Exploring EEG Signal Analysis for Perceived Number Classification and Transferability Across Individuals

Krambeer, Malte Leonard (2024) Exploring EEG Signal Analysis for Perceived Number Classification and Transferability Across Individuals. Bachelor's Thesis, Artificial Intelligence.

Preview	Text Krambeer2024Exploring-EEG-Signal-Analysis-for-Perceived-Number-Classification-and-Transferability-Across-Individuals.pdf Download (1MB) | Preview
	Text Toestemming Krambeer.pdf Restricted to Registered users only Download (154kB)

Abstract

This study explores the feasibility of classifying EEG signals recorded during number perception and the transferability of these models across individuals. Building on findings in cognitive neuroscience and machine learning, we examine distinct brain activity patterns in subjects visually perceiving numeric stimuli. EEG data were collected during a number-observation task using a consumer-grade, four-channel device. Our analysis focuses on signal distinguishability both within and across subjects to identify individual and subject-independent patterns. Although classification results do not support number-specific EEG distinguishability, statistical differences were observed in individual cases. Subjects vary in the informative value of their data, with more number-specific information found in tempo-parietal locations and 300-400 ms after stimulus presentation. The current methodology shows limitations in consistently detecting number-specific information within or across individuals. Future research should improve on our findings by employing advanced equipment, informative recording locations, temporal windows in EEG waveforms, and extensive data collection. We discuss the implications of these findings for current theory, previous EEG classification efforts, and practical brain-computer interface applications.

Item Type:	Thesis (Bachelor's Thesis)
Supervisor name:	Vortmann, L.
Degree programme:	Artificial Intelligence
Thesis type:	Bachelor's Thesis
Language:	English
Date Deposited:	16 Aug 2024 08:58
Last Modified:	16 Aug 2024 08:58
URI:	https://fse.studenttheses.ub.rug.nl/id/eprint/33975

Actions (login required)

View Item