Research & Safer Listening · Evidence Synthesis

What research shows: brain responses, attention, sleep and anxiety

Peer-reviewed research on auditory stimulation and background noise spans laboratory neuroscience, cognitive psychology, sleep medicine, and perioperative care. Separating physical acoustics and short-term brain responses from clinical outcomes clarifies what the evidence supports.

How the sound works How it is perceived What research shows Page updated: 12 November 2026 · Literature search cutoff: 10 November 2026

Key points

  • Sound can evoke measurable auditory brain responses in laboratory recordings, but those short-term signals do not automatically establish improvements in attention, sleep, or mood.
  • Evidence for broader binaural-beat brainwave entrainment remains inconsistent across peer-reviewed EEG studies using different stimulus parameters and analysis methods (Ingendoh et al., 2023).
  • Attention studies report favorable, non-significant, and sometimes unfavorable task results depending on participant groups and listening conditions (for example, Nigg et al., 2024). Some procedural-anxiety studies report benefits, but methods and results vary considerably.
  • Sleep studies report different results depending on the sound, acoustic environment, comparison group, and measured outcome; no single audio method has demonstrated reliable benefits across all of these settings.

1. Physical effects, perception and measured brain responses

When evaluating claims about rhythmic sound, noise colors, or ambient soundscapes, the most important step is separating three levels of evidence that are often blurred together:

Level 1 · Acoustics

Physical signal behavior

Two close frequencies summed in the same channel interfere at |f2 - f1|; raised-cosine pulse ramps reduce high-frequency sidebands; for ideal pink noise, power spectral density decreases by about 3 dB per octave, corresponding to equal integrated power in each octave band. These properties follow directly from physics and signal mathematics (Moore, 2012; Smith, 2010).

Level 2 · Laboratory Response

Auditory perception & EEG/MEG tracking

Under controlled headphone or speaker conditions, listeners perceive beat fluctuations and masking, and EEG/MEG recordings can detect frequency-following responses (FFR) at carrier frequencies or auditory steady-state responses (ASSR) at modulation rates, though systematic reviews of ongoing EEG band-power entrainment remain inconclusive (Galambos et al., 1981; Schwarz & Taylor, 2005; Coffey et al., 2016; Ingendoh et al., 2023).

Level 3 · Behavioral & Clinical Outcomes

Attention, sleep, anxiety & pain

Whether listening changes task accuracy, overnight sleep architecture, or anxiety depends on the population, comparator, environment, and outcome measure. Across systematic reviews and randomized trials, behavioral findings are mixed and context-dependent (Garcia-Argibay et al., 2019; Nigg et al., 2024; Riedy et al., 2021; Vazzaz et al., 2026).

Critical Distinction

Sensory tracking is not a state switch

Detecting a 40 Hz ASSR peak in primary auditory cortex confirms that auditory neurons track a 40 Hz stimulus; it does not mean the rest of the brain has been switched into a high-focus or altered mood state (Orozco Perez et al., 2020).

2. What studies measure: physiological responses, task performance, and clinical outcomes

Understanding why two papers reach seemingly different conclusions requires looking closely at what each study actually measured (Schünemann et al., 2024, Cochrane Handbook Ch. 15):

  • Electrophysiological and autonomic markers (FFR, ASSR, EEG band power, HRV, skin conductance, pupillometry): Measure short-term neural or autonomic responses during stimulation. For example, Coffey et al. (2016) used MEG to localize cortical and subcortical contributions to a 98 Hz speech fundamental frequency-following response; Pastor et al. (2002) measured scalp EEG across 12 click rates (12 to 60 Hz, n = 28) and H215O PET regional cerebral blood flow across 4 rates (12, 32, 40, and 47 Hz, n = 9); Dos Anjos et al. (2024) recorded scalp EEG in n = 28 healthy adults across 4-minute epochs of an isochronic-inspired stimulus (ITd), gamma binaural beats, and white noise; Alvarsson et al. (2010) measured skin conductance recovery and HF-HRV (N = 40); and Erfanian et al. (2025) measured pupil diameter (31 analyzed) across white, pink, and brown noise.
  • Objective cognitive tasks (ANT, Continuous Performance Tests, Raven's Progressive Matrices, MaRs-IB): Measure reaction time, vigilance decrement, or fluid reasoning accuracy under controlled or home conditions (Klichowski et al., 2023; Sudre et al., 2024; De Francesco et al., 2025; Melnichuk et al., 2025).
  • Polysomnography (PSG), EEG/EOG/EMG sleep staging, behavioral sleep-onset tasks, and subjective questionnaires: The sleep studies discussed here used different measurement methods: full laboratory polysomnography (recording N1, N2, N3 slow-wave sleep, REM sleep, and cortical arousals; Basner et al., 2026; Fan et al., 2024), AASM-scored laboratory sleep staging using EEG, EOG, and EMG rather than a full polysomnography protocol (Jirakittayakorn & Wongsawat, 2018), smartphone behavioral response-omission tasks (Shumov et al., 2023), and self-report questionnaires such as PROMIS Sleep Disturbance or the Pittsburgh Sleep Quality Index (Vazzaz et al., 2026; Ding et al., 2025). These outcomes are not interchangeable.
  • Situational anxiety and acute pain scales (MDAS, DFS, STAI, VAS): Measure short-term distress before or during a specific medical or dental procedure (Xiong et al., 2025; Fatima et al., 2026; O. Akkurt & Ekici, 2026; Kolhe et al., 2024), which differs from chronic generalized anxiety disorder.

3. Attention and task performance

  • Broadband noise in ADHD vs. non-ADHD populations (Nigg et al., 2024; Chee et al., 2024; Basu & Banerjee, 2023): In a meta-analysis of 13 studies (N = 335) in children and young adults with ADHD or elevated attentional symptoms, white and pink noise yielded a small, statistically significant improvement in task performance (g = 0.249, 95% CI 0.135 to 0.363, p < 0.0001, GRADE moderate certainty). Conversely, across 11 studies (N = 335) in non-ADHD participants, white/pink noise slightly impaired performance (abstract: g = −0.212, 95% CI −0.355 to −0.069, p = 0.0036; main text results: g = −0.206, 95% CI −0.330 to −0.081, p = 0.001; no separate GRADE rating stated for the non-ADHD contrast). No qualifying experimental trials of brown noise existed within that review's search window. Chee et al. (2024) systematically reviewed 31 studies across music, ambient/white noise, and binaural beats on autonomic arousal and cognitive performance, and Basu & Banerjee (2023) meta-analyzed 15 binaural-beat cognition studies (31 effect sizes, overall g = 0.40), both noting heterogeneous, task-dependent results.
  • Brown noise and music tempo in a stationary car cabin (De Francesco et al., 2025): In N = 26 healthy adults completing the Attention Network Test behaviorally (n = 21 for heart rate) across slow-beat music (≈ 50 BPM), fast-beat music (≈ 155 BPM), brown noise, and silence, slow-beat music significantly improved attentional orienting and executive control, lowered heart rate, and increased subjective pleasantness, relaxation, and perceived concentration relative to silence, whereas brown noise did not outperform silence or music on objective attentional network measures.
  • Home cognitive testing and expectancy framing (Klichowski et al., 2023): Across 1,000 tested participants (920 analyzed: 369 in Study 1 and 551 in Study 2, where each participant completed a single two-part session on Raven's Progressive Matrices or MaRs-IB), within-condition test scores declined significantly from part 1 to part 2 during binaural-beat listening regardless of whether the stimulus was framed as enhancing brain activity, neutral, or undefined (and across both 5 Hz and 15 Hz binaural beats), whereas the Study 2 control arms (silence, classical music, and a 240 Hz pure tone) showed no statistically significant pre-to-post score change.
  • Parametric EEG and sustained attention (Melnichuk et al., 2025) & bilateral comparisons (Sudre et al., 2024): Across 64 analyzed adults in a 16-cell factorial design (16 Hz beta vs. 40 Hz gamma × 340 vs. 400 Hz carrier × pre-task vs. task-start onset × white noise present vs. absent, with 2 crossover visits per participant against a pure-tone control on the IP-CPT), Melnichuk et al. (2025) reported higher average target detection in separate subgroup contrasts against the matched control for 40 Hz gamma with the 340 Hz carrier (p = 0.002) and when background white noise was present (p = 0.033; with stimulation starting at task onset, p = 0.013), whereas overall vigilance decrement across task blocks was not attenuated (p = 0.751). In the EEG recordings, background white noise attenuated the measured beat-frequency EEG response, particularly for gamma beats (four-way noise-moderation interaction p = 0.028; reported interactions and subgroup comparisons in References). In 38 analyzed adults (G6 n = 19; G40 n = 19), Sudre et al. (2024) found no significant improvement on the objective Not-X-CPT sustained-attention task for 6 Hz or 40 Hz binaural beats (C1), smooth panning (C2), or alternating beeps (C3) compared with the monaural-beat control (C0).

4. Sleep and overnight sleep architecture

  • Hospital and ICU masking vs. home bedrooms (Riedy et al., 2021; Ding et al., 2025; Khandelwal et al., 2025): Riedy et al. (2021) rated overall continuous-noise sleep evidence as very low under GRADE across 38 studies. In a meta-analysis of 12 white-noise RCTs (N = 1,301), Ding et al. (2025) found mixed infant/toddler endpoints (longer 24-hour sleep and higher 12-hour sleep efficiency, but null 12-hour nocturnal total sleep time p = 0.16, 24-hour sleep efficiency p = 0.46, and 24-hour WASO p = 0.46) and lower PSQI scores in adults, older adults, ICU, and non-ICU subgroups with substantial heterogeneity in adult PSQI. In a narrative systematic review of 7 RCTs in hospitalized adults (N = 496; meta-analysis not feasible due to heterogeneity), Khandelwal et al. (2025) noted improvements in sleep efficiency and study-level subjective sleep questionnaires (such as PSQI, RCSQ, and VSH) alongside 8 participants who discontinued because of discomfort or found the intervention less acceptable.
  • Continuous pink noise at 50 dBA vs. quiet control in laboratory PSG (Basner et al., 2026): In a 7-night crossover polysomnography trial (1 noise-free adaptation night + 6 experimental condition nights in n = 25 healthy adults using bedroom loudspeakers, where body-text p-values are adjusted for multiple testing via Benjamini–Hochberg FDR unless otherwise specified), continuous pink noise at 50 dBA (PN50) throughout a quiet night significantly reduced REM sleep by 18.6 minutes (SE 4.7 min in Table 3, p = 0.0003) compared with quiet control. Under the tested environmental-noise exposure (EN+earplugs vs. EN), physical earplugs mitigated N3 slow-wave sleep loss without the REM reduction observed under PN50; on secondary morning questionnaires, participants reported worse subjective sleep quality and alertness after EN and EN+PN nights versus quiet control (partially mitigated by earplugs), while next-morning cognitive, cardiovascular, and hearing-threshold tests did not differ significantly across conditions (all p > 0.65).
  • Decentralized home trials of bedtime soundscapes and binaural/monaural beats (Vazzaz et al., 2026; Shumov et al., 2023): In the 4-week USleep decentralized RCT (Vazzaz et al., 2026), recruitment ended before the planned sample size was reached because of budget constraints, with N = 495 working adults with poor sleep (Jenkins Sleep Scale ≥ 7) enrolled in the intention-to-treat sample (n = 408 at week 4) across Bedtime Stories, Sleep Sounds (music, nature sounds, and noise textures), Sleep Skills (guided behavioral and relaxation exercises), and an active digital control arm (Daily Boost). None of the three audio intervention groups demonstrated statistically significant superiority over the active control on the primary week-4 PROMIS Sleep Disturbance outcome (for Sleep Sounds vs. Daily Boost: g = 0.14, 95% CI −0.11 to 0.39, p = 0.281; indicating non-detected superiority over an active control rather than equivalence or comparison vs. silence). In a counterbalanced home trial (221 enrolled / 202 analyzed across 3 evening bedtime attempts), Shumov et al. (2023) found no significant difference in behavioral sleep-onset latency across monotonous sound with binaural beats, monaural beats, or sham audio without beats (ANOVA p = 0.21; pairwise adjusted p = 1.0). Small laboratory sleep-staging studies used stage-monitored or pre-screened protocols that differ from unmonitored home playback: in a 3-night laboratory study using EEG, EOG, and EMG for sleep staging, Jirakittayakorn & Wongsawat (2018) (N = 24 enrolled, 23 analyzed; 3 Hz stimulus on a 250 Hz carrier initiated at the first epoch of Stage N2 and stopped at the first epoch of Stage N3) reported longer N3 duration, shorter N3 latency, and higher whole-night sleep delta-band EEG power from baseline Night 2 to experimental Night 3 (within-group p = 0.0293; between-group change vs. control p = 0.0261). In a daytime nap polysomnography trial, Fan et al. (2024) (N = 12 pre-screened adults comparing sham and 0, 0.25, or 1 Hz binaural beats on a 250 Hz carrier; industry funding, two sponsor-employed co-authors, and a joint patent filing disclosed) reported shorter N2 latency (p = 0.023) and N3 latency (p = 0.045) at 0.25 Hz versus sham (uncorrected Wilcoxon comparisons) without significant neural entrainment or N3 duration change.

5. Anxiety and pain in specific clinical settings

  • Perioperative, procedural, and dental syntheses and trials (Bășeanu et al., 2024; Xiong et al., 2025; Fatima et al., 2026; O. Akkurt & Ekici, 2026; Elnazer, 2026; Kolhe et al., 2024): In a narrative systematic review of 12 studies (N = 1,349 across 10 procedural-anxiety studies, 1 anxiety study in Parkinson's disease, and 1 depression study), Bășeanu et al. (2024) noted short-term anxiety reductions primarily relative to passive care while several trials showed no clear advantage over ordinary music. In 15 surgical RCTs, Xiong et al. (2025) reported pooled reductions in perioperative anxiety (14 trials vs. blank control, n = 1,047, SMD = −1.38, I2 = 91.6%; 8 trials vs. non-binaural audio, n = 598, SMD = −1.01, I2 = 95.4%) and postoperative pain (5 trials vs. blank control, n = 433, SMD = −0.61, I2 = 75.8%; 3 trials vs. non-binaural audio, n = 265, SMD = −0.52, I2 = 0%), with extreme heterogeneity across anxiety endpoints. In 13 adult theta (4 to 8 Hz) binaural-beat RCTs (n = 630), Fatima et al. (2026) found that quantitative meta-analysis was feasible only for pain (4 RCTs, n = 173, SMD = −0.53, 95% CI −0.84 to −0.23, p = 0.001, I2 = 0%, GRADE low certainty), while anxiety and cognition were synthesized narratively without pooling. Conversely, in a 3-arm randomized clinical trial during impacted lower third-molar extraction (N = 90, 30 per group: binaural beats, isochronic tones, and no-audio control), O. Akkurt & Ekici (2026) found no statistically significant between-group differences in preoperative or postoperative anxiety across MDAS, DFS, STAI-S, or VAS scales (with isolated physiological differences at single intraoperative timepoints: lower incision SBP in the binaural-beat group p = 0.041 and lower anesthesia respiratory rate in the isochronic-tone group p = 0.015, and 78.3% patient preference for audio in future dental visits). Elnazer (2026) reviewed 10 trials combining music and binaural beats in young adults aged 19 to 24.

6. Why study results differ and what remains uncertain

Across the literature, five methodological factors explain why studies often report conflicting outcomes:

  1. Choice of comparator: Comparing headphone audio with silence estimates the effect of the tested listening package relative to silence under those conditions, without by itself establishing whether any change comes from acoustic masking, distraction, expectation, or a beat parameter. An active audio control reduces some confounding, but isolates a specific acoustic parameter only when the relevant conditions are otherwise adequately matched.
  2. Baseline population differences: As shown by Nigg et al. (2024), the same white or pink noise that modestly aids task performance in individuals with ADHD (g = 0.249) slightly impairs task performance in non-ADHD peers (g = −0.212 / −0.206).
  3. Acoustic environment and sound pressure level: Playing 40 to 60 dB white noise in a loud ICU ward masks disruptive peaks (Ding et al., 2025; Khandelwal et al., 2025), whereas playing continuous 50 dBA pink noise in a quiet bedroom can reduce REM sleep duration (Basner et al., 2026).
  4. Stimulus parameters and exposure duration: Studies vary from 10-second pupillometry clips (Erfanian et al., 2025) to a 14-minute sustained-attention task (Sudre et al., 2024) to 4-week home routines (Vazzaz et al., 2026).
  5. Multiple comparisons and selective non-reporting: Multi-arm EEG and behavioral studies often test many frequency bands, electrode sites, and subscales without preregistered primary endpoints. More broadly across intervention research, when the availability of studies or specific results depends on the direction or statistical significance of findings, published evidence can give a distorted picture, although the extent of missing evidence is not directly established for every audio-intervention topic (Page et al., 2024, Cochrane Handbook Ch. 13).

7. Comparative summary of key peer-reviewed studies and reviews

Question / study Population and design Intervention and comparator Main finding Key limitation
EEG entrainment to binaural beats
Ingendoh et al. (2023)
Systematic review of 14 human EEG studies Dichotic binaural beats across delta, theta, alpha, beta, and gamma rates vs. baseline/control Inconclusive: 5 studies consistent with entrainment, 8 negative/contradictory, 1 mixed. High methodological heterogeneity in carriers, durations, and EEG baselines.
Monaural vs. binaural ASSR & mood
Orozco Perez et al. (2020)
Within-subject laboratory EEG study (N = 16 healthy adults) 7 Hz and 40 Hz monaural and binaural beats (~400 Hz carriers) vs. baseline Both elicited carrier FFR and beat-rate ASSR (larger ASSR for monaural in tested conditions); functional connectivity patterns differed with beat frequency and stimulus type without supporting broad state entrainment; no statistically significant change in POMS mood was detected (vigilance not tested). Small laboratory sample; single-session exposure.
Home fluid intelligence & framing
Klichowski et al. (2023)
Two 2-part home trials (1,000 tested / 920 analyzed: 369 in Study 1; 551 in Study 2) Study 1: 15 Hz binaural beats after self-selected Neutral vs. Stimulating Sounds framing; Study 2: 5 Hz or 15 Hz binaural beats vs. silence, classical music, or 240 Hz pure tone on RPM & MaRs-IB Within-condition fluid-reasoning scores declined significantly from part 1 to part 2 during binaural-beat sessions; silence, music, and pure-tone controls showed no significant score change. Unsupervised home listening; self-selected framing in Study 1.
Parametric EEG & sustained attention
Melnichuk et al. (2025)
80 enrolled / 64 analyzed across a 2 × 2 × 2 × 2 between-subjects factorial design (2 crossover visits each) 16 Hz beta vs. 40 Hz gamma × 340 vs. 400 Hz carrier × onset (10 min prior vs. task start) × white noise (present vs. absent) vs. pure-tone control on IP-CPT Separate subgroup contrasts vs. matched control showed higher average target detection for 40 Hz + 340 Hz carrier (p = 0.002) and with white noise (p = 0.033; with task-onset stimulation, p = 0.013); background white noise attenuated the beat-frequency EEG response, particularly for gamma beats (four-way noise-moderation interaction p = 0.028); no overall vigilance-decrement attenuation (p = 0.751). Small cell sizes (average 4 analyzed, range 2 to 6 per between-subject cell).
Monaural vs. binaural, panning & alternating beeps
Sudre et al. (2024)
42 recruited / 38 analyzed (G6: n = 19 at 6 Hz; G40: n = 19 at 40 Hz) Binaural beats (C1), smooth panning (C2), and alternating beeps (C3) vs. monaural-beat control (C0) in randomized order Null effect on objective Not-X-CPT attention; within-condition pre–post relaxation increases in specific arms. Short exposures; within-arm pre–post change does not prove a between-condition difference.
White & pink noise for attention
Nigg et al. (2024)
Meta-analysis: 13 ADHD studies (N = 335) & 11 non-ADHD studies (N = 335) White or pink background noise during cognitive tasks vs. quiet/no-noise control Small benefit in ADHD (g = 0.249, p < 0.0001, GRADE moderate); small detriment in non-ADHD (g = −0.212 / −0.206). Zero brown-noise trials met inclusion criteria; predominantly short laboratory tasks.
Continuous pink noise vs. sleep architecture
Basner et al. (2026)
7-night laboratory PSG crossover trial (1 adaptation + 6 condition nights, n = 25 healthy adults) Quiet control, environmental noise (EN), pink noise 50 dBA (PN50), EN+PN40, EN+PN50, EN+earplugs via bedroom speakers PN50 alone reduced REM sleep by 18.6 min (SE 4.7 min in Table 3, p = 0.0003) vs. quiet control; under EN+earplugs vs. EN, earplugs mitigated N3 loss without the REM reduction seen under PN50; worse morning subjective ratings after EN and EN+PN nights (partially mitigated by earplugs); cognitive/cardiovascular/hearing tests null (all p > 0.65). Healthy adult lab cohort over single-night condition exposures.
White/masking noise for sleep in clinical & age subgroups
Ding et al. (2025); Khandelwal et al. (2025)
Meta-analysis of 12 white-noise RCTs (N = 1,301) in Ding; narrative systematic review of 7 RCTs (N = 496 hospitalized adults) in Khandelwal Overnight white noise (or USASI masking noise in 1 trial) vs. routine care / control Mixed infant endpoints in Ding (24-h TST and 12-h SE improved; 12-h nocturnal TST, 24-h SE, and 24-h WASO null); lower adult/ICU PSQI with high heterogeneity; Khandelwal noted improvements in sleep efficiency and study-level subjective sleep questionnaires (PSQI, RCSQ, VSH) alongside 8 participants who discontinued for discomfort or found the intervention less acceptable. High heterogeneity (preventing meta-analysis in Khandelwal); hospital ward masking does not generalize to quiet bedrooms.
Bedtime audio interventions at home (USleep)
Vazzaz et al. (2026)
4-week decentralized 4-arm RCT (recruitment ended early for budget constraints; N = 495 enrolled in ITT, n = 408 at week 4; working adults with Jenkins Sleep Scale ≥ 7) Bedtime Stories (narrated stories), Sleep Sounds (music, nature sounds, and noise textures), Sleep Skills (guided mindfulness and relaxation exercises) vs. Daily Boost active control (short daytime educational/motivational tracks) None of the 3 intervention arms was significantly superior to active control on primary week-4 PROMIS Sleep Disturbance (Sleep Sounds vs. control: g = 0.14, 95% CI −0.11 to 0.39, p = 0.281). Tests superiority vs. active control (not equivalence or vs. silence); self-reported outcomes; author industry affiliation disclosed.
Perioperative & dental procedural anxiety and pain
Xiong et al. (2025); Fatima et al. (2026); O. Akkurt & Ekici (2026)
Surgical meta-analysis (15 RCTs in Xiong), theta BB review (13 RCTs, n = 630 in Fatima), and 3-arm dental RCT (N = 90 in Akkurt & Ekici) Binaural beats or isochronic tones vs. blank/no-audio control or non-binaural audio Xiong pooled anxiety (SMD = −1.38 vs. blank, I2 = 91.6%) and pain reductions; Fatima pooled pain only (4 RCTs, SMD = −0.53, GRADE low); Akkurt & Ekici found null between-group anxiety differences across all 4 scales (N = 90). Extreme heterogeneity in pooled anxiety (I2 > 90%), unblinded controls, and null anxiety findings in individual RCTs.
Detailed subgroup numbers and additional experimental studies
  • Ding et al. (2025) subgroup details: Across 12 white-noise RCTs (N = 1,301: 356 infants/toddlers 0 to 3 y, 648 adults 18 to 64 y, 297 older adults ≥ 65 y; 544 ICU and 401 non-ICU adults), infant/toddler analyses in the PubMed-indexed abstract reported longer 24-hour total sleep time (MD = 137.51 min, 95% CI 67.80 to 207.23, p = 0.0001), higher 12-hour sleep efficiency (MD = 6.62 percentage points, 95% CI 1.72 to 11.52, p = 0.008), and fewer 12-hour nocturnal awakenings (MD = −1.83, 95% CI −3.12 to −0.54, p = 0.006), but no significant effect on 12-hour nocturnal total sleep time (MD = 102.56 min, p = 0.16), 24-hour sleep efficiency (MD = 7.77%, p = 0.46), or 24-hour wake after sleep onset (MD = −8.65 min, p = 0.46). Adult (MD = −3.70, 95% CI −4.90 to −2.50, p < 0.001), older adult (MD = −2.71, 95% CI −4.98 to −0.44, p = 0.02), ICU (MD = −4.04, 95% CI −6.35 to −1.73, p < 0.001), and non-ICU (MD = −2.61, 95% CI −3.84 to −1.38, p < 0.001) subgroups showed lower PSQI scores with high statistical heterogeneity in the adult PSQI analysis.
  • Nigg et al. (2024) exact reporting note: For the non-ADHD comparison group (k = 11, N = 335), the published article's abstract reports g = −0.212 (95% CI −0.355 to −0.069, p = 0.0036), whereas the main Results text reports g = −0.206 (95% CI −0.330 to −0.081, p = 0.001). Both figures represent the same small negative direction of effect.
  • Erfanian et al. (2025) pupillometry: In 38 recruited / 31 analyzed adults listening to 10-second white, pink, and brown noise clips, no significant difference in pupil-size modulation was detected between noise colors.
  • Kolhe et al. (2024) pediatric dental extraction: Randomized 40 children aged 8 to 12 years (4 groups of 10: white, pink, brown noise, and cartoon music) during primary tooth extraction, measuring acute dental anxiety and pain rather than sleep or sustained attention.