GR-519 - Meditation as an Intervention to Improve Student Attention - An EEG Study Based on Machine Learning Prediction and Spectral Ratio Analysis

Aug 5, 2026

This research develops a machine learning model using EEG recordings to detect student inattention and support early intervention. Spectral ratio analysis focuses on frontal-lobe attention shifts, linking meditation effects to measurable changes in attention states. Training uses an attention dataset with young-adult EEG recordings and preprocessing steps including filtering and ICA. An intervention protocol based on guided meditation with music is evaluated through EEG and online test comparisons.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages1

Reading Time3 min

AccessFree to download

Abstract

Objective

Background

Method

Experiment

Results

Introduction

Why early detection? -Statistics

Meditation as an intervention: CM-II Stages

Emotional review

Analyzing challenges

Rehearsing solutions for a day

Methodology

Experiment

Experimental Design

Results

Conclusion

References

GR-519 Meditation as an intervention to improve Student Attention: An EEG study based on Machine learning prediction and Spectral Ratio Analysis Abstract Objective: This research aims to develop a machine learning model using EEG data to identify student inattention, serving as an early intervention tool. The model also does details spectral ratio analysis in frontal lobes to identify shift in attention. Background: Attention deficit, influenced by social media, adversely affects student performance. ADHD, characterized by inattention, hyperactivity, and impulsivity, is linked to academic challenges. Early detection in attention in academics is crucial to provide appropriate intervention. Method: A Machine Learning model was designed, trained on an attention dataset with 34 EEG recordings of young adults. The raw EEG data was pre-processed and filtered, ICA was applied, and spectral analysis was done. Guided meditation with music was developed as an intervention to improve attention. Experiment: EEG recordings from 15 young adults during a visual reasoning test and meditation assessed the model's efficacy by comparing model output to test scores. Results: ML: ML model achieved a 98% accuracy rate to classify attention vs inattention states. During the evaluation, it was found that the model predicted with similar accuracy in detecting people inattention states. CM-II: Key findings include Theta-to-Beta Ratio (TBR) indicating deepened introspection, Theta-toAlpha Ratio (TAR) showing enhanced relaxation, Delta-to-Theta Ratio (DTR) reflecting deep introspection, and Theta-to-Gamma Ratio (TGR) suggesting increased internal reflection and reduced thinking. ML model verified that post-meditation, all participants showed increased attention, supported by online tests, marking a significant improvement from initial inattention. Introduction • The rising prevalence of attention deficits in students, often linked to social media, adversely impacts their academic achievements. Inattention, characterized by a lack of focus or oversight of details, has been frequently correlated with academic underperformance in numerous studies. • Students diagnosed with Attention Deficit Hyperactivity Disorder (ADHD) - a neurodevelopmental condition marked by inattention, impulsivity, and hyperactivity - face higher risks of underachieving. Many of these students might even drop-out without securing a final degree. • An early detection can help manage attention levels and mitigate these effects. Why early detection? -Statistics • ADHD results in societal and joblessness costs for children and adolescents, approximated at $19.4 billion and $13.8 billion respectively in the US [1] . • In 2022, a 40% college dropout rate led to a $31 billion financial hit nationally [1] . • Traits of ADHD were observed to hinder academic success (e.g. , GPA) in engineering courses [2] . • Symptoms of inattention can lead to challenges in organizing, maintaining effort, and managing time [2] . Meditation as an intervention: CM-II Stages Guided meditation with background music was developed for daily practice . 1. Emotional review • Exploring memories of fear, hurt, etc. related to inattention. 2. Analyzing challenges • Reviewing past events of the day related to various aspects of life. • Finding possible solutions and actions within. 3. Rehearsing solutions for a day • Visualizing the solutions with positivity and indulging in related tasks with creativity, seeing solutions, and positive outcomes internally; Feeling gratitude. Methodology Raw EEG dataset filtered, feature engineered, extracted band waves and trained ML model to predict attention states. 89% accuracy (NN) and 98% accuracy (Ensemble) [4] . The model was trained using EEG attention dataset [6] that was labelled as attention vs inattention among 34 recordings of 5 participants [6] . Raw EEG dataset filtered, feature engineered, extracted band waves a

FAQs

What is the main goal of GR-519 in this study?

To build a machine learning model that uses EEG data to identify student inattention as an early intervention tool and to analyze attention shifts using frontal-lobe spectral ratios.

How was the meditation intervention (CM-II) designed and evaluated?

Guided meditation with background music was developed for daily practice, and efficacy was assessed using EEG recordings before, during, and after meditation while comparing model outputs with test scores.

What were the key outcomes of the machine learning model and the spectral ratio analysis?

The model achieved up to 98% accuracy to classify attention vs inattention, and spectral ratios (including TBR, TAR, DTR, and TGR) indicated cognitive effects consistent with improved attention after meditation.

Aug 5, 2026

Upload by Ava Thompson