Thirty professional FPS players reacted to a beep in 248.33 ms on average, only 8.68 ms ahead of 30 sports science students, and that gap was not statistically significant (p = .278; Saygı and Odabaş, Eurasian Research in Sport Science 2023). The same players beat the students by 144.36 ms on an aiming test, which suggests competitive practice shows up in the hand before it shows up in the ear. In a Korean shooter task built around footsteps on leaves, 16 adults needed about 7 seconds to find a sound and fire at it, far longer than any button-press figure (Kim et al., Journal of the Acoustical Society of Korea 2025). We aggregated data from the Journal of the Acoustical Society of America, the Journal of the Acoustical Society of Korea, Scientific Reports, Frontiers in Human Neuroscience, Mensch und Computer, BMJ Public Health, and university repositories. This page covers reaction speed and the use of sound cues during play; for how headphones render direction, see our gaming headset HRTF and virtual surround localization roundup.
A short glossary, because these terms repeat below. Reaction time is the delay between a stimulus and a response. Signal-to-noise ratio (SNR) compares the level of a sound with the noise around it; a negative SNR means the noise is louder. A cue is any sound that gives the player information, such as a footstep that reveals an enemy.
TL;DR
- Auditory reaction time was 248.33 ms in 30 professional FPS players vs 257.01 ms in 30 sports science students, p = .278 (Saygı and Odabaş, Eurasian Research in Sport Science 2023).
- Players above 14 hours a week scored 258.73 ms vs 255.76 ms on auditory reaction time, p = .397, yet visual reaction time split them 227.13 ms vs 278.67 ms (Ersin et al., Kinesiology 2022).
- Elite badminton players responded to sound in 236.6 ms vs 287.7 ms to motion, p < 0.001 (Hülsdünker et al., Frontiers in Human Neuroscience 2021).
- Finding a footstep or noise burst and shooting at it took 6.91 s with a personalized profile vs 7.51 s with a dummy-head profile (Kim et al., Journal of the Acoustical Society of Korea 2025).
- Action video game players had an auditory cuing cost of 46 ms vs 73 ms in non-players, p = .049 (Mancarella et al., Computers in Human Behavior 2022).
- Across 80 adults, action game play gave no auditory benefit in reaction or speech-in-noise tests (Stewart et al., Scientific Reports 2020).
- Spatial 3-D audio cut visual search time by about 25% with moving displays and 22% with static ones (Watamaniuk et al., Human Factors 2010).
- Added audio delay of up to 500 ms had no significant effect on Counter-Strike: Global Offensive score, p = 0.154 (Halbhuber et al., Mensch und Computer 2022).
- Cutting 100 ms of local system latency raised accuracy by 6 percent, against no significant audio delay effect above (Liu et al., ACM EHPHCI 2021).
- A machine model located footsteps behind a corridor corner with 98.83% classification accuracy (Wang et al., Journal of the Acoustical Society of America 2026).
- Four shooter games averaged 85.6 to 91.2 dB, with impulses up to 119 dB (Dillard et al., BMJ Public Health 2024).
1. Simple Auditory Reaction Time: The Baseline
The textbook claim is that the ear beats the eye, and controlled lab work supports it. Nineteen elite badminton players responded to sound in 236.6 ms vs 287.7 ms to a motion stimulus, a gap of 51.1 ms (287.7 - 236.6; p < 0.001) (Hülsdünker et al., Frontiers in Human Neuroscience 2021). The auditory figure includes a deliberate 26 ms silent lead-in that models sound traveling about 9 m, so the ear’s raw advantage in that test was larger.
For shooter players, the useful finding is the third row: sound and sight together were faster than either alone. A footstep plus a glimpse is a stronger trigger than a footstep alone. Yet the Ohio Journal of Science comparison found the opposite order in its sound cue test, a warning that the method can change the answer.
| Metric | Value | Source |
|---|---|---|
| Mean simple reaction time to a beep vs to a visual box (14 subjects, space-bar press on a laptop) | About 284 ms vs about 331 ms | Shelton and Kumar, Neuroscience and Medicine 2010 |
| Mean reaction time to sound vs to radial motion (19 elite badminton players, button press, earphones) | 236.6 ms vs 287.7 ms (p < 0.001) | Hülsdünker et al., Frontiers in Human Neuroscience 2021 |
| Mean reaction time to sound plus motion together vs sound alone (same 19 players) | 224.4 ms vs 236.6 ms (p = 0.002) | Hülsdünker et al., Frontiers in Human Neuroscience 2021 |
| Mean reaction to the sound of a smash vs the sound of a drop shot (same 19 players) | 223 ms vs 250 ms | Hülsdünker et al., Frontiers in Human Neuroscience 2021 |
| Auditory responses vs visual or tactile responses (36 college football athletes, esports competitors and controls) | Auditory slower, F(140, 3) = 286.5, p = 0.0000 | Luu et al., Ohio Journal of Science 2021 |
| Accepted average simple reaction time for college-age adults: light vs sound | About 190 ms vs about 160 ms | Jain et al. 2015, as cited in Saygı and Odabaş 2023 |
Context note: most recent available data for the plain beep-vs-box design: Shelton and Kumar, 2010. That page reports no standard deviations or p value, so treat its 47 ms gap (331 - 284) as indicative only.
2. Esports and FPS Players: Auditory Speed vs Everyone Else
The headline here is an absence. Hours of play moved visual reaction time by 51.54 ms and aim time by 138.22 ms, but auditory reaction time differed by only 2.97 ms (258.73 - 255.76), in the wrong direction (Ersin et al., Kinesiology 2022). All 53 men in that sample played FPS games. The authors suggest two causes: games expose players to far more visual than auditory feedback, and some players switch sound off.
The professional FPS sample tells the same story. Those players were tied with students on visual reaction time, slightly ahead on auditory, and far ahead on aiming. They also reacted to the beep 18.81 ms slower than to the color change (248.33 - 229.52), the reverse of the lab order in section 1. Browser tools add their own audio delay, which likely explains part of that reversal.
| Metric | Value | Source |
|---|---|---|
| Auditory reaction time, 30 professional FPS players vs 30 sports science students (browser beep test, five runs, wired earbuds) | 248.33 ms vs 257.01 ms (p = .278) | Saygı and Odabaş, Eurasian Research in Sport Science 2023 |
| Visual reaction time, same two groups (browser color-change test) | 229.52 ms vs 229.51 ms (p = .999) | Saygı and Odabaş, Eurasian Research in Sport Science 2023 |
| Mean time per target on a 30-target aim test, same two groups | 414.70 ms vs 559.06 ms (p = .000) | Saygı and Odabaş, Eurasian Research in Sport Science 2023 |
| Auditory reaction time, 28 players above 14 hours a week vs 25 below (53 men, online test) | 258.73 ms vs 255.76 ms (p = .397) | Ersin et al., Kinesiology 2022 |
| Visual reaction time, same two groups | 227.13 ms vs 278.67 ms (p < .001) | Ersin et al., Kinesiology 2022 |
| Aim reaction time, same two groups | 440.61 ms vs 578.83 ms (p < .001) | Ersin et al., Kinesiology 2022 |
| Composite reaction time, 12 esports competitors vs 12 noncompetitive controls | Esports group faster, p = 0.0042 | Luu et al., Ohio Journal of Science 2021 |
| Slowing from auditory distraction across action players, in-between players, other players and non-players (80 adults, attend-location task) | No group difference, p = 0.93, BF10 = 0.087 | Stewart et al., Scientific Reports 2020 |
Context note: a BF10 (Bayes factor) below 1 favors no difference over a difference. Most recent available data for lab-grade auditory testing in action players: Stewart et al., Scientific Reports 2020. We found no published study that measured auditory reaction time in professional CS2 or Valorant rosters with calibrated lab equipment.
3. Finding a Sound Cue in a Shooter Task
Lab buttons measure detection. Shooter play asks for detection, direction and aim, and the time cost changes by an order of magnitude. In a Unity shooter task using footsteps on leaves and noise bursts, 16 adults took 6.91 s to 7.51 s per trial (Kim et al., Journal of the Acoustical Society of Korea 2025). Those seconds include searching and shooting, so they cannot be compared with the milliseconds in sections 1 and 2.
The personalized profile cut 0.60 s against the dummy head (7.51 - 6.91), or 8.0% by our math (0.60 / 7.51), without improving where shots landed. Players found the direction faster, not more exactly. The authors’ model suggests beginners gain the most, perhaps because they lean harder on sound to infer enemy positions, a hypothesis they say needs further testing.
Attention research adds a second lens. Its cuing cost is the extra time needed when a warning sound comes from the opposite ear to the target. A smaller cost means a listener recovers faster from a misleading sound. Still, a 2026 study of 53 undergraduates found no link between action play and auditory spatial memory.
| Metric | Value | Source |
|---|---|---|
| Mean time to find a sound and shoot toward it (16 adults, 336 trials each): personalized time-difference profile vs Steam Audio default vs MIT KEMAR dummy head | 6.91 s vs 7.37 s vs 7.51 s | Kim et al., Journal of the Acoustical Society of Korea 2025 |
| Aiming error across those three audio profiles | No significant difference, F(2, 5264) = 0.9947, p = 0.3699 | Kim et al., Journal of the Acoustical Society of Korea 2025 |
| Modeled time saved with the personalized profile vs KEMAR at a gaming skill score of zero, and loss of that saving per skill point | 2.694 s saved; 0.082 s less per point (p < 0.001) | Kim et al., Journal of the Acoustical Society of Korea 2025 |
| Auditory cuing cost, 25 action video game players vs 23 non-players (240 trials, left or right ear, headphones) | 46 ms vs 73 ms (p = .049, d = -.593) | Mancarella et al., Computers in Human Behavior 2022 |
| Accuracy at naming the target ear, action players vs non-players (same task) | 94.27% vs 92.54% (p = .321) | Mancarella et al., Computers in Human Behavior 2022 |
| Link between action game play and auditory spatial span (53 undergraduates, Audio-Corsi task, forward and backward) | Not significant, all p > 0.1 | Pasescu et al., Frontiers in Psychology 2026 |
| Visual search time with 3-D audio cues vs without (8 participants, one target among 15 distractors), moving vs static display | About 25% lower vs about 22% lower | Watamaniuk et al., Human Factors 2010 |
Context note: the Pasescu study could only detect correlations of about r = 0.37 or larger, so a small true effect remains possible. Most recent available data for 3-D audio search: Watamaniuk et al., 2010, which found no further gain from practice after brief first exposure.
4. What a Footstep Carries
Footsteps are richer than a position marker. Listeners and machines can pull identity and even location around corners from them. Trained on walking sounds, a simple classifier told three people apart with 73.53% accuracy on unseen data (Diapoulis et al., Euronoise 2018). With three walkers, chance is one in three, so the signal is real even in a small test.
Game design has already put that richness to work. An inspection of the early chapters of Half-Life 2 found audio doing six separate jobs, from locating sources to signaling threat, and footstep surface changes hinted at nearby enemy types (Coutinho et al., SBGames 2011).
| Metric | Value | Source |
|---|---|---|
| Footsteps in each trial of the classic walker-gender listening study | 4 footsteps on one surface; listeners could identify gender (rate not in the public abstract) | Farrington and Pastore, ASA meeting abstract 1996, on Li et al., JASA 1991 |
| Classifier accuracy naming 3 walkers on unseen data, from 1,357 sound events: 3-feature vs 6-feature model | 73.53% vs 53.43% | Diapoulis et al., Euronoise 2018 |
| Upper frequency where walking activity was visible in the spectrogram | Up to 8000 Hz | Diapoulis et al., Euronoise 2018 |
| Classification accuracy of a neural network locating real footsteps behind a corridor corner | 98.83% | Wang et al., Journal of the Acoustical Society of America 2026 |
| Reduction in average position error, same network vs a reflection-aware method | 2.56 m | Wang et al., Journal of the Acoustical Society of America 2026 |
| Feeling lighter when own footsteps had 1 to 4 kHz boosted 12 dB vs low band boosted (25 women vs 26 men) | Women Z = -3.43, p < 0.001; men Z = -1.31, p = 0.19 | Clausen et al., Frontiers in Human Neuroscience 2021 |
| Separate jobs audio performs in Half-Life 2 (first 3 of 14 chapters inspected) | 6 communication strategies | Coutinho et al., SBGames 2011 |
| Retailer’s stated footstep bands for equalizer tuning (vendor’s own claim, no source given) | Thud 125 to 250 Hz; scuff 2 to 4 kHz | ATTACKSHARK blog 2026 |
Context note: the vendor row is the only equalizer guidance we found, and it cites no measurement. The Wang et al. abstract describes a machine, not a player, and the Clausen study tested how walkers felt, not whether they could detect anything. Most recent available data for human walker-gender listening: Li et al., JASA 1991, summarized in an ASA meeting abstract.
5. Hearing Cues Through Noise
Gunfire, abilities and teammates all compete with a footstep, and noise does not damage every judgment equally. Noise hit front-back judgments hardest and left-right judgments least, across 239 source positions (Good and Gilkey, JASA 1996). For a shooter, that suggests a side may stay readable in a firefight after front and back have blurred, though no game test has confirmed it.
The hopeful finding is that movement helps. In a 30-person study, listeners who turned their heads more had lower errors in every noise condition. Head rotation does not transfer to a flat-screen shooter directly, though turning the in-game camera may serve a similar purpose, an idea worth testing. Noise in voice chat is a related problem, covered in our neural noise suppression latency statistics.
| Metric | Value | Source |
|---|---|---|
| Which judgment noise damaged most (3 listeners, 239 source positions, 9 SNR levels across 23 dB) | Left-right least, up-down more, front-back most | Good and Gilkey, JASA 1996 |
| Smallest separation between two sources that normal-hearing listeners could resolve in quiet (6 listeners, 11 loudspeakers, 69.1% criterion) | 1.4 to 5.1 degrees | Kerber and Seeber, Ear and Hearing 2012 |
| Resolution of normal-hearing listeners with noise 5 dB louder than the signal | Roughly equal to bilateral cochlear implant users in quiet, 8.3 to 43.6 degrees | Kerber and Seeber, Ear and Hearing 2012 |
| Effect of head movement on localization error (30 normal-hearing listeners, nature, traffic and cocktail party scenes in virtual reality) | More and wider head rotation, lower error in all three scenes | Valzolgher et al., Cognitive Research: Principles and Implications 2024 |
| Benefit from spatially separated talkers across action players and three comparison groups (80 adults, LiSN-S test) | No group difference, p = 0.48, BF10 = 0.17 | Stewart et al., Scientific Reports 2020 |
| Speech-in-noise threshold across the same groups (BKB-SiN test) | No group difference, p = 0.80, BF10 = 0.10 | Stewart et al., Scientific Reports 2020 |
Context note: none of these tests used game audio, and the Kerber listeners judged only the frontal horizontal plane. Most recent available data for the noise-by-judgment ordering: Good and Gilkey, 1996, on just three listeners. Kerber and Seeber tested SNRs from +10 to -7 dB.
6. Audio Delay in Counter-Strike
Sound reaches the player late on many setups, and the surprise is how little that seemed to matter for score. Across 0, 40, 270 and 500 ms of added audio delay, Counter-Strike: Global Offensive score did not change significantly (p = 0.154), though expert tension nearly tripled at 500 ms (Halbhuber et al., Mensch und Computer 2022). Experts did trend downward, losing 30.9 points from 0 to 270 ms (348.7 - 317.8), or 8.9% by our math (30.9 / 348.7). With 12 players per group, a real effect of that size could go undetected.
The contrast with visual and input delay is sharp. Removing 100 ms of local latency raised accuracy by about 6 percent in 43 skilled players. For the wider picture of system delay, see our game latency statistics. The study cannot rule out harm from delay in live team play, because participants fought medium-difficulty bots in Deathmatch rather than listening for human opponents.
| Metric | Value | Source |
|---|---|---|
| Mean score at 0 ms vs 270 ms of added audio delay (24 players, 12 experts above 100 hours, 12 novices, Deathmatch vs bots) | Experts 348.7 vs 317.8; novices 151.1 vs 148.7; delay effect p = 0.154 | Halbhuber et al., Mensch und Computer 2022 |
| Tension rating at 0 ms vs 500 ms of added delay (Game Experience Questionnaire) | Experts 0.56 vs 1.65 (p = 0.041); novices 0.72 vs 1.28 (no significant pairwise difference) | Halbhuber et al., Mensch und Computer 2022 |
| Players who could tell that audio delay was being changed | All 12 experts vs 1 of 12 novices | Halbhuber et al., Mensch und Computer 2022 |
| Shortest delay some users detected after a button press: sound vs visual feedback | 19 ms vs 32 ms | Kaaresoja et al. 2014, as cited in Halbhuber et al. 2022 |
| Gain from removing 100 ms of local system latency (43 skilled CS:GO players, 125 ms baseline) | 6 percent accuracy, 3 points per minute, 1.6 experience points on a 5-point scale | Liu et al., ACM EHPHCI 2021 |
| Gain from removing 100 ms of network latency (25 skilled CS:GO players, same baseline) | 2 percent accuracy, 2 points per minute, 0.7 experience points on a 5-point scale | Liu et al., ACM EHPHCI 2021 |
Context note: most recent available data for added audio delay in a commercial shooter: Halbhuber et al., 2022. Liu et al. studied local and network delay only, with Intel co-authors on the paper.
7. The Cost of Playing by Ear: Loudness and Access
Turning the game up to catch quiet steps also raises every gunshot. Four shooter games averaged 85.6 to 91.2 dB, and impulse sounds reached up to 119 dB (Dillard et al., BMJ Public Health 2024). The margin is thin: that review’s adult example allows 2.5 hours a week at 92 dB, close to the loudest shooter figure, and the average exposure it found was 3 hours a week.
Sound as a competitive tool also shuts some players out. Among deaf and hard of hearing respondents in Brazil who had played games, FPS titles were disliked far more often than liked. Respondents suggested vibration in controllers among their fixes, and our haptic audio and tactile feedback statistics track that hardware.
| Metric | Value | Source |
|---|---|---|
| Evidence base of the gaming and hearing review (peer-reviewed studies, people, countries) | 14 studies, 53,833 people, 9 countries | Dillard et al., BMJ Public Health 2024 |
| Average level of four shooter games | 88.5, 87.6, 85.6 and 91.2 dB | Dillard et al., BMJ Public Health 2024 |
| Peak impulse sounds during gameplay vs permissible limits for children and adults | Up to 119 dB vs about 100 dB (children) and 130 to 140 dB (adults) | Dillard et al., BMJ Public Health 2024 |
| Weekly exposure allowance at 92 dB: adults vs children | 2.5 hours vs 45 minutes | Dillard et al., BMJ Public Health 2024 |
| Studies linking gaming to hearing loss or tinnitus, of those that tested it | 4 of 5 found significant links | Dillard et al., BMJ Public Health 2024 |
| FPS games cited as disliked vs liked (68 deaf and hard of hearing players, Brazil) | 44% vs 11% | Coutinho et al., SBGames 2011 |
| Deaf and hard of hearing players reporting difficulty due to hearing loss, vs no difficulty (same 68 players) | 28% vs 57% | Coutinho et al., SBGames 2011 |
Context note: the review flags esports as under-studied, and only 2 of its sources from the past 10 years measured sound levels objectively. These figures describe exposure and association, not diagnosis, and this roundup gives no medical advice. Most recent available data for FPS attitudes among deaf players: Coutinho et al., 2011.
Summary: Competitive FPS Footstep Audio and Sound Cue Reaction Time by the Numbers
| Metric | Value | Source |
|---|---|---|
| Auditory reaction time, professional FPS players vs sports science students | 248.33 ms vs 257.01 ms (p = .278) | Saygı and Odabaş 2023 |
| Time per target on a 30-target aim test, same groups | 414.70 ms vs 559.06 ms (p = .000) | Saygı and Odabaş 2023 |
| Auditory reaction time, above vs below 14 hours of play a week | 258.73 ms vs 255.76 ms (p = .397) | Ersin et al. 2022 |
| Reaction to sound vs motion, elite badminton players | 236.6 ms vs 287.7 ms (p < 0.001) | Hülsdünker et al. 2021 |
| Reaction to sound plus motion together | 224.4 ms | Hülsdünker et al. 2021 |
| Simple reaction to a beep vs a visual box | About 284 ms vs about 331 ms | Shelton and Kumar 2010 |
| Time to find a sound and shoot toward it, personalized vs dummy-head profile, and aiming error between them | 6.91 s vs 7.51 s; error p = 0.3699, not significant | Kim et al. 2025 |
| Auditory cuing cost, action players vs non-players | 46 ms vs 73 ms (p = .049) | Mancarella et al. 2022 |
| Slowing from auditory distraction across action and non-action groups | p = 0.93, BF10 = 0.087 | Stewart et al. 2020 |
| Visual search time saved by 3-D audio, moving vs static display | About 25% vs about 22% | Watamaniuk et al. 2010 |
| Classifier accuracy naming 3 walkers from their walking sound | 73.53% | Diapoulis et al. 2018 |
| Neural network accuracy locating footsteps behind a corner | 98.83% | Wang et al. 2026 |
| Listener resolution in quiet, normal hearing | 1.4 to 5.1 degrees | Kerber and Seeber 2012 |
| Effect of up to 500 ms audio delay on shooter score | p = 0.154, not significant | Halbhuber et al. 2022 |
| Accuracy gained by removing 100 ms of local latency | 6 percent | Liu et al. 2021 |
| Peak impulse sounds in game audio | Up to 119 dB | Dillard et al. 2024 |
| FPS games cited as disliked vs liked by deaf and hard of hearing players | 44% vs 11% | Coutinho et al. 2011 |
Methodology and Sources
We prioritized peer-reviewed measurements with a named sample and task. Where we derived a figure, the subtraction or division is shown beside it. Rows marked ‘as cited’ come from the citing paper, because the original was not read directly.
- Saygı and Odabaş, Comparing the Accuracy and Reaction Times of Esports Players and Sport Sciences Students, Eurasian Research in Sport Science, December 2023.
- Ersin et al., The Relationship Between Reaction Time and Gaming Time in E-Sports Players, Kinesiology 2022.
- Hülsdünker et al., Auditory Information Accelerates the Visuomotor Reaction Speed of Elite Badminton Players in Multisensory Environments, Frontiers in Human Neuroscience 2021.
- Shelton and Kumar, Comparison between Auditory and Visual Simple Reaction Times, Neuroscience and Medicine 2010.
- Luu et al., Reaction Times for Esport Competitors and Traditional Physical Athletes are Faster than Noncompetitive Peers, Ohio Journal of Science 2021.
- Stewart et al., Auditory Cognition and Perception of Action Video Game Players, Scientific Reports 2020.
- Mancarella et al., Enhanced Disengagement of Auditory Attention and Phonological Skills in Action Video Gamers, Computers in Human Behavior 2022 (accepted manuscript).
- Pasescu et al., Game Over? Examining Associations between Video Game Play and Visual and Auditory Spatial Ability, Frontiers in Psychology 2026.
- Kim et al., Interplay of Auditory Cue Optimization and Gaming Proficiency to Reduce Reaction Times in a First-Person Shooter Game, Journal of the Acoustical Society of Korea 2025.
- Watamaniuk et al., Visual Search Performance With 3-D Auditory Cues, Human Factors 2010.
- Diapoulis et al., Person Identification from Walking Sound on Wooden Floor, Euronoise 2018.
- Wang et al., A Dual-Branch Fusion Network for Footstep Sound Source Localization in Non-Line-of-Sight Corridors, JASA 2026.
- Li et al., Perception of Acoustic Source Characteristics: Walking Sounds, JASA 1991.
- Farrington and Pastore, Perceiving Source Characteristics from Complex Sounds, ASA meeting abstract 1996.
- Clausen et al., Action Sounds Informing Own Body Perception Influence Gender Identity and Social Cognition, Frontiers in Human Neuroscience 2021.
- Good and Gilkey, Sound Localization in Noise: The Effect of Signal-to-Noise Ratio, JASA 1996.
- Kerber and Seeber, Sound Localization in Noise by Normal-Hearing Listeners and Cochlear Implant Users, Ear and Hearing 2012.
- Valzolgher et al., Sound Localization in Noisy Contexts, Cognitive Research: Principles and Implications 2024.
- Halbhuber et al., The Effects of Auditory Latency on Experienced First-Person Shooter Players, Mensch und Computer 2022.
- Liu et al., Comparing the Effects of Network Latency versus Local Latency on Competitive First Person Shooter Game Players, ACM EHPHCI 2021.
- Dillard et al., Risk of Sound-Induced Hearing Loss from Exposure to Video Gaming or E-Sports, BMJ Public Health 2024, BMJ Group press summary.
- Coutinho et al., An Analysis of Information Conveyed through Audio in an FPS Game and Its Impact on Deaf Players Experience, SBGames 2011.
- ATTACKSHARK, Audio Tuning for Esports, February 3, 2026 (vendor blog, unsourced claim).
- Data watch: published data on this topic is thin, and we found no study that timed footstep reactions in professional players during live CS2 or Valorant matches. Browser reaction tests include device and audio delay, so their absolute values run high and should only be compared inside one study. Ersin et al. print the same SD (33.15) for visual and auditory scores in one group, show the auditory p value as 397 in the table but .397 in the text, and let players test on their own varied devices. The Counter-Strike audio delay paper mentions 400 ms once where every other passage says 500 ms, and its conclusion places expert effects at 270 ms although expert post-hoc tests were significant only against 500 ms. Kim et al. differ between figure and text on positions per layer (12 vs 7) and publish no accuracy percentages. Hülsdünker reaction times include a 26 ms silent lead-in. Luu et al. report p = 0.0000 and the page we read gives no millisecond means. Mancarella et al. and Stewart et al. conflict on auditory benefit; both are cross-sectional, so cause cannot be assigned. Coutinho et al. print 61% and 68% for the share of respondents who had played (68 of 111 is 61%). A 2026 Frontiers mini review on game audio in esports reported no numeric footstep or localization findings. We could not read Johanson and Mandryk’s CHI 2016 paper on audio cues in shooters, so no figure from it appears here.
Last updated: October 31, 2026. We update this roundup quarterly, and the next refresh is expected when the next ACM CHI PLAY proceedings are published.