The challenge of controlling an auditory BCI in the case of severe motor disability

Table 3 Controls and patients’ event-related potentials

Subjects	Online accuracy*	Response to deviance	Attentional modulation
Subjects	Online accuracy*	Response to deviance	Standards	Deviants	Total
S1	100	✓	✓	✓	✓
S3	68	✓	Ø	Ø	Ø
S4	80	✓	✓	✓	✓
S5	88	✓	✓	Ø	✓
S6	92	✓	✓	✓	✓
S7	92	✓	✓	✓	✓
S8	96	✓	✓	✓	✓
S9	82	✓	✓	✓	✓
S10	96	✓	✓	✓	✓
S11	96	✓	✓	Ø	✓
S12	72	✓	✓	✓	✓
S13	94	✓	✓	✓	✓
S14	80	✓	✓	Ø	✓
S15	60	✓	Ø	✓	✓
S16	96	✓	✓	✓	✓
S17	94	✓	✓	✓	✓
S18	96	✓	✓	✓	✓
S19	74	Ø	Ø	Ø	Ø
Prevalence of healthy subjects with BCI control	94%	94%	83%	72%	89%
LIS1	Chance level	Ø	Ø	Ø	Ø
LIS2	Chance level	Ø	Ø	Ø	Ø
LIS3	Chance level	Ø	Ø	Ø	Ø
ALS1	Chance level	Ø	Ø	Ø	Ø
ALS2	Chance level	Ø	Ø	Ø	Ø
ALS3	100	✓	✓	✓	✓
ALS4	100	✓	✓	✓	✓
Prevalence of patients with BCI control	29%	29%	29%	29%	29%

✓: presence Ø: absence
Response to deviance: presence or absence of a significant difference of event-related potential for standards versus deviant sounds
Attentional modulation: presence or absence of significant difference of event-related potential for attended versus ignored condition
* Note that for patients LIS1 and LIS2, the online accuracy could not be measured as they did not undergo any test session. This is because the calibration phase proved inconclusive (cross-validation was at chance level). We thus report a chance level performance here (see Table 4 for more information)

ISSN: 1743-0003