The physical changes that occur when the brain makes a new memory have been observed for the first time, say researchers, who hope to go on to map the distribution of memory across brain regions.

Gary Lynch of the University of California, Irvine, and his colleagues examined the junctions between neurons — synapses — in three dimensions using a technique called restorative deconvolution microscopy (RDM). This consists of a sensitive light microscope with computer algorithms that analyse light scattered above and below the focal point, producing a three-dimensional 'trace' of an object's structure.

In previous work, the group developed a fluorescent marker that attaches to synapses in the brain that have recently undergone a certain type of neuron-to-neuron connection believed to be responsible for encoding memory, called long-term potentiation (LTP) (L. Y. Chen et al. J. Neurosci. 27, 5363–5372; 2007).

The team exposed live rats to a novel environment and allowed them to learn its layout. They then removed the animals' brains to examine the hippocampus — a region involved in memory — using RDM to observe individual synapses. A second group of rats was shown the new environment but not allowed to explore it before their brains were examined.

Only rats that had undergone learning and memory acquisition showed new synaptic growth, Lynch says (V. Fedulov et al. J. Neurosci. 27, 8031–8039; 2007). And the hippocampal synapses to which the LTP fluorescent marker attached were 50% larger than other synapses not involved in LTP.

We saw that the synapses had actually changed shape as a result of the new memory.

Furthermore, when the group looked at hippocampal slices from a third group of rats, which had been allowed to learn the same new environment but been given a drug to block LTP, the synapses showed no new growth, and were similar to those of the second group. This indicates that the new synaptic growth observed in the study is a result of LTP, Lynch asserts.

“We saw that the synapses had actually changed shape as a result of the new memory,” Lynch says. “They went from oval to circles, which have a greater surface area.” He now aims to use the technique to see which other areas of the brain might be involved in memory.

Being able to look at memory at the synaptic level is a major advance, says Mark Bear, a neuroscientist at the Massachusetts Institute of Technology in Cambridge. But he hesitates to accept the conclusions Lynch's team has drawn. “I don't think it's been proven that these [synaptic] changes represent the memory,” he says.