The Role of Transient Target Stimuli in a Steady-State Somatosensory Evoked Potential-Based Brain–Computer Interface Setup

Abstract

In earlier literature, so-called twitches were used to support a user in a steady-state somatosensory evoked potential (SSSEP) based brain–computer interface (BCI) to focus attention on the requested targets. Within this work, we investigate the impact of these transient target stimuli on SSSEPs in a real-life BCI setup. A hybrid BCI was designed which combines SSSEPs and P300 potentials evoked by twitches randomly embedded into the streams of tactile stimuli. The EEG of fourteen healthy subjects was recorded, while their left and right index fingers were simultaneously stimulated using frequencies selected in a screening procedure. The subjects were randomly instructed by a cue on a screen to focus attention on one or none of the fingers. Feature for SSSEPs and P300 potentials were extracted and classified using separately trained multi-class shrinkage LDA classifiers. Three-class classification accuracies significantly better than random could be reached by nine subjects using SSSEP features and by 12 subjects using P300 features respectively. The average classification accuracies were 48.6% using SSSEP and 50.7% using P300 features. By means of a Monte Carlo permutation test it could be shown that twitches have an attenuation effect on the SSSEP. Significant SSSEP blocking effects time-locked to twitch positions were found in seven subjects. Our findings suggest that the attempt to combine different types of stimulation signals like repetitive signals and twitches has a mutual influence on each other, which may be the main reason for the rather moderate BCI performance. This influence is originated at the level of stimulus generation but becomes apparent as physiological effect in the SSSEP. When designing a hybrid BCI based on SSSEPs and P300 potentials, one has to find an optimal tradeoff depending on the overall design goals or individual subjects' performance. Our results give therefore some new insights that may be useful for the successful design of hybrid BCIs.

Classification of Selective Attention Within Steady-State Somatosensory Evoked Potentials From Dry Electrodes Using Mutual Information-Based Spatio-Spectral Feature Selection

“…In our experimental results (Fig. 6), the proposed method showed higher performances (56.4%, and 57.9%) compared to the previous study as 48.6% [29]. Despite considering the differences in stimuli inducing SSSEP and in the number of subjects from the previous study and the current study, our approach is valuable.…”

“…Our study showed that the proposed method can help to improve the classification accuracy of the selective attention and inattention and show spatially clear and distinguishable patterns for vibration stimuli-induced SSSEP-based BCIs. In the previous twitch-based multiclass SSSEP study, Pokorny et al [29], the cross-validation accuracy for multi-class (concentrating on left-hand, right-hand, and idle class) were averaged 48.6% within 14 subjects using C-2 tactor (Engineering Acoustics, Inc., Casselberry, Florida, USA) with finger clip to induce SSSEP [29]. Unlike our experimental setup, the EEG signals from 29 channels focused on the somatosensory cortex and the EOG signals from 3 channels were recorded [29].…”

“…In the previous twitch-based multiclass SSSEP study, Pokorny et al [29], the cross-validation accuracy for multi-class (concentrating on left-hand, right-hand, and idle class) were averaged 48.6% within 14 subjects using C-2 tactor (Engineering Acoustics, Inc., Casselberry, Florida, USA) with finger clip to induce SSSEP [29]. Unlike our experimental setup, the EEG signals from 29 channels focused on the somatosensory cortex and the EOG signals from 3 channels were recorded [29]. In order to have constant contact pressure between the tactors and fingers, finger clips were used [29].…”

“…Unlike our experimental setup, the EEG signals from 29 channels focused on the somatosensory cortex and the EOG signals from 3 channels were recorded [29]. In order to have constant contact pressure between the tactors and fingers, finger clips were used [29]. The tactile stimulation pattern consisted of a twitch-based sinusoidal carrier signal which was amplitude modulated with a rectangular stimulation signal of the respective stimulation frequency.…”

“…The tactile stimulation pattern consisted of a twitch-based sinusoidal carrier signal which was amplitude modulated with a rectangular stimulation signal of the respective stimulation frequency. For the data analysis in [29], the features were extracted with the logarithmic lock-in amplifier system (LAS) and the multi-class shrinkage LDA was also used as the classifier. The LAS is an algorithm of filtering small sinusoidal signals out of random noise, or out of a mixture of different signals [16].…”