This review highlights the strengths and weaknesses of the three categories of SSVEP training methods. Training-free systems are more practical but their performance is limited due to inter-subject variability resulting from the complex EEG activity. Feature extraction methods that incorporate some training data address this issue and in fact have outperformed training-free methods: subject-specific BCIs are tuned to the individual yielding the best performance at the cost of long, tiring training sessions making these methods unsuitable for everyday use; subject-independent BCIs that make use of training data from various subjects offer a good trade-off between training effort and performance, making these BCIs better suited for practical use.

The use of brain signals for person recognition has in recent years attracted considerable interest because of the increased security and privacy these can offer when compared to conventional biometric measures. The main challenge lies in extracting features from the EEG signals that are sufficiently distinct across individuals while also being sufficiently consistent across multiple recording sessions. A range of EEG phenomena including eyes open and eyes closed activity, visual evoked potentials (VEPs) through image presentation, and other mental tasks have been studied for their use in biometry.

No abstract

No abstract

Objective. This study analyses the effect of distractors on steady-state visually evoked potentials (SSVEP) to establish how these nuisance signals, typically present in the real-world, would effect an SSVEP-based brain-computer interface (BCI). The distractors introduced to the SSVEP-based experiments are auditory, visual and movement distractors, specifically selected to reflect the use of various BCIs outside the laboratory. Approach. An assessment on the influence of these nuisance signals on SSVEP data as compared to SSVEP data with no distractors is presented. This is done by examining the frequency spectrum of the SSVEP responses followed by the implementation of feature extraction techniques, specifically canonical correlation analysis (CCA) and power spectral density analysis (PSDA), and a statistical analysis on the results obtained. As an added contribution, this study is designed to have each subject repeat the same experiments on different days such that the variation in performance with and without distractors may be investigated longitudinally. Main results. The results revealed that the spectra and the classification performance of the auditory distractors condition are comparable with the no distractors condition. On the other hand, there is a decrease in the signal-to-noise ratios (SNRs) of the visual and movement distractors conditions. This is reflected with a significant decrease in the SSVEP performance of 4.29% and 12.14% for the two distractor conditions respectively compared to the no distractors condition. A significant above-chance level SSVEP performance could still be obtained with all the distractor conditions. The longitudinal analysis revealed that the difference in performance between the SSVEP experimental conditions was consistent across sessions. Significance. This study provides a comparison on the performance of an SSVEP-based BCI with various external distractors. This analysis is necessary for the transition of BCI applications to be used every day in uncontrolled environments.