Skip navigation Learning: It's a memory thing
1. Brain: Watch me learn!

2. Eye on the neuron

3. Hip, hip, hippocampus!

4. Synapses: Watch 'em grow!

5. What babies remember




All microscopic images, this page, courtesy Wen-Biao Gan, New York University (Scale bars are one micron - one-millionth of a meter - long)


Using genetic engineering and a nifty microscope, scientists are watching nerve cells make connections in the brain.




Neurons make a hookup
It's the ultimate question for scientists who study learning. Exactly how do neurons -- nerve cells -- change in a way that makes memories? If the brain is a giant network of neurons, there are essentially three possibilities:

New neurons form in the brain. Once thought impossible, this does happen, although rarely;

Existing connections between neurons get stronger; or

New connections appear between existing neurons.

While neurons do form in the hippocampus, a brain memory center, they are too rare to explain learning, which is a common event (unless you overdose on TV...). And although many neuroscientists have theorized that learning may result when existing brain pathways speed up, there's now evidence for a basic rewiring -- for the formation of new connections between existing neurons.

Such connections, called synapses, allow nerve impulses to jump from one neuron to the next, and like everything else in a living brain, they are difficult to study. Now, however, two groups of researchers have watched synapses grow in living mice. Although they have come to opposite conclusions as to the relevance for memory, the ability to watch synapses grow is key to understanding learning at the cellular level.

 Image of two long, knobby neurons. In microscope images of a neuron in a young mouse, taken a month apart, arrowheads show a slender "filopodium" that converted into a spine -- one half of a synapse. The long arrow shows a filopodium that disappeared. Courtesy Wen-Biao Gan

The new studies depend on genetically engineered mice, whose neurons make a fluorescent protein, and the two-photon microscope, which can examine living tissue.

Wen-Biao Gan of the New York University School of Medicine studied the formation of synapses in the visual cortex of mice, by looking at two related structures:

Spines, tiny protrusions on the receiving end of a synapse.

Filopodia, a fancy name that means "filament-like feet," even smaller widgets that sometimes develop into spines.

Images show long, criss-crossed neurons. Images made three days apart show no change in the size or number of these spines in an adult mouse. Courtesy Wen-Biao Gan

The microscope revealed a range of changes in the neurons. In young, but not adult, mice, filopodia grew fast at a time when the visual cortex was developing quickly. And while more than 99 percent of the little feet disappeared before maturity, some did develop into spines.

Memory and the scrap heap
The scenario emerging from Gan's laboratory could represent a neural basis for learning: It seems that the young brain is testing synapses, and scrapping those that don't make the grade. Overall, he says, "Our idea was that you actually don't need to make many new synapses and get rid of old ones when you learn, memorize. You just need to modify the strength of the preexisting synapses for short-term learning and memory. However, it's likely that few synapses are made or eliminated to achieve long-term memory" says Gan. In essence, it looks as if the brain is making more synapses than it needs during development, and scrapping those that, for whatever reason, donĀ¹t help many based on early experience.

Images show change in one knob on a long neuron.In an adult mouse, this spine (arrow) grew in length and diameter in one month. Courtesy Wen-Biao Gan

While the discovery that synapses come and go does not prove that the new synapses participate in learning, the evidence is suggestive. But in a similar mouse study, Karel Svoboda of the Cold Spring Harbor Laboratory made images of synapses (see "Long-term ..." in the bibliography), and found that nearly all disappeared rather quickly. That makes them an unlikely candidate for long-term memory.

While the two studies have contrasting results, which Gan says may be due to differences in experimental technique. His lab is now studying mouse brains while the animals are learning little mousie facts.

Even if Gan is correct -- that new synapses play a role in learning -- the strengthening-of-synapses theory may still hold water. In both cases, he says, "You want to understand the extent, the degree of change" in the connection of brain cells.

Furthermore, because the combination of two-photon microscopy and fluorescent brain cells gives such a good view of living neurons, he says, "This approach will be very important to study not only learning and memory, but also disease-related changes in structure and function of the brain, like Alzheimer's, stroke and epilepsy. You can look at very fine changes in the brain, can understand better the disease mechanism, and perhaps study drug treatment."

If memory dims with age, why don't you remember infancy?

The Why Files

There are 1 2 3 4 5 pages in this feature.
Bibliography | Credits | Feedback | Search

©2003, University of Wisconsin, Board of Regents.