EEG-Based Acrophobia Detection for Pilot Candidate Screening

Abstract

Acrophobia, or an excessive fear of heights, may compromise situational awareness, decision-making, and operational performance, making its early identification particularly relevant in pilot candidate screening. This study develops an electroencephalography (EEG)-based framework for distinguishing individuals with and without acrophobic tendencies using a Decision Tree classifier implemented independently in MATLAB and Python. EEG signals were acquired from 16 participants and preprocessed using WinEEG to obtain discriminative brainwave features for binary classification. The two implementations were evaluated under the same experimental conditions using classification accuracy as the primary performance metric. The results demonstrate that the Python-based Decision Tree achieved an accuracy of 98.68%, compared with 94.73% for the MATLAB Classification Learner implementation, corresponding to an improvement of 3.95 percentage points. These findings indicate that the Python implementation provided more effective discrimination between acrophobic and non-acrophobic participants under the evaluated experimental conditions. The results further demonstrate the feasibility of combining EEG-derived features with an interpretable machine-learning classifier for objective assessment of acrophobic responses. Although validation with a larger and more diverse cohort is required, the proposed framework provides a promising basis for developing data-driven neurophysiological screening tools to complement conventional psychological assessment in pilot candidate selection.