Evaluation of Artifact Subspace Reconstruction for Automatic EEG Artifact Removal

One of the greatest challenges that hinder the decoding and application of electroencephalography (EEG) is that EEG recordings almost always contain artifacts - non-brain signals. Among existing automatic artifact-removal methods, artifact subspace reconstruction (ASR) is an online and realtime capable, component-based method that can effectively remove transient or large-amplitude artifacts. However, the effectiveness of ASR and the optimal choice of its parameter have not been evaluated and reported, especially on real EEG data. This study systematically validates ASR on ten EEG recordings in a simulated driving experiment. Independent component analysis (ICA) is applied to separate artifacts from brain signals to allow a quantitative assessment of ASR's effectiveness in removing various types of artifacts and preserving brain activities. Empirical results show that the optimal ASR parameter is between 10 and 100, which is small enough to remove activities from artifacts and eye-related components and large enough to retain signals from brain-related components. With the appropriate choice of the parameter, ASR can be a powerful and automatic artifact removal approach for offline data analysis or online real-time EEG applications such as clinical monitoring and brain-computer interfaces.

Evaluating the Performance of Non-Hair SSVEP-Based BCIs Featuring Template-Based Decoding Methods

Our previous study has demonstrated the feasibility of employing non-hair-bearing electrodes to build a Steadystate Visual Evoked Potential (SSVEP)-based Brain-Computer Interface (BCI) system, relaxing technical barriers in preparation time and offering an ease-of-use apparatus. The signal quality of the SSVEPs and the resultant performance of the non-hair BCI, however, did not close upon those reported in the state-of-the-art BCI studies based on the electroencephalogram (EEG) measured from the occipital regions. Recently, advanced decoding algorithms such as task-related component analysis have made a breakthrough in enhancing the signal quality of the occipital SSVEPs and the performance of SSVEP-based BCIs in a well-controlled laboratory environment. However, it remains unclear if the advanced decoding algorithms can extract highfidelity SSVEPs from the non-hair EEG and enhance the practicality of non-hair BCIs in real-world environments. This study aims to quantitatively evaluate whether, and if so, to what extent the non-hair BCIs can leverage the state-of-art decoding algorithms. Eleven healthy individuals participated in a 5-target SSVEP BCI experiment. A high-density EEG cap recorded SSVEPs from both hair-covered and non-hair-bearing regions. By evaluating and demonstrating the accessibility of nonhair-bearing behind-ear signals, our assessment characterized constraints on data length, trial numbers, channels, and their relationships with the decoding algorithms, providing practical guidelines to optimize SSVEP-based BCI systems in real-life applications.

Online Automatic Artifact Rejection using the Real-time EEG Source-mapping Toolbox (REST)

Non-brain contributions to electroencephalographic (EEG) signals, often referred to as artifacts, can hamper the analysis of scalp EEG recordings. This is especially true when artifacts have large amplitudes (e.g., movement artifacts), or occur continuously (like eye-movement artifacts). Offline automated pipelines can detect and reduce artifact in EEG data, but no good solution exists for online processing of EEG data in near real time. Here, we propose the combined use of online artifact subspace reconstruction (ASR) to remove large amplitude transients, and online recursive independent component analysis (ORICA) combined with an independent component (IC) classifier to compute, classify, and remove artifact ICs. We demonstrate the efficacy of the proposed pipeline using 2 EEG recordings containing series of (1) movement and muscle artifacts, and (2) cued blinks and saccades. This pipeline is freely available in the Real-time EEG Sourcemapping Toolbox (REST) for MATLAB (The Mathworks, Inc.).

Transferring Shared Responses Across Electrode Montages for Facilitating Calibration in High-Speed Brain Spellers

Recent studies have shown that using the user's average steady-state visual evoked responses (SSVEPs) as the template to template-matching methods could significantly improve the accuracy and speed of the SSVEP-based brain- computer interface (BCI). However, collecting the pilot data for each individual can be time-consuming. To resolve this practical issue, this study aims to explore the feasibility of leveraging pre- recorded datasets from the same users by transferring common electroencephalogram (EEG) responses across different sessions with the same or different electrode montages. The proposed method employs spatial filtering techniques including response averaging, canonical correlation analysis (CCA), and task- related component analysis (TRCA) to project scalp EEG recordings onto a shared response domain. The transferability was evaluated by using 40-class SSVEPs recorded from eight subjects with nine electrodes on two different days. Three subsets of electrode montages were selected to simulate different scenarios such as identical, partly overlapped, and non-overlapped electrode placements across two sessions. The target identification accuracy of the proposed methods with transferred training data significantly outperformed a conventional training-free algorithm. The result suggests training data required in the BCI speller could be transferred from different EEG montages and/or headsets.

Evaluation of Artifact Subspace Reconstruction for Automatic Artifact Components Removal in Multi-Channel EEG Recordings

OBJECTIVE

Artifact subspace reconstruction (ASR) is an automatic, online-capable, component-based method that can effectively remove transient or large-amplitude artifacts contaminating electroencephalographic (EEG) data. However, the effectiveness of ASR and the optimal choice of its parameter have not been systematically evaluated and reported, especially on actual EEG data.

METHODS

This paper systematically evaluates ASR on 20 EEG recordings taken during simulated driving experiments. Independent component analysis (ICA) and an independent component classifier are applied to separate artifacts from brain signals to quantitatively assess the effectiveness of the ASR.

RESULTS

ASR removes more eye and muscle components than brain components. Even though some eye and muscle components retain after ASR cleaning, the power of their temporal activities is reduced. Study results also showed that ASR cleaning improved the quality of a subsequent ICA decomposition.

CONCLUSIONS

Empirical results show that the optimal ASR parameter is between 20 and 30, balancing between removing non-brain signals and retaining brain activities.

SIGNIFICANCE

With an appropriate choice of parameter, ASR can be a powerful and automatic artifact removal approach for offline data analysis or online real-time EEG applications such as clinical monitoring and brain-computer interfaces.

Facilitating Calibration in High-Speed BCI Spellers via Leveraging Cross-Device Shared Latent Responses

OBJECTIVE

This paper proposes a novel device-to-device transfer-learning algorithm for reducing the calibration cost in a steady-state visual evoked potential (SSVEP)-based brain-computer interface (BCI) speller by leveraging electroencephalographic (EEG) data previously acquired by different EEG systems.

METHODS

The transferring is done by projecting the scalp-channel EEG signals onto a shared latent domain across devices. Three spatial filtering techniques, including channel averaging, canonical correlation analysis (CCA), and task-related component analysis (TRCA), were employed to extract the shared responses from different devices. The transferred data were integrated into a template-matching-based algorithm to detect SSVEPs. To evaluate its transferability, this paper conducted two sessions of simulated online BCI experiments with ten subjects using 40 visual stimuli modulated by joint frequency-phase coding method. In each session, two different EEG devices were used: first, the Quick-30 system (Cognionics, Inc.) with dry electrodes, and second, the ActiveTwo system (BioSemi, Inc.) with wet electrodes.

RESULTS

The proposed method with CCA- and TRCA-based spatial filters achieved significantly higher classification accuracy compared with the calibration-free standard CCA-based method.

CONCLUSION

This paper validated the feasibility and effectiveness of the proposed method in implementing calibration-free SSVEP-based BCIs.

SIGNIFICANCE

The proposed method has great potentials to enhance practicability and usability of real-world SSVEP-based BCI applications by leveraging user-specific data recorded in previous sessions even with different EEG systems and montages.

Electroencephalographic and peripheral temperature dynamics during a prolonged psychomotor vigilance task

Attention lapses and fatigue are a main source of impaired performance that can lead to accidents. This study analyzed electroencephalographic (EEG) dynamics and body skin temperature as markers of attentional fluctuations in non-sleep deprived subjects during a 45min Psychomotor Vigilance Task (PVT). Independent Component Analysis and time-frequency analysis were used to evaluate the EEG data. Results showed a positive association between distal and distal-to-proximal gradient (DPG) temperatures and reaction time (RT); increments in EEG power in alpha-, theta- and beta-band frequencies in parieto-occipital, central-medial and frontal components, were associated with poor performance (slower RT) in the task. This generalized power increment fits with an increased activity in the default mode network, associated with attention lapses. This study highlights the potential use of the PVT as a tool to obtain individual physiological indices of vigilance and fatigue that could be applied to other vigilance tasks typically performed in occupational settings.

Eye fixation-related fronto-parietal neural network correlates of memory retrieval

Eye movements are considered to be informative with regard to the underlying cognitive processes of human beings. Previous studies have reported that eye movements are associated with which scientific concepts are retrieved correctly. Moreover, other studies have also suggested that eye movements involve the cooperative activity of the human brain's fronto-parietal circuits. Less research has been conducted to investigate whether fronto-parietal EEG oscillations are associated with the retrieval processing of scientific concepts. Our findings in this study demonstrated that the fronto-parietal network is indeed crucial for successful memory retrieval. In short, significantly lower theta augmentation in the frontal midline and lower alpha suppression in the right parietal region were observed at the 5th eye fixation for physics concepts that were correctly retrieved than for those that were incorrectly retrieved. Moreover, the visual cortex in the occipital lobe exhibits a significantly greater theta augmentation followed by an alpha suppression following each eye fixation, while a right fronto-parietal asymmetry was also found for the successful retrieval of presentations of physics concepts. In particular, the study results showed that eye fixation-related frontal midline theta power and right parietal alpha power at the 5th eye fixation have the greatest predictive power regarding the correctness of the retrieval of physics concepts.

Multi-modal Approach for Affective Computing

Throughout the past decade, many studies have classified human emotions using only a single sensing modality such as face video, electroencephalogram (EEG), electrocardiogram (ECG), galvanic skin response (GSR), etc. The results of these studies are constrained by the limitations of these modalities such as the absence of physiological biomarkers in the face-video analysis, poor spatial resolution in EEG, poor temporal resolution of the GSR etc. Scant research has been conducted to compare the merits of these modalities and understand how to best use them individually and jointly. Using multi-modal AMIGOS dataset, this study compares the performance of human emotion classification using multiple computational approaches applied to face videos and various bio-sensing modalities. Using a novel method for compensating physiological baseline we show an increase in the classification accuracy of various approaches that we use. Finally, we present a multi-modal emotion-classification approach in the domain of affective computing research.

Brain Network Changes in Fatigued Drivers: A Longitudinal Study in a Real-World Environment Based on the Effective Connectivity Analysis and Actigraphy Data

The analysis of neurophysiological changes during driving can clarify the mechanisms of fatigue, considered an important cause of vehicle accidents. The fluctuations in alertness can be investigated as changes in the brain network connections, reflected in the direction and magnitude of the information transferred. Those changes are induced not only by the time on task but also by the quality of sleep. In an unprecedented 5-month longitudinal study, daily sampling actigraphy and EEG data were collected during a sustained-attention driving task within a near-real-world environment. Using a performance index associated with the subjects' reaction times and a predictive score related to the sleep quality, we identify fatigue levels in drivers and investigate the shifts in their effective connectivity in different frequency bands, through the analysis of the dynamical coupling between brain areas. Study results support the hypothesis that combining EEG, behavioral and actigraphy data can reveal new features of the decline in alertness. In addition, the use of directed measures such as the Convergent Cross Mapping can contribute to the development of fatigue countermeasure devices.

A Wearable Multi-Modal Bio-Sensing System Towards Real-World Applications

Multi-modal bio-sensing has recently been used as effective research tools in affective computing, autism, clinical disorders, and virtual reality among other areas. However, none of the existing bio-sensing systems support multi-modality in a wearable manner outside well-controlled laboratory environments with research-grade measurements. This paper attempts to bridge this gap by developing a wearable multi-modal bio-sensing system capable of collecting, synchronizing, recording, and transmitting data from multiple bio-sensors: PPG, EEG, eye-gaze headset, body motion capture, GSR, etc., while also providing task modulation features including visual-stimulus tagging. This study describes the development and integration of various components of our system. We evaluate the developed sensors by comparing their measurements to those obtained by a standard research-grade bio-sensors. We first evaluate different sensor modalities of our headset, namely, earlobe-based PPG module with motion-noise canceling for ECG during heart-beat calculation. We also compare the steady-state visually evoked potentials measured by our shielded dry EEG sensors with the potentials obtained by commercially available dry EEG sensors. We also investigate the effect of head movements on the accuracy and precision of our wearable eye-gaze system. Furthermore, we carry out two practical tasks to demonstrate the applications of using multiple sensor modalities for exploring previously unanswerable questions in bio-sensing. Specifically, utilizing bio-sensing, we show which strategy works best for playing "Where is Waldo?" visual-search game, changes in EEG corresponding to true vs. false target fixations in this game, and predicting the loss/draw/win states through bio-sensing modalities while learning their limitations in a "Rock-Paper-Scissors" game.

Theta and Alpha Oscillations in Attentional Interaction during Distracted Driving

Performing multiple tasks simultaneously usually affects the behavioral performance as compared with executing the single task. Moreover, processing multiple tasks simultaneously often involve more cognitive demands. Two visual tasks, lane-keeping task and mental calculation, were utilized to assess the brain dynamics through 32-channel electroencephalogram (EEG) recorded from 14 participants. A 400-ms stimulus onset asynchrony (SOA) factor was used to induce distinct levels of attentional requirements. In the dual-task conditions, the deteriorated behavior reflected the divided attention and the overlapping brain resources used. The frontal, parietal and occipital components were decomposed by independent component analysis (ICA) algorithm. The event- and response-related theta and alpha oscillations in selected brain regions were investigated first. The increased theta oscillation in frontal component and decreased alpha oscillations in parietal and occipital components reflect the cognitive demands and attentional requirements as executing the designed tasks. Furthermore, time-varying interactive over-additive (O-Add), additive (Add) and under-additive (U-Add) activations were explored and summarized through the comparison between the summation of the elicited spectral perturbations in two single-task conditions and the spectral perturbations in the dual task. Add and U-Add activations were observed while executing the dual tasks. U-Add theta and alpha activations dominated the posterior region in dual-task situations. Our results show that both deteriorated behaviors and interactive brain activations should be comprehensively considered for evaluating workload or attentional interaction precisely.

Neural Oscillation Correlates Chemistry Decision-Making

This study explored the electroencephalography (EEG) dynamics during a chemistry-related decision-making task and further examined whether the correctness of the decision-making performance could be reflected by EEG activity. A total of 66 undergraduate students’ EEG were collected while they participated in a chemistry-related decision-making task in which they had to retrieve the relevant chemistry concepts in order to make correct decisions for each task item. The results showed that it was only in the anterior cingulate cortex (ACC) cluster that distinct patterns in EEG dynamics were displayed for the correct and incorrect responses. The logistic regression results indicated that ACC theta power from 300[Formula: see text]ms to 250[Formula: see text]ms before stimulus onset was the most informative factor for estimating the likelihood of making correct decisions in the chemistry-related decision-making task, while it was the ACC low beta power from 150[Formula: see text]ms to 250[Formula: see text]ms after stimulus onset. The results suggested that the ACC theta augmentation before the stimulus onset serves to actively maintain the relevant information for retrieval from long-term memory, while the ACC low beta augmentation after the stimulus onset may serve the function of mapping the encoded stimulus onto the relevant criteria that the given participant has held within his or her mind to guide the decision-making responses.

Enhancing Detection of SSVEPs for a High-Speed Brain Speller Using Task-Related Component Analysis

OBJECTIVE

This study proposes and evaluates a novel data-driven spatial filtering approach for enhancing steady-state visual evoked potentials (SSVEPs) detection toward a high-speed brain-computer interface (BCI) speller.

METHODS

Task-related component analysis (TRCA), which can enhance reproducibility of SSVEPs across multiple trials, was employed to improve the signal-to-noise ratio (SNR) of SSVEP signals by removing background electroencephalographic (EEG) activities. An ensemble method was further developed to integrate TRCA filters corresponding to multiple stimulation frequencies. This study conducted a comparison of BCI performance between the proposed TRCA-based method and an extended canonical correlation analysis (CCA)-based method using a 40-class SSVEP dataset recorded from 12 subjects. An online BCI speller was further implemented using a cue-guided target selection task with 20 subjects and a free-spelling task with 10 of the subjects.

RESULTS

The offline comparison results indicate that the proposed TRCA-based approach can significantly improve the classification accuracy compared with the extended CCA-based method. Furthermore, the online BCI speller achieved averaged information transfer rates (ITRs) of 325.33 ± 38.17 bits/min with the cue-guided task and 198.67 ± 50.48 bits/min with the free-spelling task.

CONCLUSION

This study validated the efficiency of the proposed TRCA-based method in implementing a high-speed SSVEP-based BCI.

SIGNIFICANCE

The high-speed SSVEP-based BCIs using the TRCA method have great potential for various applications in communication and control.

Sustained Attention in Real Classroom Settings: An EEG Study

Sustained attention is a process that enables the maintenance of response persistence and continuous effort over extended periods of time. Performing attention-related tasks in real life involves the need to ignore a variety of distractions and inhibit attention shifts to irrelevant activities. This study investigates electroencephalography (EEG) spectral changes during a sustained attention task within a real classroom environment. Eighteen healthy students were instructed to recognize as fast as possible special visual targets that were displayed during regular university lectures. Sorting their EEG spectra with respect to response times, which indicated the level of visual alertness to randomly introduced visual stimuli, revealed significant changes in the brain oscillation patterns. The results of power-frequency analysis demonstrated a relationship between variations in the EEG spectral dynamics and impaired performance in the sustained attention task. Across subjects and sessions, prolongation of the response time was preceded by an increase in the delta and theta EEG powers over the occipital region, and decrease in the beta power over the occipital and temporal regions. Meanwhile, implementation of the complex attention task paradigm into a real-world classroom setting makes it possible to investigate specific mutual links between brain activities and factors that cause impaired behavioral performance, such as development and manifestation of classroom mental fatigue. The findings of the study set a basis for developing a system capable of estimating the level of visual attention during real classroom activities by monitoring changes in the EEG spectra.

Detecting Glaucoma With a Portable Brain-Computer Interface for Objective Assessment of Visual Function Loss

Importance

The current assessment of visual field loss in diseases such as glaucoma is affected by the subjectivity of patient responses and the lack of portability of standard perimeters.

Objective

To describe the development and initial validation of a portable brain-computer interface (BCI) for objectively assessing visual function loss.

Design, Setting, and Participants

This case-control study involved 62 eyes of 33 patients with glaucoma and 30 eyes of 17 healthy participants. Glaucoma was diagnosed based on a masked grading of optic disc stereophotographs. All participants underwent testing with a BCI device and standard automated perimetry (SAP) within 3 months. The BCI device integrates wearable, wireless, dry electroencephalogram and electrooculogram systems and a cellphone-based head-mounted display to enable the detection of multifocal steady state visual-evoked potentials associated with visual field stimulation. The performances of global and sectoral multifocal steady state visual-evoked potentials metrics to discriminate glaucomatous from healthy eyes were compared with global and sectoral SAP parameters. The repeatability of the BCI device measurements was assessed by collecting results of repeated testing in 20 eyes of 10 participants with glaucoma for 3 sessions of measurements separated by weekly intervals.

Main Outcomes and Measures

Receiver operating characteristic curves summarizing diagnostic accuracy. Intraclass correlation coefficients and coefficients of variation for assessing repeatability.

Results

Among the 33 participants with glaucoma, 19 (58%) were white, 12 (36%) were black, and 2 (6%) were Asian, while among the 17 participants with healthy eyes, 9 (53%) were white, 8 (47%) were black, and none were Asian. The receiver operating characteristic curve area for the global BCI multifocal steady state visual-evoked potentials parameter was 0.92 (95% CI, 0.86-0.96), which was larger than for SAP mean deviation (area under the curve, 0.81; 95% CI, 0.72-0.90), SAP mean sensitivity (area under the curve, 0.80; 95% CI, 0.69-0.88; P = .03), and SAP pattern standard deviation (area under the curve, 0.77; 95% CI, 0.66-0.87; P = .01). No statistically significant differences were seen for the sectoral measurements between the BCI and SAP. Intraclass coefficients for global and sectoral parameters ranged from 0.74 to 0.92, and mean coefficients of variation ranged from 3.03% to 7.45%.

Conclusions and Relevance

The BCI device may be useful for assessing the electrical brain responses associated with visual field stimulation. The device discriminated eyes with glaucomatous neuropathy from healthy eyes in a clinically based setting. Further studies should investigate the feasibility of the BCI device for home-based testing as well as for detecting visual function loss over time.

Improving EEG-Based Emotion Classification Using Conditional Transfer Learning

To overcome the individual differences, an accurate electroencephalogram (EEG)-based emotion-classification system requires a considerable amount of ecological calibration data for each individual, which is labor-intensive and time-consuming. Transfer learning (TL) has drawn increasing attention in the field of EEG signal mining in recent years. The TL leverages existing data collected from other people to build a model for a new individual with little calibration data. However, brute-force transfer to an individual (i.e., blindly leveraged the labeled data from others) may lead to a negative transfer that degrades performance rather than improving it. This study thus proposed a conditional TL (cTL) framework to facilitate a positive transfer (improving subject-specific performance without increasing the labeled data) for each individual. The cTL first assesses an individual’s transferability for positive transfer and then selectively leverages the data from others with comparable feature spaces. The empirical results showed that among 26 individuals, the proposed cTL framework identified 16 and 14 transferable individuals who could benefit from the data from others for emotion valence and arousal classification, respectively. These transferable individuals could then leverage the data from 18 and 12 individuals who had similar EEG signatures to attain maximal TL improvements in valence- and arousal-classification accuracy. The cTL improved the overall classification performance of 26 individuals by ~15% for valence categorization and ~12% for arousal counterpart, as compared to their default performance based solely on the subject-specific data. This study evidently demonstrated the feasibility of the proposed cTL framework for improving an individual’s default emotion-classification performance given a data repository. The cTL framework may shed light on the development of a robust emotion-classification model using fewer labeled subject-specific data toward a real-life affective brain-computer interface (ABCI).

Monitoring alert and drowsy states by modeling EEG source nonstationarity

OBJECTIVE

As a human brain performs various cognitive functions within ever-changing environments, states of the brain characterized by recorded brain activities such as electroencephalogram (EEG) are inevitably nonstationary. The challenges of analyzing the nonstationary EEG signals include finding neurocognitive sources that underlie different brain states and using EEG data to quantitatively assess the state changes.

APPROACH

This study hypothesizes that brain activities under different states, e.g. levels of alertness, can be modeled as distinct compositions of statistically independent sources using independent component analysis (ICA). This study presents a framework to quantitatively assess the EEG source nonstationarity and estimate levels of alertness. The framework was tested against EEG data collected from 10 subjects performing a sustained-attention task in a driving simulator.

MAIN RESULTS

Empirical results illustrate that EEG signals under alert versus drowsy states, indexed by reaction speeds to driving challenges, can be characterized by distinct ICA models. By quantifying the goodness-of-fit of each ICA model to the EEG data using the model deviation index (MDI), we found that MDIs were significantly correlated with the reaction speeds (r  =  -0.390 with alertness models and r  =  0.449 with drowsiness models) and the opposite correlations indicated that the two models accounted for sources in the alert and drowsy states, respectively. Based on the observed source nonstationarity, this study also proposes an online framework using a subject-specific ICA model trained with an initial (alert) state to track the level of alertness. For classification of alert against drowsy states, the proposed online framework achieved an averaged area-under-curve of 0.745 and compared favorably with a classic power-based approach.

SIGNIFICANCE

This ICA-based framework provides a new way to study changes of brain states and can be applied to monitoring cognitive or mental states of human operators in attention-critical settings or in passive brain-computer interfaces.

Polychromatic SSVEP stimuli with subtle flickering adapted to brain-display interactions

OBJECTIVE

Interactive displays armed with natural user interfaces (NUIs) will likely lead the next breakthrough in consumer electronics, and brain-computer interfaces (BCIs) are often regarded as the ultimate NUI-enabling machines to respond to human emotions and mental states. Steady-state visual evoked potentials (SSVEPs) are a commonly used BCI modality due to the ease of detection and high information transfer rates. However, the presence of flickering stimuli may cause user discomfort and can even induce migraines and seizures. With the aim of designing visual stimuli that can be embedded into video images, this study developed a novel approach to induce detectable SSVEPs using a composition of red/green/blue flickering lights.

APPROACH

Based on the opponent theory of colour vision, this study used 32 Hz/40 Hz rectangular red-green or red-blue LED light pulses with a 50% duty cycle, balanced/equal luminance and 0°/180° phase shifts as the stimulating light sources and tested their efficacy in producing SSVEP responses with high signal-to-noise ratios (SNRs) while reducing the perceived flickering sensation.

MAIN RESULTS

The empirical results from ten healthy subjects showed that dual-colour lights flickering at 32 Hz/40 Hz with a 50% duty cycle and 180° phase shift achieved a greater than 90% detection accuracy with little or no flickering sensation.

SIGNIFICANCE

As a first step in developing an embedded SSVEP stimulus in commercial displays, this study provides a foundation for developing a combination of three primary colour flickering backlights with adjustable luminance proportions to create a subtle flickering polychromatic light that can elicit SSVEPs at the basic flickering frequency.

Eye Activity Correlates of Fatigue during a Visual Tracking Task

Five concurrent eye activity measures were used to model fatigue-related changes in performance during a visual compensatory tracking task. Five subjects demonstrated considerable variations in performance level within two 53-min testing sessions during which continuous video-based eye activity measures were obtained. For each subject, moving estimates of blink duration and frequency, fixation duration and frequency, and mean pupil diameter from one session were used to train an artificial neural network to produce moving estimates of changes in mean tracking performance during the same session. Applied to eye tracking data from a second session, the same networks produced moving estimates of tracking performance that were highly correlated with actual performance changes ( R2=0.65, range 0.30–0.89 across ten sessions). The results suggest that information from multiple eye measures may be combined to produce individualized and accurate estimates of sub-minute scale changes in alertness during continuous task performance.