science:

An interesting fact I learned from a reddit today, and which is supported by the scientific literature: schizophrenics can tickle themselves. Why is that?

The brain has a predictive model of the sensory results of different motor movements. For instance, if I move my hand in this way or that way, where is it going to end up? When you make the movement, this “predictor” is informed and forms a prediction about the sensation that will result—say, the sensation of a feather moving against your skin. When this sensory data comes in, it is compared to the prediction, and if they match up, the brain assumes that it was your self-generated movement, and thus can compensate for it. This is how we can distinguish between the exact same stimulus when applied by an external force or when applied by ourselves.

In an experiment, healthy subjects were instructed to tickle themselves by moving a piece of foam on their palm with their other hand. As predicted, the subjects did not find the sensation of their own tickling ticklish. Then their hand was hooked up so that, instead of directly tickling themselves, their hand movements were replicated by a robot hand. The researchers then introduced various delays or rotational translations of the movement—in effect, increasing the difference between the brain’s predictive model of the movement and the end result. The subjects reported that the sensation grew increasingly ticklish as the delay or rotation increased. In another experiment, however, schizophrenics reported no difference in the amount of ticklishness, whether the tickling was self-applied or applied by another person.

A current hypothesis about the origin of delusions of passivity, e.g., the idea that someone else is making you say or do the things you do, a classic symptom of schizophrenia, is that it is due to faults in this predictive model. Evidence suggests that schizophrenics fail to generate this predictive model of what their own movements will result in, which manifests as the inability to distinguish external and internally generated stimuli, like in the tickling example. In healthy non-schizophrenics, the predictive model and the sensory input that results from a movement will reinforce each other; the brain predicts that the movement of a hand will result in a certain sensation, and the sensation confirms it. This also allows us to self-correct before we even begin a movement if we detect that the intended result and the predicted result don’t match up. But schizophrenics tend not to make such self-corrections. One can imagine how delusions of passivity could result if the “predictor” fails to make a prediction: when it then tries to compare the end result to a non-existent prediction, it is as if someone else had willed the action. After all, in healthy subjects, when we receive “unexpected” (i.e., not predicted by the model) stimuli, it is usually of external origin. As in, when another person tickles us.

26 April 305 science Permalink