Event-related potential

Related publications (79)

Error-Related EEG Potentials Generated during Simulated Brain-Computer Interaction

Pierre Ferrez

Brain-computer interfaces (BCIs) are prone to errors in the recognition of subject's intent. An elegant approach to improve the accuracy of BCIs consists in a verification procedure directly based on the presence of error-related potentials (ErrP) in the EEG recorded right after the occurrence of an error. Several studies show the presence of ErrP in typical choice reaction tasks. However, in the context of a BCI, the central question is: "Are ErrP also elicited when the error is made by the interface during the recognition of the subject's intent?" We have thus explored whether ErrP also follow a feedback indicating incorrect responses of the simulated BCI interface. Five healthy volunteer subjects participated in a new human-robot interaction experiment, which seem to confirm the previously reported presence of a new kind of ErrP. But in order to exploit these ErrP we need to detect them in each single trial using a short window following the feedback associated to the response of the BCI. We have achieved an average recognition rate of correct and erroneous single trials of 83.5% and 79.2%, respectively using a classifier built with data recorded up to three months earlier.

2008

Error-Related EEG Potentials in Brain-Computer Interfaces

Pierre Ferrez

Brain-computer interfaces (BCI), as any other interaction modality based on physiological signals and body channels (e.g., muscular activity, speech and gestures), are prone to errors in the recognition of subject's intent. An elegant approach to improve the accuracy of BCIs consists in a verification procedure directly based on the presence of error-related potentials (ErrP) in the EEG recorded right after the occurrence of an error. Most of these studies show the presence of ErrP in typical choice reaction tasks where subjects respond to a stimulus and ErrP arise following errors due to the subject's incorrect motor action. However, in the context of a BCI, the central question is: "Are ErrP also elicited when the error is made by the interface during the recognition of the subject's intent?" We have thus explored whether ErrP also follow a feedback indicating incorrect responses of the interface and no longer errors of the subject himself. Four healthy volunteer subjects participated in a simple human-robot interaction experiment (i.e., bringing the robot to either the left or right side of a room), which seem to reveal a new kind of ErrP. These "interaction ErrP" exhibit a first sharp negative peak followed by a broader positive peak and a second negative peak (~270, 400 and 550 ms after the feedback, respectively). But in order to exploit these ErrP we need to detect it in each single trial using a short window following the feedback that shows the response of the classifier embedded in the BCI. We have achieved an average recognition rate of correct and erroneous single trials of 83.7% and 80.2%, respectively. We also show that the integration of these ErrP in a BCI, where the subject's intent is not executed if an ErrP is detected, significantly improves the performance of the BCI.

The MIT Press2007

Towards everyday brain-computer interfacing through decoding visual cognition

Ruslan Aydarkhanov

Decoding visual cognition from non-invasive measurements of brain activity has shown valuable applications. Vision-based Brain-Computer Interfaces (BCI) systems extend from spellers to database search and spatial navigation. Despite the high performance of these systems, they are limited to the laboratory environment. Several attempts were made to improve the quality of life of patients with severe neurodegenerative disorders by helping them acquire more independence. Bringing vision-based BCI to everyday life would benefit a wide range of population from handicapped to healthy people. However, real-life applications impose additional challenges on obtaining neural signals with a good signal-to-noise ratio. The natural environments are rich, dynamic and ambiguous that challenges the visual perception and, therefore, the decoding of underlying neural correlates. The limited attentional resources require to explore and visually sample the environment by freely moving eyes and fixating on the relevant objects. Moreover, everyday activities are rarely isolated leading to a mixture and complex interactions between neural correlates in the measured signal. In this thesis, I explore various aspects of decoding brain activity by developing new protocols for realistic scenarios with challenging visual tasks. The tasks include free visual exploration, dynamic scenes, and multiple tasks, for example, recognition of a billboard category during simulated driving. The major efforts have been directed towards synchronous decoding of Event-Related Potentials (ERP) acquired by the means of electroencephalography (EEG) and Eye-tracking technology, which allowed the extraction of Eye-Fixation Related Potentials (EFRP). Additionally, I investigate the neural correlates of perceptual decision making which is tightly linked to visual recognition. The perceptual challenges can lead to higher temporal variability of obtained ERP which was also addressed from the decoding perspective. The obtained results, firstly, provide new insights on how to tackle the temporal variability of ERP. Secondly, they raise new questions on studying and decoding neural correlates of perceptual decision making. And thirdly, they show the feasibility of decoding visual recognition in a simulated realistic scenario of car driving. Although, as expected, I found a lower performance compared to classical ERP protocols, these findings hold promise and raise new questions to be investigated in order to improve the quality of decoding visual cognition for everyday application.

EPFL2020

The ability to notice erroneous behavior is crucial for effective training. Within the framework of neuroprosthetics, numerous studies in electroencephalography (EEG) confirm the existence of neural correlates when humans perceive the erroneous actions of the device. Subsequently, the decoding of this correlate has been used to correct the erroneous behavior performed by the agent or to tune the behavioral strategy of the agent, among others. However, a main limitation of current approaches is that the actions of the agent were discretized, thus restraining the usability of such systems. The main objective of this PhD study is to study, and decode, the neural correlates of error evaluation under continuous trajectories performed by external agents; and to use this decoding to tune the continuous behavior of the agent for individual users. To accomplish this goal, two essential questions will be investigated: (i) whether it is possible to infer individual preference under continuous state-action scenarios, and (ii) how to create a reliable decoding pipeline in a continuous fashion. Results obtained during the first year of the PhD have confirmed the existence of such correlates under continuous motions of a robotic arm. Furthermore, such correlates encode individual preferences, indicating that the neural prosthesis can be also customized for individual users, which may play an important rule to increase the quality of brain-computer based assistance. This property not only will increase the level of perceived assistance provided by a brain-computer interface, but also may facilitate embodiment of the brain-controlled device.

EPFL2021

EEG Error-related Potentials Detection with a Bayesian Filter

Pierre Bessière, Ricardo Andres Chavarriaga Lozano

Several studies describe evoked EEG potentials elicited when a subject is aware of an erroneous decision either taken by him or by an external interface. This paper try to detect Error-related potentials (ErrP) elicited when a human user want to monitors an external system upon which he has no control whatsoever. To this end we use a Bayesian filter to classify erroneous or correct events. On average over three subjects, the proposed probabilistic classifier achieves single-trial classification of 85% for correct trials and 71% for erroneous trials.

IEEE2009

, ,

Error-related EEG potentials (ErrP) can be used for brain-machine interfacing (BMI). Decoding of these signals, indicating subject's perception of erroneous system decisions or actions can be used to correct these actions or to improve the overall interfacing system. Multiple studies have shown the feasibility of decoding these potentials in single-trial using different types of experimental protocols and feedback modalities. However, previously reported approaches are limited by the use of long inter-stimulus intervals (ISI>2s). In this work we assess if it is possible to overcome this limitation. Our results show that it is possible to decode error-related potentials elicited by stimuli presented with ISIs lower than 1s without decrease in performance. Furthermore, the increase in the presentation rate did not increase the subject workload. This suggests that the presentation rate for ErrP-based BMI protocols using serial monitoring paradigms can be substantially increased with respect to previous works.

2015

Functional meaning of oscillatory brain activity in various frequency bands in the human electroencephalogram (EEG) is increasingly researched. While most research focuses on event-related changes of brain activity in response to external events there is also increasing interest in internal brain states influencing information processing. Several studies suggest amplitude changes of EEG oscillatory activity selectively influencing cortical excitability, and more recently it was shown that phase of EEG activity (instantaneous phase) conveys additional meaning. Here we review this field with many conflicting findings and further investigate whether corticospinal excitability in the resting brain is dependent on a specific spontaneously occurring brain state reflected by amplitude and instantaneous phase of EEG oscillations. We applied single pulse transcranial magnetic stimulation (TMS) over the left sensorimotor cortex, while simultaneously recording ongoing oscillatory activity with EEG. Results indicate that brain oscillations reflect rapid, spontaneous fluctuations of cortical excitability. Instantaneous phase but not amplitude of oscillations at various frequency bands at stimulation site at the time of TMS-pulse is indicative for brain states associated with different levels of excitability (defined by size of the elicited motor evoked potential). These results are further evidence that ongoing brain oscillations directly influence neural excitability which puts further emphasis on their role in orchestrating neuronal firing in the brain.

Hindawi Publishing Corporation2014

Vision in clutter

Vitaly Chicherov

Perception of a visual target can strongly deteriorate in the presence of flanking elements (crowding). Crowding is the main limiting factor in reading and a major contributor to poor vision in amblyopia. In crowding, elements appear jumbled and thus are hard to recognize. Classic pooling models propose that crowding occurs via pooling of information from low-level to high-level visual areas. Because of pooling, receptive fields become larger and, at final stages, encompass both the target's and flankers' locations. As a result, the brain is unable to perceive the target separately from the flankers. Pooling models predict that adding flankers or increasing the size of the flankers should always increase crowding strength. In a series of previous studies from our laboratory and other groups, this view was shown to be over-simplistic and incomplete. In many situations, crowding can be reduced and good performance recovered by adding flanking elements or by increasing their size. This can occur when flankers ungroup from the target and group with each other. Grouping between elements is thus a key component in crowding. The neural mechanisms of crowding are unknown at the moment. Here, I used EEG to study the time course of crowding and the brain areas involved in the processing of crowding. Using event-related potentials (ERPs), I showed that crowding is a late visual process. In particular, the N1 component of ERPs (around 190 ms) was suppressed in crowding while earlier components did not correlate with crowding. The N1 suppression was caused by reduced neural activity in high-level visual areas such as the lateral occipital cortex. In a second study, I attempted to replicate the results with gabor stimuli. In a third study, I used steady-state visually evoked potentials (ssVEPs) to disentangle target and flanker processing in crowding. In this study, I showed that the target is selectively suppressed when it groups with the flankers. In addition to these three projects on crowding, I worked on time-frequency analysis that I performed on backward masking data in schizophrenia patients.

EPFL2015

Topographical instability of event-related potentials in P300 wave study

Vitaly Chicherov

Microstate analysis of ERP and EEG topographies reveals stable patterns of brain activity. Little is known about the changes from one stable state to another. During such changes significant topographic instabilities are observed together with low ERP field power and high values of topographical dissimilarity. Although topographical dissimilarity indicates how stable a given topography is, it does not indicate the exact limits of stable and unstable intervals. In addition, it does not account for the influence of background noise on the degree of instability. In the current study, we introduce a measure that identifies unstable periods in topographical dynamics.We investigated the characteristics of stable and unstable periods in ERPs recorded in a target-detection paradigm where subjects counted target letters. Despite large variations among subjects, four main stable topographic periods and four unstable periods were found in the majority of ERPs. Stable periods were closel y related to the ERP waves (N1, P3a, P3b, N400). Scalar magnitudes of ERP vectors were significantly larger for the target condition than for the non-target condition in the time intervals containing the N1, P3b, and N400 stable periods. The longest unstable period occurred between the N1 and P3a waves (40 ms on average). The results suggest that our measure is a useful tool to identify and study the properties of unstable periods in ERPs.

2011

Decoding stimulus-related information from single-trial EEG responses based on voltage topographies

Michael Herzog, Gijs Plomp, Athina Tzovara

Neuroimaging studies typically compare experimental conditions using average brain responses, thereby overlooking the stimulus-related information conveyed by distributed spatio-temporal patterns of single-trial responses. Here, we take advantage of this rich information at a single-trial level to decode stimulus-related signals in two event-related potential (ERP) studies. Our method models the statistical distribution of the voltage topographies with a Gaussian Mixture Model (GMM), which reduces the dataset to a number of representative voltage topographies. The degree of presence of these topographies across trials at specific latencies is then used to classify experimental conditions. We tested the algorithm using a cross-validation procedure in two independent EEG datasets. In the first ERP study, we classified left- versus right-hemifield checkerboard stimuli for upper and lower visual hemifields. In a second ERP study, when functional differences cannot be assumed, we classified initial versus repeated presentations of visual objects. With minimal a priori information, the GMM model provides neurophysiologically interpretable features - vis a vis voltage topographies - as well as dynamic information about brain function. This method can in principle be applied to any ERP dataset testing the functional relevance of specific time periods for stimulus processing, the predictability of subject's behavior and cognitive states, and the discrimination between healthy and clinical populations. (C) 2011 Elsevier Ltd. All rights reserved.

Elsevier2012