Skip to main content
2024 BMES Annual Meeting Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.

Neural Engineering

Neural Engineering - Poster Session E

Poster Z12 - Is blink-related neural processing a stable marker of cognition? Assessing the test-retest reliability of blink-related oscillations

Saturday, October 26, 2024
10:00 AM - 11:00 AM EST
Location: Exhibit Hall E, F & G

Introduction:  Blink-related oscillations (BRO) are newly discovered neurophysiological brainwave responses associated with spontaneous blinking, and correspond to information processing that occurs as the brain evaluates new visual information that appears after eye re-opening. Prior studies have demonstrated that these responses index a wide range of neural mechanisms including visual sensory processing, memory, and information processing, and also have potential as a digital biomarker for clinical brain function assessment. However, little is known about its reliability in repeated testing, which is an essential requirement for a potential diagnostic device for brain health evaluation. The current study undertook the first investigation of the test-retest reliability of BRO measurements to determine its stability and reliability as a potential digital biomarker of cognition.

Materials and

Methods:  We used publicly available 32-channel electroencephalography (EEG) data from 26 healthy adults performing an n-back working memory task with varying levels of task difficulty (Shin et al, 2019). Participants performed three runs each of the n-back task at three difficulty levels (0-back, 2-back and 3-back), totaling 30min. Artifacts were removed using independent component analysis, and blink instances were identified using a convolution-based template matching procedure. BRO responses were then extracted by segmenting the cleaned signals time-locked to blinking and averaging across trials in both time and frequency domains. BRO responses were quantified by computing the amplitudes in the two post-blink peaks in the time domain, as well as computing the spectral power following continuous wavelet transform (CWT) in the frequency domain. Effects were compared using repeated measures ANOVA. The within-session consistency of BRO responses were also determined using intraclass correlation between pairs of runs in both the time and frequency domains.

Results, Conclusions, and Discussions: Results showed that BRO responses were present and detectable within 3 repeated runs for each n-back task, with consistent time- and frequency-domain characteristics across all 3 runs (Figure 1a). BRO peak amplitudes were also comparable across all 3 runs for each n-back task, and all runs were able to capture significant changes in BRO amplitude compared to pre-blink baseline (Figure 1b). Additionally, significant intraclass correlations were also observed between pairs of runs in both time and frequency domains, suggesting that the morphological characteristics of this response were also consistent and reliable between repeated measurements. Moreover, individual-level BRO amplitude and latency were also found to exhibit significant high correlations between pairs of runs regardless of n-back task difficulty, indicating that these measurements were also consistent and reliable at the individual level (Figure 1c). Together, these findings demonstrate for the first time that blink-related neural processing measured through BRO responses exhibit within-session stability and reliability in repeated measurements at the individual level, providing promising evidence that blink-related neural processing may be utilized as a reliable biomarker in brain function assessments.

Acknowledgements (Optional):