Brain Responses to Errors During 3D Motion in a Hapto-Visual VR
We investigated brain potentials recorded by electroencephalography (EEG) signals in response to unpredictable haptic/kinesthetic disturbances to a continuously moving object in a hapto-visual 3D virtual world that highly resembles reality. Participants m
- PDF / 1,815,755 Bytes
- 11 Pages / 439.37 x 666.142 pts Page_size
- 63 Downloads / 182 Views
Abstract. We investigated brain potentials recorded by electroencephalography (EEG) signals in response to unpredictable haptic/kinesthetic disturbances to a continuously moving object in a hapto-visual 3D virtual world that highly resembles reality. Participants moved a virtual object from an initial position to a target position in the virtual environment. A large cylinder obscured part of the motion between the origin and the target. The position of the emerging object under the cylinder is disturbed, and hence unexpected, for part of the scenarios. This disturbance is perceived as an error. We examined the EEG signals locked to the error. Our results show a consistent disturbance-locked potential with an early negative peak followed by a positive peak. Peak-to-peak amplitude increased with the disturbance magnitude. Source estimation at the time of the negative and positive peaks revealed a strong activity in the vicinity of Brodmann area (BA) 7, known to be involved in hapto-visual integration and in the neural computation of dynamic motor errors. These results demonstrate the presence of haptic-disturbance-related brain activity under conditions of continuous motion. Results further suggest a feedback signal for error detection and correction in EEG-based Brain-Computer Interfaces (BCI) in applications such as telesurgery, manipulation of remote objects and rehabilitation. Keywords: EEG
ERP BCI Error Motion 3D
1 Introduction How is motion performance affected in virtual worlds, or telemanipulation systems when the haptic interface goes wrong? Recent studies showed that haptic feedback is crucial for user performance [1, 2] of motor tasks and for Brain-Computer Interfaces (BCIs) [3]. Haptic feedback contributes to user’s intent recognition for BCI systems [3], to closure of sensory-motor loop [2, 3] and improves user awareness of interaction in the task world [1, 2]. On the other hand BCI systems are prone to misclassification of user intent up to 30 % of the cases [4, 5]. Interface errors can occur in any type of controlled motion, and are defined as execution errors. Execution errors occur when an unexpected movement was performed instead of the intended one [6–8]. Error Related Potentials © Springer International Publishing Switzerland 2016 F. Bello et al. (Eds.): EuroHaptics 2016, Part II, LNCS 9775, pp. 120–130, 2016. DOI: 10.1007/978-3-319-42324-1_12
Brain Responses to Errors During 3D Motion in a Hapto-Visual VR
121
(ErrPs) evoked by errors and embedded in the ongoing Electroencephalography (EEG) (non-invasive brain signal recording) [8] are used to decrease the misclassification rate [4]. ErrP is a type of Event Related Potential (ERP) locked to an error [8, 9]. BCI systems with integrated ErrP feedback, decreased the misclassification rate from 30 % to 7 % [4]. ErrPs are characterized by an initial negative fronto-central component with latency in range of 50–100 ms (msec), known as error-related negativity (ERN) followed usually by a positive component with centro-parietal prominence and la
Data Loading...
