Adversarial Stimuli: Attacking Brain-Computer Interfaces via Perturbed Sensory Events
This work addresses security vulnerabilities in brain-computer interfaces, which could impact users relying on these systems for communication or control, and is incremental by applying adversarial attack concepts from machine learning to a new domain.
The paper investigates whether adversarial perturbations in sensory stimuli can degrade the performance of EEG-based Motor Imagery Brain-Computer Interfaces, finding that minor adversarial stimuli significantly deteriorate BCI performance across all participants (p=0.0003) and are more effective under induced stress.
Machine learning models are known to be vulnerable to adversarial perturbations in the input domain, causing incorrect predictions. Inspired by this phenomenon, we explore the feasibility of manipulating EEG-based Motor Imagery (MI) Brain Computer Interfaces (BCIs) via perturbations in sensory stimuli. Similar to adversarial examples, these \emph{adversarial stimuli} aim to exploit the limitations of the integrated brain-sensor-processing components of the BCI system in handling shifts in participants' response to changes in sensory stimuli. This paper proposes adversarial stimuli as an attack vector against BCIs, and reports the findings of preliminary experiments on the impact of visual adversarial stimuli on the integrity of EEG-based MI BCIs. Our findings suggest that minor adversarial stimuli can significantly deteriorate the performance of MI BCIs across all participants (p=0.0003). Additionally, our results indicate that such attacks are more effective in conditions with induced stress.