The accurate identification of Photoparoxysmal Responses (PPR) is clinically important, as these epileptiform discharges are the key biomarkers used in diagnosing photosensitivity. Traditionally, PPR are identified in electroencephalography (EEG) during Intermittent Photic Stimulation, but manual detection is time-consuming and subjective, highlighting the need for reliable automated approaches. Recent studies have shown that Gamma and High-Frequency Oscillations (HFO) are promising biomarkers in epilepsy research, potentially improving the precision of diagnosis. Building on this, our study incorporates higher-frequency analysis into photosensitivity to investigate how it contributes to identifying PPR. We use an unsupervised anomaly detection method based on a Variational Autoencoder trained on EEG recordings from healthy individuals. Results show that excluding Gamma and High-Gamma rhythms greatly decreases the model’s detection Accuracy from 83% in prior full-spectrum analysis to 63%, supporting the idea that higher frequencies are key in the distinction of PPR, which aligns with broader epilepsy research, where high-frequency activity has proven to be a key marker of pathological brain dynamics. This approach could help integrate high-frequency biomarkers into clinical decision-support systems as well as support automated EEG analysis in other epilepsy syndromes beyond photosensitivity.