On the Influence of Affect in EEG-Based Subject Identification

Abstract

Biometric signals have been extensively used for user identification and authentication due to their inherent characteristics that are unique to each person. The variation exhibited between the brain signals (EEG) of different people makes such signals especially suitable for biometric user identification. However, the characteristics of these signals are also influenced by the user's current condition, including his/her affective state. In this paper, we analyze the significance of the affect-related component of brain signals within the subject identification context. Consistent results are obtained across three different public datasets, suggesting that the dominant component of the signal is subject-related, but the affective state also has a contribution that affects identification accuracy. Results show that identification accuracy increases when the system has been trained with EEG recordings that refer to similar affective states as the sample that is to be identified. This improvement holds independently of the features and classification algorithm used, and it is generally above 10% under a rigorous setting, when the training and validation datasets do not share data from the same recording days. This finding emphasizes the potential benefits of considering affective information in applications that require subject identification, such as user authentication.

Image-Evoked Affect and its Impact on Eeg-Based Biometrics

“…Results using different EEG features and classification methods showed that EEG recordings referring to similar emotional states provided consistently higher identification accuracy than recordings referring to different emotional states, independently of the features, classification method, and signal acquisition session. Furthermore, the use in this work of a portable low-cost off-the-shelf EEG device [16], in contrast to the medical-grade EEG devices used in [5], shows that our finding is also valid for EEG devices with a lower spatial resolution (less electrodes).…”

“…Nevertheless, EEG-based biometrics pose new challenges, such as the problem of permanence of the recorded signals, i.e. finding patterns that are not heavily affected by template ageing, and the problem of defining suitable signal acquisition protocols in relation to the stimuli and the devices used [5].…”

“…As a result, varying emotional states can potentially act as noise within the acquired EEG signal, thus obstructing pattern matching. In [5], the authors studied the effect of varying emotional responses elicited by video stimuli in the context of EEGbased subject identification. Results showed that regardless the EEG features used, the machine learning method, and the time between signal acquisition, the use of EEG recordings associated with similar emotional states provided higher identification accuracy than the use of EEG signals associated with different emotional states.…”

“…Regardless the theoretical interest of the research reported in [5], the exposure time required for video stimuli has a significant influence on the applicability of the method in realistic contexts. In order to overcome this major limitation, we have extended this previous research by studying whether the effect reported also holds when images are used as the stimulus.…”

Adversarial Deep Learning in EEG Biometrics

“…Longitudinal studies also confirm the feasibility of EEG as an alternative means of biometrics [3][4][5]. However, one recent study demonstrates different affective mental states potentially influencing stability of EEG as a tool for user identification [6]. As such, discriminative EEG biometric feature extractor models that can filter out specific nuisance variables are likely to enhance usability of generated invariant features for biometric identification.…”