

Abstract

Using highly interactive systems like computer games requires a lot of visual activity and eye movements. Eye movements are best characterized by visual fixation - periods of time when the eyes stay relatively still over an object. We analyzed the distributions of fixation duration of professional athletes, amateur and newbie players. We show that the analysis of fixation durations can be used to deduce the skill level in computer game players. Highly skilled gaming performance is characterized by more variability in fixation durations and by bimodal fixation duration distributions suggesting the presence of two fixation types in high skill gamers. These fixation types were identified as ambient (automatic spatial processing) and focal (conscious visual processing). The analysis of computer gamers' skill level via the analysis of fixation durations may be used in developing adaptive interfaces and in interface design.
BibTeX
@inproceedings{velichkovsky2019visual,
  author = {Boris B. Velichkovsky and Nikita Khromov and Alexander Korotin and Evgeny Burnaev and Andrey Somov},
  title = {{Visual Fixations Duration as an Indicator of Skill Level in eSports}},
  booktitle = {INTERACT},
  year = {2019},
  url = {https://openreview.net/forum?id=nrDIpGQpnD2},
}
Figure