By analyzing fMRI scans of a subject?s brain, neuroscientists are able to identify images that subjects have seen. Although the process has no practical application currently, the technology is attracting attention by both scientists and alarmists.
The research team lead by Shinji Nishimoto and Jack Gallant at UC Berkeley use functional magnetic resonance imaging (fMRI) to measure brain activity in the visual cortex to reconstruct what the subject saw. A motion encoding model was developed identifying the spatial position, motion direction and speed filters the brain uses to make sense of visual input.
They foresee the technique helping to diagnosis brain conditions and expand research in neural prostheses or brain machine interfaces. Neurological evaluation and diagnosis involving conditions such as stroke or dementia might be improved with the help of such techniques. Gallant says, "The visual system is also tightly integrated with other sensory subsystems and systems for memory and language." Applications in visual trauma and even blindness could be pursued. It could possibly assist in determining the mental state of a person who lies in a coma.
Original and Retrieved Image
However, alarmists jump to conclusions involving nefarious uses. The foremost threat, they think, is that the technology would be used to extract scenes from a person’s memory. They conjure up spy movies or science fiction novels where subjects are forced to undergo treatment to reveal covert information. Dream images might be used to uncover unconscious motives, or plans. Big Brother would not only be watching, he would be probing, and worse yet, interpreting to meet forgone conclusions.
Gallant did admit: "It is possible that decoding brain activity could have serious ethical and privacy implications downstream in, say, the 30 to 50-year time frame." For now, the process is too complicated, involves too many sophisticated devices, and requires such in depth knowledge that the casual malevolent is incapable of duplicating their work.
Jack Gallant, is Professor of Psychology at the University of California at Berkeley, and is affiliated with the graduate programs in neuroscience, bioengineering, biophysics and vision science. He says that in his lab, "Our main experimental tools are functional MRI and electrophysiology; our main theoretical tool is nonlinear system identification (‘reverse engineering’)." Gallant?s lab also is involved in the STRFPak system, a Matlab toolbox for estimating the linear and nonlinear stimulus-response mapping function of sensory systems. The software can be used to estimate quantitative encoding models for any sensory system, based on data collected using neurophysiology, EEG or fMRI.
