A novel near-infrared spectroscopy-based brain-computer interface for the detection of emotional valence in children

Erica Floreani (1,2), Tom Chau (1,2)

1. Institute for Biomaterials & Biomedical Engineering, University of Toronto

2. Bloorview Research Institute, Holland Bloorview Kids Rehabilitation Hospital


Children born with severe motor impairments are limited in their options for augmentative and alternative communication (AAC) technologies. Brain-computer interfaces (BCIs) are devices that bypass the need for motor control by directly analyzing and interpreting brain activity to control AAC applications. While BCIs have been used as an access pathway for adults with severe motor impairments, minimal research has investigated BCIs for children. In addition, existing BCI paradigms often rely on higher language skills such as spelling that would be unsuitable for pre-literate children. An alternative approach is to develop a BCI that can identify emotional states, which could be used to directly communicate feelings. So-called affective BCIs have been developed to detect emotional states in adults, although widespread methodological variations have led to inconsistent results. The primary goal of this project is to develop a BCI that can distinguish between positive and negative emotional states in typically developing children, to investigate the feasibility of using emotion as an access pathway to communication for children with severe motor impairments.


Near-infrared spectroscopy (NIRS), an imaging modality that measures activity-related changes in oxygenated blood flow to the brain, will be used to identify changes in emotional state from the prefrontal cortex, a region of the brain implicated in emotional processing. Typically developing children aged 8-14 will be recruited to undergo a series of emotion-induction trials, viewing affective stimuli to elicit an emotional response. After offline training, the BCI will be tested online, where classification results will be displayed in real-time using visual feedback as participants attend to the affective stimuli. BCI performance will be evaluated in terms of accuracy, sensitivity and specificity and inter-session statistical analyses will be performed for each participant.


There is an urgent need to develop AAC devices for children with severe motor impairments. An emotion-detecting BCI would provide these children with a means of communicating that circumvents any need for functional motor control and avoids challenges due to possible cognitive or language delays. With access to communication, these children can engage within their communities, learn how to advocate for themselves, gain independence, and overall improve their quality of life.