Thought police may be coming

Mind reading by means of electronics? The idea would sound awfully far-fetched, if so many researchers weren’t actually doing it — barely, sort of, and clunkily, but still doing it.

In recent years I’ve read a lot of accounts of somewhat successful attempts to control mentally the movements of a cursor on a computer screen, for example. Control-by-thought is coming. In fact, it has been commercialized, to control video games.

From the site of Emotiv Systems: “Project Epoc is a headset that uses a set of sensors to tune into electric signals naturally produced by the brain to detect player thoughts, feelings and expression.

“It connects wirelessly with all game platforms from consoles to PCs.” There is a lot of this. The Institute of Electrical and Electronics Engineers reports successful prediction, by electroencephalogram, of which finger a subject is about to move.

I’ve seen reports of control of a wheelchair by computer analysis of signals from electrodes implanted in the brain.

The details of such research would fill volumes. The essence is this: When people think about particular things, specific changes occur in their brains. In principle, these changes can be detected, thus revealing, at least to some extent, what they are thinking about.

In practice, such mind reading is rudimentary, but improving. We are nowhere near reading specific thoughts: “I prefer Renoir to Cezanne because …” A major difficulty is that of instrumentation. One problem is that the machines used to measure the brain’s activity often are large, as well as hard to use outside of a laboratory.

Magnetic resonance imaging (MRI) is a good example, requiring the subject to lie quietly inside a large tube. For research, this works. For practical applications, probably not.

The other problem is that the techniques for measuring changes in the brain — positron emission tomography, MRI, electroencephalography, etc., lack sufficient resolution. The situation resembles that of digital photography a few years back: A one-megapixel camera just didn’t capture enough detail.

However, one thing that has become a constant is that instrumentation improves fast.

Among consumer products, computers, cameras and IPods are good examples. Laboratory instrumentation is doing the same thing. This means that, short of an asteroid strike, in 10 or 20 years neuroscientists will be far more able to detect detailed changes in the brain.

Just how, I don’t know. Neither, probably, do they yet.

Now, what will be, or may be, the upside? Lots. For example, if you are a quadriplegic in a wheelchair, how do you control the chair, or use a computer? Well, when healthy people type, changes occur in their brains and signals exit the brain and travel to the fingers. If sensors can detect these changes, they can be used to operate the keyboard. This would have been a wacky idea 30 years ago. Today, it’s almost a question of when.

The downside? Here I’m speculating because I don’t know how well it will be possible to tell what people are thinking. It is one thing to detect a desire to move a finger, another to detect emotional states or truthfulness, and quite another to detect the thought, “I think I’ll strangle the boss.” Detailed thoughts may never be readable.

However, there is something unnerving in the thought of no longer enjoying privacy in one’s own mind.

Story Continues →