Cognition

Declarative memory—the ability to encode and recall facts, events, and arbitrary associations—critically depends on structures in the medial temporal lobe. In contrast, procedural memory, which includes the ability to learn motor skills, does not. For example, amnestic persons with bilateral medial temporal lobe lesions, like patient H.M., can learn procedural skills, whereas word list learning cannot be recalled. Both declarative and procedural memory, however, are dependent on frontal cortex activity. We studied the spatial and temporal localizations of brain regions associated with procedural memory using PET and [¹⁵O]H₂O to assess regional cerebral blood flow (rCBF). The results are illustrated in the images shown above.

The motor learning task consisted of the volunteer moving a robotic arm in a specified external force field (Field A) from a resting place to a target location. The volunteer’s movements were interfered with by a force produced by the robot that varied depending on on the subject’s hand motion. In the control condition, the volunteer’s movements were made without any perturbing forces. A second control condition used random forces to displace the reaching trajectory, but precluded the possibility of learning. Earlier work with this motor task has shown that a time-dependent pattern of consolidation occurs during the 5-6 hours after initial practice in which the subject moves the robot arm correctly. To find the neural correlates of motor memory learning and consolidation, an imaging experiment was performed. During the initial learning of Field A, the CNS regions significantly activated above the control condition were in the primary motor cortex and prefrontal cortical areas. Both regions showed the greatest activation early in the learning session with a gradual reduction in activation over 40 minutes of training in Field A. At 5.5 hours, consolidation had been allowed to take place in the memory of Field A, and when subjects were retested in this field, they showed increased activations in the parietal, premotor, and cerebellar cortices. These data suggest that new regions in the brain are involved in recalling motor memory at 5.5 hours, regions that did not demonstrate significant activation changes during memory acquisition. As the images demonstrate, regions in the prefrontal cortex appear to be important for this memory process. At 5.5 hours, after memory of A has undergone a functional reorganization, there is little interference in learning of another motor task. These studies emphasize the time-dependent changes that occur in the brain after initial acquisition of a motor skill and suggest that motor memory consolidation involves different regions of the brain at different times.

Image is courtesy of Dr. Holcomb.