Single Finger Trajectory Prediction From Intracranial Brain Activity Using Block-Term Tensor Regression With Fast and Automatic Component Extraction.

Multiway-or tensor-based decoding techniques for brain-computer interfaces (BCIs) are believed to better account for the multilinear structure of brain signals than conventional vector-or matrix-based ones. However, despite their outlook on significant performance gains, the used parameter optimization approach is often too computationally demanding so that conventional techniques are still preferred. We propose two novel tensor factorizations which we integrate into our block-term tensor regression (BTTR) algorithm and further introduce a marginalization procedure that guarantees robust predictions while reducing the risk of overfitting (generalized regression). BTTR accounts for the underlying (hidden) data structure in a fully automatic and computationally efficient manner, leading to a significant performance gain over conventional vector-or matrix-based techniques in a challenging real-world application. As a challenging real-world application, we apply BTTR to accurately predict single finger movement trajectories from intracranial recordings in human subjects. We compare the obtained performance with that of the state-of-the-art.

Phase-Spatial Beamforming Renders a Visual Brain Computer Interface Capable of Exploiting EEG Electrode Phase Shifts in Motion-Onset Target Responses.

OBJECTIVE: in this work, we aim to develop a more efficient visual motion-onset based Brain-computer interface (BCI). Brain-computer interfaces provide communication facilities that do not rely on the brain's usual pathways. Visual BCIs are based on changes in EEG activity in response to attended flashing or flickering targets. A less taxing way to encode such targets is with briefly moving stimuli, the onset of which elicits a lateralized EEG potential over the parieto-occipital scalp area called the motion-onset visual evoked potential (mVEP). METHODS: We recruited 21 healthy subjects for an experiment in which motion-onset stimulations translating leftwards (LT) or rightwards (RT) were encoding 9 displayed targets. We propose a novel algorithm that exploits the phase-shift between EEG electrodes to improve target decoding performance. We hereto extend the spatiotemporal beamformer (stBF) with a phase extracting procedure, leading to the phase-spatial beamformer (psBF). RESULTS: we show that psBF performs significantly better than the stBF (p < 0.001 for 1 and 2 stimulus repetitions and p < 0.01 for 3 to 5 stimulus repetitions), as well as the previously validated linear support-vector machines (p < 0.001 for 5 stimulus repetitions and p < 0.01 for 1,2 and 6 stimulus repetitions) and stepwise linear discriminant analysis decoders (p < 0.001 for all repetitions) when simultaneously addressing timing and translation direction. CONCLUSION: We provide evidence of decodability of joint direction and target in mVEP responses. SIGNIFICANCE: the described methods can aid in the development of a faster and more comfortable BCI based on mVEPs.

Accurate Offline Asynchronous Detection of Individual Finger Movement From Intracranial Brain Signals Using a Novel Multiway Approach.

Asynchronous motor Brain Computer Interfacing (BCI) is characterized by the continuous decoding of intended muscular activity from brain signals. Such applications have gained widespread interest for enabling users to issue commands volitionally. In conventional motor BCIs features extracted from brain signals are concatenated into vector- or matrix-based (or one-/two-way) representations. Nevertheless, when accounting for the original multimodal or multiway signal structure, decoding performance has been shown to improve jointly with result interpretability. However, as multiway decoders are notorious for the extensive computational cost to train them, conventional ones are still preferred. To curb this limitation, we introduce a novel multiway classifier, called Block-Term Tensor Classifier that inherits the improved accuracy of multiway methods while providing fast training. We show that it can outperform state-of-the-art multiway and two-way Linear Discriminant Analysis classifiers in asynchronous detection of individual finger movements from intracranial recordings, an essential feature to achieve a sense of dexterity with hand prosthetics and exoskeletons.

Measuring brand association strength with EEG: A single-trial N400 ERP study.

Companies need to ensure that customers perceive their brands as intended, with strong and unique associations, when facing a competitive market. Traditionally, brand associations are measured using conventional techniques such as surveys and questionnaires albeit both conscious and unconscious factors can influence the collected data and the outcome of a campaign. Neuromarketing can shed light on how the customer's brain processes marketing stimuli. We report here on an EEG study aimed at gauging mental associations with brands. We focus on the N400 event-related potential, an EEG component most strongly elicited in response to a concept unrelated to a preceding concept. We considered two video on demand brands, Netflix and Rex&Rio, and selected a set of words grouped in 4 categories that were either related (Television, Relaxation, and Price), in varying degrees, or unrelated (Unrelated) to the said brands. The experiment started with both brands' TV commercials, as a common reference for our participants. We then applied a semantic priming paradigm in which a brand logo ("prime") was followed by a word ("target"), and the strength of the N400 response to the word used as an inverted measure of the association strength with the brand logo. We clustered N400 responses to identify, for each brand, natural groups of associated words. As a result, for Netflix the cluster with the smallest N400 responses (i.e., strongest associations) consisted of words related to Television but for Rex&Rio it consisted of words related to Relaxation. We also evaluated the relationship between the two brands and determined which associations they share or which ones not. It turned out that associations related to Relaxation and Television distinguish the two brands. Interestingly, survey data did not show any difference between the two brands as they were equally associated with Television and Relaxation. These findings show that our N400 technique can reveal brand associations, and natural categories thereof, that would otherwise go unnoticed when using conventional surveys.

Fast Multiway Partial Least Squares Regression.

OBJECTIVE: Multiway array decomposition has been successful in providing a better understanding of the structure underlying data and in discovering potentially hidden feature dependences serving high-performance decoder applications. However, the computational cost of multiway algorithms can become prohibitive, especially when considering large datasets, rendering them unsuitable for time-critical applications. METHODS: We propose a multiway regression model for large-scale tensors with optimized performance in terms of time complexity, called fast higher order partial least squares (fHOPLS). RESULTS: We compare fHOPLS with its native version, higher order partial least squares (HOPLS), the state-of-the-art in multilinear regression, under different noise conditions and tensor dimensionalities using synthetic data. We also compare their performance when used for predicting scalp-recorded electroencephalography signals from invasively recorded electrocorticography signals in an oddball experiment. For the sake of exposition, we evaluated the performance of standard unfolded partial least squares (PLS) and linear regression. CONCLUSION: Our results show that fHOPLS is significantly faster than HOPLS, in particular for big data. In addition, the regression performances of fHOPLS and HOPLS are comparable and outperform both unfolded PLS and linear regression. Another interesting result is that multiway array decoding yields more accurate results than epoch-based averaging procedures traditionally used in the brain computer interfacing community.

Decoding Steady-State Visual Evoked Potentials From Electrocorticography.

We report on a unique electrocorticography (ECoG) experiment in which Steady-State Visual Evoked Potentials (SSVEPs) to frequency- and phase-tagged stimuli were recorded from a large subdural grid covering the entire right occipital cortex of a human subject. The paradigm is popular in EEG-based Brain Computer Interfacing where selectable targets are encoded by different frequency- and/or phase-tagged stimuli. We compare the performance of two state-of-the-art SSVEP decoders on both ECoG- and scalp-recorded EEG signals, and show that ECoG-based decoding is more accurate for very short stimulation lengths (i.e., less than 1 s). Furthermore, whereas the accuracy of scalp-EEG decoding benefits from a multi-electrode approach, to address interfering EEG responses and noise, ECoG decoding enjoys only a marginal improvement as even a single electrode, placed over the posterior part of the primary visual cortex, seems to suffice. This study shows, for the first time, that EEG-based SSVEP decoders can in principle be applied to ECoG, and can be expected to yield faster decoding speeds using less electrodes.

ERP Response Unveils Effect of Second Language Manipulation on First Language Processing.

Lexical access in bilinguals has been considered either selective or non-selective and evidence exists in favor of both hypotheses. We conducted a linguistic experiment to assess whether a bilingual's language mode influences the processing of first language information. We recorded event related potentials during a semantic priming paradigm with a covert manipulation of the second language (L2) using two types of stimulus presentations (short and long). We observed a significant facilitation of word pairs related in L2 in the short version reflected by a decrease in N400 amplitude in response to target words related to the English meaning of an inter-lingual homograph (homograph-unrelated group). This was absent in the long version, as the N400 amplitude for this group was similar to the one for the control-unrelated group. We also interviewed the participants whether they were aware of the importance of L2 in the experiment. We conclude that subjects participating in the long and short versions were in different language modes: closer to monolingual mode for the long and closer to bilingual mode for the short version; and that awareness about covert manipulation of L2 can influence the language mode, which in its turn influences the processing of the first language.

Stable Voice Clusters Identified When Using the Maximum versus Minimum Intensity Curve in the Phonetogram.

OBJECTIVE: To examine whether the minimum as well as the maximum voice intensity (i.e. sound pressure level, SPL) curves of a voice range profile (VRP) are required when discovering different voice groups based on a clustering analysis. In this approach, no a priori labeling of voice types is used. PATIENTS AND METHODS: VRPs of 194 (84 male and 110 female) professional singers were registered and processed. Cluster analysis was performed with the use of features related to (1) both the maximum and minimum SPL curves and (2) the maximum SPL curve only. RESULTS: Features related to the maximum as well as the minimum SPL curves showed three clusters in both male and female voices. These clusters, or voice groups, are based on voice types with similar VRP features. However, when using features related only to the maximum SPL curve, the clusters became less obvious. CONCLUSION: Features related to the maximum and minimum SPL curves of a VRP are both needed in order to identify the three voice clusters.