Designing AAC Interfaces for Commercial Brain-Computer Interaction Gaming Hardware
Stephen Steward · 2009 · Proceedings of the 11th International ACM SIGACCESS Conference on Computers and Accessibility (Assets '09) · doi:10.1145/1639642.1639708
Summary
This paper explores repurposing a low-cost commercial brain-computer interface (BCI) gaming device — the OCZ Technologies Neural Impulse Actuator (NIA) — as an input device for augmentative and alternative communication (AAC). The NIA is a headband that reads electrical signals from facial muscles and translates them into keystrokes via its software interface. Unlike traditional EEG skull caps that interpret low-level brain signals directly, the NIA works at a higher level, converting configurable facial actions (glances left or right, cheek tension, jaw tension) into keystroke outputs. Sold at roughly $140 versus $2,100 for research-grade BCI devices, the NIA represents a dramatically cheaper entry point. The author designed a text-entry interface modeled on row-column scanning — a familiar AAC paradigm — that maps to the NIA's reliable input actions. The research was motivated by the observation that people with severe speech and motor impairments often have interaction limited to a single switch, which severely constrains communication rate, and that BCI technology could offer additional input channels.
Key findings
The author developed two interface variants — a two-button and a three-button row-column scanning interface — tailored to the NIA's strengths and limitations. The NIA reliably supported three distinct facial actions (left glance, right glance, and muscle tension), so the interfaces were designed around at most three inputs. In initial testing, the two-button and three-button interfaces achieved equal speed, but the three-button version produced fewer errors. A key usability issue emerged with the glance-based input: users had a tendency to trigger an unintended glance in the opposite direction when returning their eyes to the screen, causing false inputs in the three-button interface. Two mitigation strategies were identified — adding a built-in delay after each action (effective but slower) or training users to slowly return their gaze (effective but requiring practice). The two-button interface was unaffected by this issue since it used only right glance and muscle tension. The author notes that a diamond-shaped letter layout may be better suited to three-input navigation than the traditional row-column grid.
Relevance
This work is an early example of the creative reuse of consumer gaming hardware for assistive purposes — a theme that has grown significantly as gaming peripherals have become more sophisticated. The core insight that affordable commercial devices can be adapted for AAC use remains highly relevant, especially as consumer EEG headsets and other biosignal devices continue to improve. The paper highlights important interface design considerations for BCI-based AAC: the number of reliable distinct inputs determines the interaction paradigm, and error prevention is as important as speed. For practitioners, the study underscores that the interface design must be matched to the specific capabilities and failure modes of the input device, rather than simply mapping new hardware onto existing interaction patterns. The work also points to the ongoing challenge of making BCI technology accessible and affordable for people who would benefit most from it.
Tags: augmentative and alternative communication · brain-computer interface · assistive technology · motor impairment · speech impairment · switch access · interface design