Behavior/ J. of NIH Research, Nov.,1995 With Practice, The Motor Cortex May Reorganize Itself It's not easy to type 100 words per minute or play a Chopin sonata, but it can be done--with practice. Now, in the Sept. 14 issue of Nature, Avi Karni and his colleagues at the National Institute of Mental Health (NIMH) shed some light on just how practice makes perfect. To study the way people learn, Karni, Leslie Ungerleider, and Gundela Meyer of NIMh used functional magnetic resonance imaging (MRI) to measure blood flow to specific brain regions. Their results suggest that over time, the primary motor cortex undergoes a reorganization as people develop expertise with a new motor skill. In previous studies, Karni and others had shown that over the course of a single motor-learning session, as a subject's performance improves, brain blood flow in motor and sensorimotor regions of the cortex first increases, then declines. That finding led some researchers to speculate that the reason for the declining blood flow is that as subjects improve at a task, they no longer need to devote so much of their brain to it. In other words, it becomes more "automatic" and requires less "thought." Karni, however, suspected that the blood-flow pattern might change with time because one's ability to perform complex motor tasks such as touch-typing or playing the piano improves over weeks or months -- or years. He and his colleagues trained six adult men to tap their fingers in a specific sequence -- pinkie (identified as finger no. 4), index finger (1), ring finger (3), middle finger (2), pinkie (4) -- as quickly and accurately as they could. As expected, their performance improved during their first training session, and blood flow to the primary motor cortex increased, then declined as they got better at the finger-movement task. As subjects returned to the lab for testing each week over a period of five weeks, their performance continued to improve-- but now, blood flow to the primary motor cortex began to climb again, and the increased blood flow had spread to a larger area of the motor cortex. Another surprise was that there was no transfer of learning between similar motor tasks. Practicing on the sequence 4-1-3-2-4 did not help the subjects' performance on the sequence 4-2-3-1-4. Clearly, the change in motor skill was not related to a general improvement in the subject's ability to move their fingers, something that might be expected to be controlled by the primary motor cortex, but rather to an improvement in executing a particular sequence of movements, something that would require extensive motor control. Karni says these results challenge accepted wisdom about motor learning. "The primary motor cortex is supposed to be hardwired since puberty," he says. Yet for his subjects, something about the primary motor cortex was changing over time. Now, Karni and his colleagues intend to turn their functional MRI on areas of the motor cortex that are thought to control more complex movements to catch a glimpse of how they, too, change over time.