F1000Research F1000Research 2046-1402 F1000 Research Limited London, UK 10.12688/f1000research.174232.1 Research Article Articles Emotion Detection of People with Autism Ages 10–20: Using Beta-Left and Gamma-Right Temporal

[version 1; peer review: 1 approved with reservations]

 Murad Omayya Conceptualization Data Curation Formal Analysis Investigation Methodology Resources Software Validation Visualization Writing – Original Draft Preparation https://orcid.org/0000-0003-0527-3035 a 1 Malkawi Mohammad Conceptualization Data Curation Formal Analysis Investigation Methodology Project Administration Supervision Validation Writing – Original Draft Preparation Writing – Review & Editing 2 Khadum Methaq Formal Analysis Investigation Methodology Validation Visualization Writing – Review & Editing 3 Faouel Nour Methodology Validation 4 Batterjee Khalid Writing – Original Draft Preparation 5 Dashi Samer Supervision 6 Leberrara Hala Investigation 7 Arafat Ahmad Data Curation Formal Analysis Resources Software 8 Computer Science, Middle East University Faculty of Information Technology, Amman, Amman Governorate, Jordan Jordan University of Science and Technology, Irbid, Irbid Governorate, Jordan Mustansiriyah University, Baghdad, Baghdad Governorate, Iraq Psychiatry, University of Monastir Tunis, Tunis, Tunisia Saudi German Hospital Jeddah, Jeddah, Makkah Province, Saudi Arabia The University of Texas at Dallas, Richardson, Texas, USA University of Ahmed DRAIA, Adrar, Algeria Artificial Intelligence, Middle East University Faculty of Information Technology, Amman, Amman Governorate, Jordan o.murad@meu.edu.jo

No competing interests were disclosed.

 28 1 2026 2026 15 131 14 1 2026 Copyright: © 2026 Murad O et al. 2026 This is an open access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Abstract* Background

Autism Spectrum Disorder (ASD) diagnosis currently relies on behavioral assessments, which can be time-consuming and subjective. This study investigates whether minimal EEG band measurements can differentiate individuals with ASD from neurotypical peers, providing a potential rapid screening tool.

 Methods

A proof-of-concept study was conducted with twelve participants aged 10 to 20 years, including five neurotypical controls and seven ASD residents from the Arabic Village for Special Education Centre. Brain activity was recorded using four-sensor Muse 2 headband covering the temporal and front regions. The primary EEG frequency bands- Alpha, Beta, Gamma, Delta and Theta- were measured and statically analysed for each participant.

 Results

The EEG signals from the temporal regions were the most informative for distinguishing between groups. Specifically, gamma activity at the right temporal sensor (TP10) and beta activity at the left temporal (TP9) demonstrated the highest capacity to differentiate ASD individuals from controls.

 Conclusion

These preliminary findings suggest that brief, simple EEG measurements, focusing on specific frequency bands in temporal regions, could rapid screening or monitoring of ASD behaviour in clinical and educational settings. Future research with larger samples and enhanced data collection is necessary to validate and expand upon these results.

 Autism Spectrum Disorder Emotion Detection EEG Muse 2 Physiological Signals Youth with Autism Behavioural Monitoring Early Warning Federation of Arab Scientific Research Councils (FASRC) This research was supported by a $30,000 grant from the Federation of Arab Scientific Research Councils (FASRC). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Introduction

AI has recently had a major impact on human medical research and applications that call for sophisticated representation, computationally demanding calculations, and decision-making procedures. 1– 4 The primary goal of this work is to identify and detect human emotional responses in individuals with Autism Spectrum Disorder (ASD), which encompasses a wide range of neurodevelopmental conditions. People with ASD frequently display repetitive behavioural patterns, and their emotional responses are strongly associated with difficulties in communication and social interaction. 5, 6 Autistic people frequently exhibit irregular emotional behaviour, which can be challenging to understand or identify using standard observational techniques. Self-harm, hand flapping, rocking, and prolonged staring are common behaviours associated with autism. Rather of being misbehaviour, such conduct usually reflects stress, overburden, or unfulfilled needs. Giving physicians and caregivers a novel physiological approach to deduce the internal states of individuals with ASD is one of the goals of our work. 1– 4 We investigate an option that uses a simple EEG montage and wearable neurophysiology. Numerous researchers have used EEG signals extensively to analyse human-specific behaviour. 7, 8

By developing a distinct differentiator between ASD behavioural response and otherwise normal reactions, this work seeks to build a reference base for autistic behaviour response. The present study creates a testbed for further research in which we assess the EEG signals of people with and without ASD. Keep in mind that the objective is to find tiny, comprehensible signal features that accurately identify symptoms of ASD rather than to construct a model for brain activity in general. In order to facilitate early screening and emotion-aware intervention, such identification is needed.

This method expands on earlier research on emotion recognition utilizing viable, small signal sets, where the researchers looked at fifteen distinct physiological markers. EEG, skin conductivity, blood pressure, heart rate variability (HRV), and other factors were assessed. The importance of EEG signals, which frequently independently disclose respondents' inner emotional status, was one of the study's main conclusions. 1– 4

In order to determine if a small collection of EEG characteristics may distinguish ASD from neurotypical controls, this study offers a proof-of-concept analysis using a four-sensor headband. Over a period of ten to fifteen minutes, the EEG bands (alpha, beta, delta, gamma, and theta) are measured. Four distinct sensors are positioned at the AF7/AF8 (frontal) and TP9/TP10 (temporo-parietal) to measure each band. 9

Five neurotypical controls and seven people with autism spectrum disorder (ASD) are involved in the pilot experiment. The findings set the stage for a longer follow-up with more participants and recordings that are controlled for artifacts.

In summary, the objective of this article aligns with several sustainable development goals (SDGs), such as: SDG 3: Good Health and Well-Being, by helping to identify the emotions of individuals with ASD, which may help clinicians and parents understand their needs. By empowering the educational process for people with ASD through the use of emotion detection tools, it also advances SDG 4: Quality Education. It also aligns with SDG 10: Reduced Inequalities, which is exemplified by reducing educational and social barriers. Additionally, our endeavour advances SDG 9: Industry, Innovation, and Infrastructure by introducing a novel health data-driven innovation.

 Methods Participants and data analysis

The authors of this study examined the EEGs of twelve patients, including five neurotypical controls (n = 5) and seven individuals with ASD (n = 7). Alpha, beta, delta, gamma, and theta bands are the five band-isolated time series that each participant's deployed sensors monitor and report. The data taken by the patients using a written ethical approval received by the ‘Arab Village for Special Challenges’ under approval number (ز/1810). Two sensors were positioned in front of the subject's head for this arrangement, specifically the AF7 sensor on the left and the AF8 sensor on the right, or above the eyes. However, as seen in Figure 1, two additional sensors, TP9 and TP10, were placed toward the rear, or close to the ears (TP9 on the left mastoid region and TP10 on the right mastoid region). The sensor data is reported and stored as channels 1, 2, 3, and 4 using Muse 2 technology. Before statistical testing and modeling, the sessions that some participants completed were combined inside the subject.

Muse sensors setup for EEG measurement.

The most challenging aspect of this study was gathering data from participants with ASD; the authors intend to include at least 30 people with ASD (ages 10–20) from nearby clinics and centers in their next investigation. Participants who can tolerate wearing a head-mounted sensor and have a confirmed diagnosis of ASD will be included. People with other neurological conditions, including epilepsy, that could interfere with EEG recordings will not be included by the authors. 10, 11

To put it succinctly, all study methods will be carried out with Institutional Review Board (IRB) permission, and prior to participation, participant assent and parental agreement will be sought.

The Muse 2 headband is used in this study to take measurements. 12, 13 Although Muse 2 may record many metrics, the pilot study solely looks at EEG data. The authors of a prior study 4 employed a NeXus Q32 system with 14 physiological indicators, including EEG, to identify human emotions. EEG signals (alpha, beta, delta, gamma, and theta) may be adequate to identify the emotion without the need to measure additional signals, according to earlier thorough research. Actually, the primary problem with depending solely on EEG is that it is very challenging, if not impossible, to use EEG signals in a non-invasive way.

The measuring sessions take place in a pleasant environment with video observation, and they last only ten to fifteen minutes. Every session involving subjects with ASD was carried out under the guidance and assistance of a therapist, who assisted in keeping the sessions organized and ended them when needed. As a result, this illustrates the challenges the writers encountered when gathering information for their paper.

When using this type of data, where participation is voluntary and withdrawal is possible at any time, confidentiality is obviously essential. An example of the data is shown in Table 1. With an average duration of 15 minutes, each data capture yielded over 250,000 readings, or a reading every 3.6 milliseconds for one normal/healthy (X) and one ASD (Y).

Sample data (mean power (μV <sup>2</sup>) per electrode per band).
Group Subject Session Band P TP9 P AF7 P AF8 P TP10
Healthy X_Person_ Session1 alpha 103.3872 48.77104 77.27053 116.2682
Healthy X_Person_ Session1 beta 82.67007 60.86879 66.44008 102.7144
Healthy X_Person_ Session1 delta 1825.627 2872.767 954.7735 3646.692
Healthy X_Person_ Session1 gamma 62.09157 47.28045 35.31806 71.73652
Healthy X_Person_ Session1 theta 304.5876 182.4597 203.7933 371.7244
Autism Y_Person Session1 alpha 3619.203 3450.599 2157.096 2391.242
Autism Y_Person Session1 beta 2936.616 3227.962 1660.67 1790.658
Autism Y_Person Session1 delta 23297.29 25694.04 18576.49 16746.58
Autism Y_Person Session1 gamma 2316.774 2975.258 1704.566 1336.96
Autism Y_Person Session1 theta 7521.203 7677.439 4044.538 4844.26

The power per band per electrode (P band,e) is computed as the mean of squares of the time series. For each subject and band: P band , e = 1 N ∑ i = 1 N ( x i , e ) 2 Where: P band,e is the band power at electrode e (TP0, AF7, AF8, TP10); N is the number of reading samples; x i , e is the power at electrode (e) at sample time i. For each subject, 250,000 samples were collected with 5 milliseconds separation between successive readings.

To reduce inter-subject scale differences, the relative power at each electrode is given by RelativePower band , e = P band , e ∑ b ∈ { δ , θ , α , β , γ } P b , e

Statistical analysis

This study uses a number of additional statistics to give a thorough comparison of EEG properties between the Autism and Healthy groups. To get a statistically sound conclusion, the authors used the relevant statistical analysis in addition to the relative power band displayed above. When data are independent and group variances are roughly similar (pooled-variance assumption), the independent-samples t-test, which assesses whether the mean value varies between the groups, is appropriate. Welch's t-test (unequal variances) is recommended if variances are obviously different. 14, 15

Hedges' g is also used to measure the standardized effect size, or the practical magnitude of the difference. Hedges' g is a standardized mean difference that, after applying a small-sample bias correction, indicates the distance between two group means in units of their pooled standard deviation. Hedge’s g is computed as: g = j x d ; where j = 1- 3 4 ( n 1 + n 2 ) − 9 ; and d = ( x 1 − x 2 ) Sp ;

x1, x2 are group means and Sp is the standard deviation Sp = ( n 1 − 1 ) s 1 2 + ( n 2 − 1 ) s 2 2 n 1 + n 2 − 2

“g” is expressed in units of pooled standard deviations and corrected for small-sample bias. In fact, g values are considered to be as follows; small as |g| ≈ 0.2, medium as |g| ≈ 0.5 and large as |g| ≈ 0.8. The power value and Hedges’ g are helping to understand the statistical significance as well as the magnitude and potential practical relevance of the effect. 16, 17

For multiple tests (ADS and Control with 5 different bands each), we use the false discovery rate (FDR) using Benjamini et al. method. 18, 19

We employ the Area Under the ROC Curve (AUC), where ROC stands for Receiver Operating Characteristic, to account for discrimination performance. The AUC calculates the likelihood that a randomly selected positive rate will be ranked higher than a randomly selected negative rate using the Mann–Whitney formulation. 20– 22 AUC (area under the curve) describes discrimination (1.0 = perfect, 0.5 = chance); ROC, according to Mann-Whitney, is a curve that displays True Positive Rate (sensitivity) vs. False Positive Rate (1 − specificity) across all decision thresholds. The following section goes into more detail on the application of the AUC and ROC. Future studies will employ a bigger sample size to examine the existing results utilizing the AUC approach in more detail.

We employed leave-one-subject-out (LOSO) cross-validation to estimate performance on unseen individuals and avoid leaking from repeated sessions. In each test, the model was trained on the remaining recordings while all of the recordings from a single person were kept out for testing. After that, we assessed three lightweight models and presented three ROC curves:

Two-feature model using relative beta at TP9 and relative gamma at TP10 (features selected a priori from univariate screening).

Log-band model using the log10 of mean power in alpha, beta, gamma, delta, and theta.

The Combined model included the five log-band features plus the two relative features (7 total). All analyses were run using CSV files exported from the Muse 2 Application.

No further artifact rejection was applied beyond the preprocessed data.

 Results & Discussion

Figures 2A– 2E show the difference in average power alpha, beta, delta, gamma and theta bands; each figure shows the overall power average for the four sensors/channels (AF7/AF8 and TP9/TP10). The blue bars in the figure represent the normal control subjects, while the yellow bars represent the ADS) subjects. The average power of a given band, e.g., alpha, is computed as the average across all participants (e.g. with ADS) over a given sensor, for example TP9. Then we take the average across all ADS participants. Since the sample is small, we use an error range by computing the deviation of the average from the standard deviation. Figures 2A– 2E shows the summary of the results, with the yellow bar indicating the relative power average for the ADS sample, while the blue bar represents the normal control sample. The thin black line represents the error bar for each result. For example the Alpha@TP9 autism mean ≈ 661.4 with SE (error) ≈ ∓ 495.8. In other words, the Alpha@TP9 range is (661.4-495.8) to (661.4+495.8) or minimum of 165.6 to maximum of 1157.2.

Alpha band, Beta band, Gamma band, Delta band, Theta band: Group Means ± SE.

Across all the Figures 2A– 2E, the Autism group’s exhibit higher power than the Healthy group, indicating higher absolute power on average. However, the relative β at TP9 and relative γ at TP10, are the clearest discriminators of ASD vs. controls.

The absolute-power differences observed for gamma at TP10 (right temporal) with a Hedges’ g ≈ 0.94 (Autism > Healthy) indicating large effect. For beta at TP9 (left temporal) with g ≈ 0.79 (Autism > Healthy), the difference index is moderate to large.

Using the single–band discrimination based on log10 overall band power (per subject), the gamma band ranked highest (AUC ≈ 0.71), with beta at AUC ≈ 0.69 next highest. Alpha showed modest overall power at (AUC ≈ 0.63), and theta/delta at (AUC ≈ 0.57/0.54).

Figure 3 summarizes the results for the Area under ROC curve, where we utilized the leave-one-subject-out (LOSO) principle. The figure illustrates the Receiver Operating Characteristic (ROC) for the three models (two-feature, log band, and combined 7 features)

Muse sensors setup for EEG measurement.

The two-feature model (Beta and Gamma bands) with AUC = 0.77, reveals a focused classifier using the beta band at TP9 and gamma band at TP10. Note that the entire 2-factor curve sets above the diagonal line, which represents the medium chance (FPR=TPR). The two feature model exhibited the best discrimination between ADS subjects and normal control subjects. Hence, the two features model may be sufficient to do most of the work in distinguishing between the groups. This findings, however, needs to be further investigated with more experiments and subjects, especially in light of other research results, which emphasize the alpha band. 23

The log-bands baseline (AUC = 0.20) is a broader feature set using log-transformed band power across all sensors (alpha, beta, gamma, theta, delta). LOSO. The entire ROC curve falls below the diagonal, which represents the 0.5 chance (FPR=TPR). This indicates weak or inconsistent signal in these generic summaries for this dataset.

The combined model with AUC = 0.71 is a representation of both models, the 2 feature model and the log band model. The combined model outperformed the log-bands baseline. However, it trailed behind the two-feature model. This behavior indicates that adding weak or collinear features may weaken the signal and reduce generalization.

Even, at low false-positive rates, performance remains meaningful despite the small sample. For example, at FPR ≈ 0.20 (1/5 healthy misclassified), the two-feature model detects ≈57% of Autism cases. The combined model (at FPR = 0.2) detects ≈71%. At FPR ≈ 0.40, the same pattern holds (two-feature ≈57%, combined ≈71%).

The alpha effects in our dataset are modest ( Figure 4), although, several studies detect alpha abnormalities in Autism. Figure 4 shows that only at >0.8 FPR, the Alpha band begins to detect autism cases. This does not necessarily contradict the alpha-centric literature, were these studies frequently rely on other sensors such as eyes-closed reactivity, individual alpha peak frequency, or connectivity measures. Our experiments continue to reliably detect autism to a good degree without the use of these extra parameters. Within these constraints, the temporal beta/gamma measures provide strong enough and reliable discrimination between the two sets.

Alpha ROC Plot.

The ROC overlay demonstrates that a minimal, two-feature sensors, namely the relative beta at TP9 and the relative gamma at TP10, deliver the highest cross-validated AUC and favorable sensitivity. This avoids the need to rely on a larger sensor array or complex preprocessing. This simple design is practical for real-world screening. The two temporal sensors are fast to place, and relative features aren’t easily thrown off by changes in overall amplitude. Table 2 provides a summary of key performance/results.

Summary of key parameters.
Model/Feature (sensor) Metric Value Notes
Two-feature model (TP9 β + TP10 γ) AUC (LOSO) 0.77 Highest cross-validated discrimination.
Combined model (7 features) AUC (LOSO) 0.71 Mixed relative + log-band features.
Log-bands baseline (5 features) AUC (LOSO) 0.2 ROC mostly below chance.
Single-band: γ (overall power) AUC 0.71 Per-band screening.
Single-band: β (overall power) AUC 0.69 Per-band screening.
Single-band: α/θ/δ (overall power) AUC 0.63/ 0.57/0.54 Per-band screening.
Relative γ @ TP10 Hedges’ g ≈0.94 Large effect (Autism > Healthy).
Relative β @ TP9 Hedges’ g ≈0.79 Moderate–large effect (Autism > Healthy).
Band-wise group tests p-values Not reported Independent/Welch t-tests; FDR controlled across bands.

Even with ROC curves and LOSO cross-validation, models can appear overly optimistic especially with relatively low sample. However, in our study this risk is acknowledged, and remedied by aggregating repeated sessions within subject, used leave-one-subject-out (LOSO) to hold out entire participants, and controlled multiple testing via FDR, steps intended to reduce overfitting and selection bias. We also emphasized effect sizes (Hedges’ g) alongside ROC to avoid relying on a single metric and noted that findings are preliminary pending larger, multi-site validation with stricter artifact control and broader coverage, which we outline as future work.

Ethical considerations

Because the study involved minors, informed consent was obtained from the participants’ legal guardians. The study received ethical approval from ‘Arab Village for Special Challenges’ under approval number (ز/1810), and all procedures followed the Declaration of Helsinki guidelines. The consent was verbal, as approved by the ethics committee at Arab Village for Special Challenges, due to cultural considerations, and logistical reasons. The approval of the research was granted after review of the research protocol to ensure compliance with ethical standards for research involving human participants. Generally, all disabilities centers in Jordan are following the rules under ‘The Higher Council for the Rights of Persons with Disabilities’, and Arab Village for Special Challenges is listed under this council and follow its rules. The centre has as ethics review process that reviews research proposals prior to approval that aligned with the Higher Council for the Rights of Persons with Disabilities specifications. All legislation including (laws, instructions and international agreements) are available in Higher Council for the Rights of Persons with Disabilities website.

 Data availability

Due to the high level of privacy and secrecy surrounding the individuals who participated in the research's experiments, the data are not publicly accessible. However, they can be obtained from the corresponding author upon reasonable request.

Reporting guidelines

This study involves observational research following the STROBE guidelines. The completed checklists and flowcharts are available upon request from the corresponding author. Due to privacy considerations, the underlying data are not publicly available but can be shared with qualified researchers upon reasonable request. However, access to the data can be requested by contacting the research team at ( o.murad@meu.edu.jo; mimalkawi@just.edu.jo; methaq@uomustansiriyah.edu.iq), and approval will be granted upon review of the request and agreement to maintain participant confidentiality and comply with ethical standards.

 Acknowledgements

We are appreciative of the clinical therapists, data science volunteers, autistic groups, and participating families that helped make this project feasible. Above all, we thank the Federation of Arab Scientific Research Councils (FASRC) for providing support and funding this research. We would also want to thank Dr. Taima Hiyari, the director of Arab Village for Special Challenges, for helping to make the study possible.

 References Malkawi M Murad O : Artificial neuro fuzzy logic system for detecting human emotions. HCIS. 2013;3(1):3. Albashiti A Malkawi M Khasawneh M : A Novel Optimization Algorithm To Detect Human Emotions Using Artificial Intelligence Techniques. Proceedings of The 25th International Business Information Management Association Conference «Innovation Vision 2020: From Regional Development Sustainability to Global Economic Growth», May 7-8, 2015. Amsterdam, Netherlands:2015;1256:1237. Murad O Malkawi M : An Optimization Method Using Clustering Technique for the Human Emotions Detection Artificial Neuro-Fuzzy Logic System. Int. J. 2016;15(9). Albashiti AI Malkawi M Khasawneh A : A novel neuro-fuzzy model to detect human emotions using different set of vital factors with performance index measure. Journal of Communications Software and Systems. 2018;14(1):121–129. Stretch LS Eldredge K : This the Diagnostic and Statistical Manual of Mental Disorders (text rev.; DSM-5-TR ®). DSM-5-TR ® and Family Systems. 2023;209. Mazefsky CA Herrington J Siegel M : The role of emotion regulation in autism spectrum disorder. J. Am. Acad. Child Adolesc. Psychiatry. 2013;52(7):679–688. 23800481 10.1016/j.jaac.2013.05.006 Alarcao SM Fonseca MJ : Emotions recognition using EEG signals: A survey. IEEE Trans. Affect. Comput. 2017;10(3):374–393. Lotte F Bougrain L Cichocki A : A review of classification algorithms for EEG-based brain–computer interfaces: a 10 year update. J. Neural Eng. 2018;15(3):031005. 29488902 10.1088/1741-2552/aab2f2 Oostenveld R Praamstra P : The five percent electrode system for high-resolution EEG and ERP measurements. Clin. Neurophysiol. 2001;112(4):713–719. 11275545 Tuchman R Rapin I : Epilepsy in autism. The Lancet Neurology. 2002;1(6):352–358. 12849396 Lukmanji S Manji SA Kadhim S : The co-occurrence of epilepsy and autism: A systematic review. Epilepsy Behav. 2019;98:238–248. 31398688 10.1016/j.yebeh.2019.07.037 InteraXon Inc: Muse 2: The brain sensing headband [Hardware]. InteraXon Inc.;2018. Reference Source InteraXon Inc: Muse 2 user manual. Toronto, Canada: InteraXon Inc;2018. Welch BL : The generalization of ‘STUDENT'S’ problem when several different population varlances are involved. Biometrika. 1947;34(1-2):28–35. 20287819 Delacre M Lakens D Leys C : Why psychologists should by default use Welch's t-test instead of Student's t-test. International Review of Social Psychology. 2017;30(1):92–101. Hedges LV : Distribution theory for Glass's estimator of effect size and related estimators. J. Educ. Stat. 1981;6(2):107–128. Hedges LV Olkin I : Statistical methods for meta-analysis. Orlando, FL: Academic Press;1985. Benjamini Y Hochberg Y : Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal statistical society: series B (Methodological). 1995;57(1):289–300. Benjamini Y Yekutieli D : The control of the false discovery rate in multiple testing under dependency. Ann. Stat. 2001;1165–1188. Bamber D : The area above the ordinal dominance graph and the area below the receiver operating characteristic graph. J. Math. Psychol. 1975;12(4):387–415. Hanley JA McNeil BJ : The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology. 1982;143(1):29–36. 7063747 DeLong ER DeLong DM Clarke-Pearson DL : Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988;44:837–845. 3203132 Hammed SH Albahri AS : Unlocking the potential of autism detection: Integrating traditional feature selection and machine learning techniques. Applied Data Science and Analysis. 2023;2023:42–58. 10.5256/f1000research.192116.r464742 Reviewer response for version 1 Gunturu Sasidhar 1 Referee https://orcid.org/0000-0002-6867-2394 Bronx-Lebanon Hospital Center, New York, New York, USA

Competing interests: No competing interests were disclosed.

 28 3 2026 Copyright: © 2026 Gunturu S 2026 This is an open access peer review report distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. recommendation approve-with-reservations

1. TITLE

The title "Using Beta-Left and Gamma-Right Temporal" is grammatically incomplete and reads like a fragment. It does not form a coherent noun phrase. Revise to something like: "Detecting Emotional Responses in Youth with Autism Spectrum Disorder Using Temporal Beta and Gamma EEG Features from a Wearable Headband: A Pilot Study."

The phrase "Emotion Detection" is somewhat misleading. The study does not detect specific emotions (e.g., happiness, fear, sadness); it attempts to classify ASD vs. neurotypical status. 

2. ABSTRACT

The claim that findings "suggest brief, simple EEG measurements could [support] rapid screening" is a strong translational claim not fully supported by n=12, no artifact control, verbal consent only, and a single-site design.

3. INTRODUCTION

The introduction uses "emotion detection" and "ASD classification" interchangeably throughout. These are fundamentally different tasks. Detecting that someone has ASD is a diagnostic/screening problem; detecting what emotion a person with ASD is experiencing is a decoding problem. The paper conflates both without ever explicitly defining the target variable. If the goal is group discrimination (ASD vs. neurotypical), the study should be clearly framed as a classification feasibility study, not an emotion detection study. This inconsistency undermines the scientific coherence of the entire manuscript.

The references lean heavily on the authors' own prior work (references 1–4 cited repeatedly). The EEG-ASD literature is rich and growing; the introduction makes no mention of competing or complementary approaches (e.g., EEG connectivity biomarkers, mu-rhythm suppression, P300-based ASD studies, or prior consumer-EEG ASD work). This creates an incomplete scholarly context and risks the appearance of self-citation bias.

Why specifically AF7/AF8 and TP9/TP10 ? The Muse 2 is constrained to these positions by hardware design, but the scientific motivation for these electrode positions in the context of ASD/emotion should be articulated.

A proof-of-concept study should at minimum state directional hypotheses (e.g., "We hypothesized that gamma and beta power would be elevated in ASD participants relative to controls").

4. METHODS

The authors explicitly state: "No further artifact rejection was applied beyond the preprocessed data." EEG recorded from children with ASD during unconstrained sessions is typically dominated by motion artifacts, muscle activity (EMG), eye blinks, and jaw clenching. The Muse 2's onboard preprocessing does not constitute a validated artifact rejection pipeline. Without artifact removal, band power values — especially gamma (which is highly susceptible to EMG contamination) and delta (susceptible to motion) — cannot be reliably attributed to neural sources. The finding that gamma at TP10 is the top discriminator is therefore potentially an artifact detection result rather than a neuroscientific finding.

The power computation formula (mean of squares of time series) is described, but it is not clarified whether the time series are already bandpass-filtered before this computation (as would be expected from Muse 2 output). This needs to be made explicit.

The Muse 2's proprietary preprocessing is not described. The device applies internal signal conditioning that is not fully transparent to me. This is a reproducibility concern.

Table 2 explicitly states "p-values: Not reported" for band-wise group tests. This is a fundamental omission. If t-tests were performed, the p-values and their FDR-adjusted counterparts must be reported in the manuscript. The reader cannot evaluate statistical significance claims without them.

5. RESULTS & DISCUSSION

Results and Discussion are merged without clear demarcation. Interpretive statements appear within the results section, and result values appear in the discussion. These should either be cleanly separated or explicitly labeled as a combined section with subsections.

Figure 2 error bars are labeled as SE but are described inconsistently. In the text: "we compute the deviation of the average from the standard deviation" — this describes SD, not SE. The example given (Alpha@TP9: mean ≈ 661.4, error ≈ 495.8) represents an enormous spread relative to the mean, suggesting the error bar is SD. The text and figure labels must be consistent and correct.

Individual-level data are not shown. With n=12, it is standard practice to show individual data points overlaid on group summaries (e.g., strip plots, beeswarm plots). This would allow readers to evaluate the distribution shape, outlier influence, and group separation directly.

The title claim "Emotion Detection" is not empirically supported in the Results section. No emotion-labeled stimuli were used, no emotional states were induced or verified, and no emotion categories were classified. The Results section reports only group-level ASD vs. control power differences. The paper should either add emotion elicitation methodology or rebrand the research question.

The discussion references one external paper (Hammed & Albahri, 2023) but does not systematically compare AUC, electrode selection, or sample characteristics with the broader EEG-ASD classification literature.

Clinical interpretation is overextended. Statements like "could [support] rapid screening or monitoring of ASD behaviour in clinical and educational settings" are not supported by n=12, no artifact control, single-site data, and no test of external validity.

The condition is repeatedly referred to as "ADS" (e.g., "ADS subjects," "ADS sample") instead of "ASD." This error appears in figures, tables, and body text and must be corrected uniformly throughout the manuscript.

The manuscript would benefit from professional English-language editing before resubmission.

Figure captions are minimal and do not describe what is shown in sufficient detail for standalone interpretation (e.g., Figure 2 captions do not state group definitions, units, or error bar type)

Is the work clearly and accurately presented and does it cite the current literature?

Partly

If applicable, is the statistical analysis and its interpretation appropriate?

Partly

Are all the source data underlying the results available to ensure full reproducibility?

Partly

Is the study design appropriate and is the work technically sound?

Partly

Are the conclusions drawn adequately supported by the results?

Partly

Are sufficient details of methods and analysis provided to allow replication by others?

Partly

Reviewer Expertise:

Psychiatry

I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.