Adaptive parameter setting in a code modulated visual evoked potentials BCI

Federica Turi (1, 2) , Maureen Clerc (2)

1 COMUE UCA - COMUE Université Côte d'Azur (2015-2019)
2 ATHENA - Computational Imaging of the Central Nervous System

Code-modulated visual evoked potentials (c-VEPs) BCI are designed for high-speed communication. The setting of stimulus parameters is fundamental for this type of BCI, because stimulus parameters have an influence on the performance of the system. In this work we design a c-VEP BCI for word spelling, in which it is possible to find the optimal stimulus presentation rate per each subject thanks to an adaptive setting parameter phase. This phase takes place at the beginning of each session and allows to define the stimulus parameters that are used during the spelling phase. The different stimuli are modulated by a binary m-sequence circular-shifted by a different time lag and a template matching method is applied for the target detection. We acquired data from 4 subjects in two sessions. The results obtained for the offline spelling show the variability between subjects and therefore the importance of subject-dependent adaptation of c-VEP BCI.

Domaines: Interface homme-machine [cs.HC]; Informatique; Traitement du signal et de l'image [eess.SP]; Neurosciences

Federica Turi, Maureen Clerc. Adaptive parameter setting in a code modulated visual evoked potentials BCI. 8th Graz Brain-Computer Interface Conference 2019, Sep 2019, Graz, Austria. ⟨hal-02303562⟩