It is more than 80 years since German psychiatrist Hans Berger first measured human brain activity using an early electroencephalograph that took electrode readings at the skull's surface. Since then, it has become possible to insert electrodes into the brain — and even, in animals, inside individual cells. But to establish the roles of single neurons in specific brain activities, such as sensory processing and behaviour, we need to be able to take concurrent readings from within neighbouring neurons in awake animals. Now, for the first time, two researchers in Switzerland have achieved just that.

Carl Petersen, a neurobiologist at the Swiss Federal Institute of Technology (EPFL) in Lausanne, set out to understand how information is processed by individual neurons, and how these cells communicate with one another. Using electrophysiological techniques that capture high-speed 'snapshots' of the electrical activity in single brain cells, Petersen and postdoc James Poulet succeeded in simultaneously recording the electrical activity inside pairs of neighbouring neurons in awake mice (see page 881).

By recording changes in the electrical potential difference that exists across each cell's membrane, the two were able to study the electrical correlations between neighbouring neurons when an animal changes its behaviour. They focused on the barrel cortex, a brain area responsible for processing tactile sensory information from the whiskers, during two different brain 'states': quiet and active. The active state was distinguished by 'whisking' — rapid, rhythmic waving of the whiskers back and forth — a feature of mouse exploratory behaviour. Whisking is akin to the way that humans visually scan or physically touch their environments, says Peterson, and experimental work has shown mouse whiskers to be as sensitive to touch and texture as human fingertips. Whisking ceases in mice that are in a quiet, relaxed state.

In mice in the quiet state, the membrane potentials recorded from neighbouring cell pairs swung dramatically in a slow, almost perfectly synchronized pattern. When the mice became active, the cells desynchronized and their electrical potentials fluctuated at higher frequencies. Petersen believes these results are analogous to electrical oscillations reported by Berger in 1929. He noted that humans relaxing with their eyes closed yielded patterns of slow, large-amplitude electrical waves, which disappeared upon eye opening.

Obtaining simultaneous recordings from two specific neurons in awake animals was technically challenging. A glass electrode had to be held perfectly still in cells just 10 micrometres wide. Petersen estimates that he and Poulet managed the feat about 5% of the time. To get the big picture of how every cell in the brain is involved in particular processes, Petersen says, “we somehow need to be able to measure from thousands or millions of cells at any given time, and that is going to be a big challenge.”

But the ability to record from neighbouring cells is an important step, he says, because it will allow further enquiry into how cells respond in relation to one another as sensory information comes in. Petersen hopes that by piecing together information one small region at a time, he and his colleagues will turn up important clues about brain function. “In my lifetime, I don't think we'll understand how the whole thing works, but we may be able to understand very small, isolated circuits.”