In some respects, fruit fly brains work very much as do the brains of other animals, including humans. Both flies and humans are capable of the kind of "associative learning" made famous by Pavlov's dogs. (After ringing a bell and presenting dogs with food several times over a few days, the Russian physiologist Ivan Pavlov [1849-1936] found that eventually his dogs would display dinnertime behavior [drooling, excitement] upon hearing the sound of the bell alone.) Josh and Tim study olfactory associative learning by training flies to avoid a particular odor and later measuring the flies' odor avoidance behavior. They found that memories based on olfactory associative learning can be acquired and stored in the absence of electrical activity in mushroom bodies. However, those memories cannot be recalled in the absence of such electrical activity.

These results suggest that memories are acquired and stored by chemical processes in mush-room bodies, and that electrical activity in this part of the brain is necessary only to recall those memories. In some instances, when it comes to learning and memory, the fly brain (and perhaps ours) must function electrically to recall what is stored chemically.

Brain Imaging: Karel Svoboda continues to pioneer the application of a high-resolution, real-time imaging technique called two-photon microscopy to brain research. Recently, Karel and his col-leagues have used two-photon microscopy to examine abnormalities during brain development in a mouse model of Fragile X syndrome. In humans, Fragile X syndrome is the most commonly inherited cause of mental impairment. The syndrome is typically caused by a mutation in the FMR1 gene leading to absence of the Fragile X mental retardation protein or FMRP.

By using two-photon microscopy to compare the earliest stages of postnatal brain develop-ment in normal and FMR1 knock-out mice, Karel and Esther Nimchinsky have detected dramatic differences in the density and length of all-important neuronal structures called dendritic spines. Such spines form the familiar connections between neurons called synapses. When examined 1 week after birth, the "barrel" cortex in the brains of Fragile X mice displayed significantly increased density and length of spines. (Barrel cortex interprets neurological signals from the whiskers of newborn animals as they explore their world for the first time.) However, by 4 weeks after birth, dif-ferences in the density and length of spines between normal and Fragile X mice were largely undetectable.

The transient nature of the spine abnormality in Fragile X mice suggests that FMRP functions soon after birth during a critical window of opportunity for brain development. In addition, Karel's findings indicate that FMRP probably has a role in linking experience (e. g., whisker signals) to the growth of dendritic spines and that such links are required to specify the elaboration of a fully functional neuronal architecture in the brain.