(posted by S.Mason from email note) I have read your technical report from BCI-Info.org, and I have a few comments:
* I'm not fond of the use of the word "automated" in 2.1.2.1. Something with "habituated" or "committed to procedural memory" flavour would meet my expectation. * To reduce confusion, remove figure 9b. 9a is sufficient to show the difference you are describing. 9b draws your attention away from the point of interest. Also, I'd mention the differences in performance which may result from real-time feedback to the user. Real-time feedback for the off-line analysis won't result during its testing, however, the prior real-time feedback of the "old" analysis will contaminate (perhaps not significantly) the off-line analysis. * 2.2.3 - "we felt that BCI users..." - It's not really the case that the BCI user gets to pick and choose the transducer and match it to their AT. Nor do we currently test one transducer against another with the same AT and the same user. The user's point of view sounded inappropriate to me. I would recommend changing the text to outline that TPa characteristics can be objectively measured most easily, and that their impact on the output at any other TP(b or c) can be rationalized by determining the "transfer functions" of those parts of the system (and the user prefernce the operation at those levels). All-in-all... a good read... good work! Steve Badelt _______________________________________________ BCI-info mailing list [email protected] https://mlist.tugraz.at/mailman/listinfo/bci-info
