BrainInSight : a brain-computer interface to decode attention and perception in real time – BrainInSight

At the junction between neuroscience, computer science, and robotics, brain-computer interfaces (BCIs) and neural prosthetics have emerged as a tremendous opportunity to build on core scientific knowledge to alleviate the life of patients with motor disabilities (Lebedev and Nicolelis, 2006; Nicolelis and Lebedev, 2009; Green and Kalaska, 2010; Fetz, 2007; Hatsopoulos and Donoghue, 2009). The general aim of this rapidly developing field is to use preserved electrophysiological nervous activities in order to counter specific dysfunctions or deficits by driving external palliative devices such as a cursor on a computer screen, a robotic arm or a wheelchair, thus restoring mobility and independence to patients with central or peripheral motor disabilities. One of the most representative examples of a palliative neural prosthesis is the use of the neuronal activity of the motor cortex of a tetraplegic patient to help him control a computer driven environment, thus allowing him a certain degree of assisted mobility and independence (Donoghue et al., 2007; Hochberg et al., 2006). Another important advance in the field is the demonstration that other regions than the motor cortex can be used to drive BCIs, such as parietal cortex (Musallam et al., 2004), or dorso-lateral prefrontal cortex (Vansteensel et al., 2010) thus providing a potential substitute of motor cortex activities when these are not available, following for example an acute injury of this region. The foreseen impact of this field on the life of thousands of patients is huge and directed to encompass a diversity of pathological conditions including injuries of the motor system (e.g. lesions of the motor cortex, the corticospinal pyramidal tract or the spinal cord), certain types of neurodegenerative conditions (e.g. cerebral palsy, amyotrophic lateral sclerosis) or distal limb injuries (e.g. amputations).
However, while most of the research effort in neural prosthetics has concentrated on the use of motor signals to drive external devices, new directions in the field of BCIs are also emerging. For example, a recent study has demonstrated that incorporating sensory feedback to a motor BCI improves its performance (Suminski et al., 2010). On another line, Musallam et al. (2004) have shown, in the context of a motor behavior, that cognitive signals such as the expected value of a reward, i.e. the subject’s motivation, can be decoded from parietal neural activity. The work of Jerbi and collaborators also demonstrates that such signals as attention orientation signals and mental calculation signals can be used to drive a cognitive BCI (Jerbi et al., 2009).
Here, we propose to develop a brain-computer interface in the non-human primate that allows for a real-time readout of endogenous cognitive variables as opposed 1) to sensory variables reflecting the sensory inputs to the nervous system or 2) motor variables reflecting the motor outputs i.e. the endpoint choice of action of the subject onto the environment. Our objective is to achieve reliable beforehand prediction of the animal’s overt behaviour and covert representations on the basis of single trial activities collected simultaneously at two different cortical sites and analyze the susceptibility of this readout to changes in the cognitive demands, in the sensory inputs and in the functional integrity of the cortical network of interest. The cognitive variables we will focus on are: spatial selection/attention, perception and response selection. As a result, our recordings will be targeted to the fronto-parietal attentional network and more specifically simultaneously to the lateral intraparietal area LIP and to the frontal eye field FEF.
We believe that accessing this type of information will pave the way to a new generation of intelligent BCIs with a high potential to improve the life of patients affected by acute, developmental or degenerative neurological deficits.