A variety of technologies are being pressed into service to “read minds,” and Canadian researchers have found they can determine a subject’s preference with 80% accuracy using infrared brain imaging. According to Sheena Luu, a doctoral student who led the research, “This is the first system that decodes preference naturally from spontaneous thoughts. Preference is the basis for everyday decisions.”
The research was published in the Journal of Neural Engineering. From the press release describing the work, it appears the goal of the work isn’t to enhance neuromarketing capabilities, but rather to provide a non-invasive brain-computer interface for use by handicapped individuals.
The 80% accuracy number was obtained when researchers asked subjects to choose between one of two drinks. Four times out of five, they were able to predict the subject’s choice from the infrared brain images.
It appears that the current infrared technology produces coarser images than fMRI or EEG. Devin Coldewey of CrunchGear comments,
Patients are trained to think of something that lights up a certain area of their brain — a strong memory, a song sung in the head, or a particular image — and this is picked up by the sensor and the patient can decide whether that’s a “yes” or “no” answer. That’s about the limit of things, but consider that it can be used by someone with absolutely no motor control in their body, including eyes and mouth, often the last resort for getting information through. Being able to say “yes” and “no” may sound basic, and it is, but it’s the basic things we take for granted. Hopefully this relatively straightforward and inexpensive method of thought detection will see a rollout sometime soon in hospitals and long-term care facilities.
In short, this may be an exciting and useful technology, but don’t expect infrared brain imaging to supplant fMRI or EEG techniques for neuromarkting studies.