From: Craig Hubley Subject: Normative Media (was Re: Emotion sensors) Date: Sun, 13 Oct 1991 05:12:09 -0400 Organization: UTCS Public Access In article <1991Oct10.194622.9368@milton.u.washington.edu> billon@chopin.udel. edu (John Billon) writes: >But emotions are telegraphed in a way that could perhaps be deciphered by >elecctronic sensors. A shoulder sensor to detect tension, and >changes in breathing and heart beat to imply a further change in emotion. I didn't read the original posting but certainly David Rokeby's Very Nervous System qualifies as a sensitive enough system to do this sort of interpretation. David tells a very funny story about a beltway bandit type coming up to him at SIGGRAPH, staring him in the face and then coming back with two guys in uniform, babbling to them about "immense military applications" and "instantaneous personality detection"... >However, an extension of this idea (I've seen stranger stuff in >this group) would imply a great deal. Agreed. I have a paper on this subject, which I call Normative Media. These differ from communicative media in that they can help to mold the responses of the user. My interest (and not coincidentally, the original interest of B.F. Skinner) was to empower individuals with tools to change their own behavior and improve themselves. For example, the "LifeSign" stop smoking computer or various time management systems. Or biofeedback systems in general. Main issues are how to personalize the medium to each particular user, and ensure that information is kept secure by the device/program/object that uses it. Privacy issues are the most pervasive. >A computer may become a more sensitive companion than the 'average' person. >A program using standard methods of statistical ai could... I used to call this the Sensitive Engine (after Kay's Reactive Engine) but now I simply call it a Very Personal Computer. >...concievably manipulate the emotions of the user, if it was so designed... Yes, although it might be more useful simply to "label" the emotional state and have heuristics regarding which tasks should and should not be performed in that state, etc. It would optimize your matching of time and mood to task. >It could concievably have a beneficial or damaging effect. This is just an >extension of interface technology, but I think a very interesting one that >could really use further study. > >Does anyone know of research being done in this area? There is quite a bit regarding various body sensors including EEG, EKG, GSR, body temperature, etc. This tends to be scattered among medical researchers and psychologists. I don't know if there is a special group or conference on body sensing technology. Certainly the glove, U-Force, Polhemus, etc., are simply more physically-oriented body sensors. Not to mention cameras and microphones. The Access.bus that DEC has proposed as an ACE standard (and put in the public domain) will probably drastically accelerate the use of non-traditional input devices. This is an equivalent to the Apple Desktop Bus, using shielded RJ12 connectors (Apple wouldn't license the ADB) and an object-oriented device definition protocol. Support chips cost about .50 per device and several vendors are already building them. There are lots of projects regarding gathering information on individuals in a traditional office environment: The Xerox PARC/Olivetti active badge. The Cavecat groupware system at the U of Toronto's Dynamic Graphics Project. Bellcore Cruiser. See Proceedings of CSCW for more. The intent of these is to gather cues (e.g. office door open or closed, where is chair facing, person on telephone ?) to aid message routing, interruption, group forming and dissolving. These systems tend to allow a lot of personalization and support individual definitions of privacy. > Thanks Alot, > John Billon Hope this helps. -- Craig Hubley -- Consultants in object-oriented technology & techniques, -- Craig Hubley & Associates -- user interface design & user productivity -- craig@gpu.utcs.Utoronto.CA UUNET!utai!utgpu!craig craig@utorgpu.BITNET craig@gpu.utcs.toronto.EDU {allegra,bnr-vpa,decvax}!utcsri!utgpu!craig 28 First Avenue, Toronto, Ontario M4M 1W8 Canada (US TC) call 416-466-4097 [MODERATOR'S NOTE AND REQUEST: We would love to supplement the usual engineering-oriented discussions of virtual worlds contained in our archived citations list with "softer" scientific investigations and findings. Please, if you have such citations, please send them to co- moderator Mark De Loura (madsax@hitl.washington.edu) for incorporation in the citations list. Thank you very much for this service. It helps move the state of the art forward. (Same goes, of course, for ANY cita- tions you think are of interest and value to other participants. [Thank you, Craig, for these leads. -- Bob Jacobson]