As we look at the world around us, our eyes are constantly on the move. But, somehow, what we see is not a series of disjointed, disconnected still pictures but a seamless film. How our brain manages to compensate for our rapid eye movements has puzzled neuroscientists for many years. On page 374, Marc Sommer and Robert Wurtz offer evidence that may help solve the mystery.

Wurtz, a neuroscientist at the US National Eye Institute in Bethesda, Maryland, had tried and failed to resolve the question back in 1968. But when Sommer joined his lab in 1998, a surprising result in a related project led them back to the problem — and offered them a solution.

The pair had been using monkeys to examine signals that travel from part of the brain's cortex that processes visual stimuli to an area of the brainstem linked with eye movement. Every now and then they saw something strange happen: a signal went in the opposite direction. A neuron in the brainstem fired, causing a response in the cortex. “The brainstem was talking back to the cortex,” says Sommer.

This result puzzled the two researchers, until they realized that they might be looking at the pathway that eluded Wurtz all those years ago. Maybe the brainstem was sending a signal to the cortex to alert it to an upcoming eye movement. Based on anatomical layout, they suspected that the signal was passing through 'relay' neurons in the thalamus on its way to the cortex. A series of preliminary experiments suggested that they were right, and the thalamus was involved. But going on to prove that the whole signal pathway existed was no easy task.

Credit: M. A. SOMMER

The pair set up a conceptually simple, if technically difficult, experiment. They used three probes implanted into the brain of a monkey: one to stimulate nerve cells in the brainstem, one to switch off neurons in the thalamus, and one to record the activity of a neuron in the cortex. “These are fundamental techniques of neurophysiology, but the challenge was to do all three at the same time in an awake animal,” says Sommer.

To get his measurements, Sommer had to painstakingly identify a single cell in the cortex that he could monitor. He then had to record its activity for three hours in a live, moving animal. As if all this was not difficult enough, Sommer had by this time begun to set up his own laboratory at the University of Pittsburgh. “I would fly back at weekends hoping to get one more neuron,” he recalls.

It took six months to record signals from eight neurons. But that was enough to show that the pathway exists. The brainstem seems to send a signal via the thalamus to the cortex to warn it of an impending eye movement. In response, neurons in the cortex adjust the position of the field of vision to where the eye will move next. When Sommer used the probe in the thalamus to switch off the relay neurons, the signal didn't get through and the cortex didn't take any steps to compensate for eye movement.

Now, Sommer and Wurtz want to find out what effect blocking this signal has on vision. Will the monkeys end up seeing the world as a series of disjointed snapshots? Sommer is currently devising tests to answer this question — but he first needs to find a way to measure what the monkeys are actually perceiving.