Abstracts

fMRI TECHNIQUES TO ASSESS LEARNING AND EPISODIC MEMORY

Abstract number : 1.067
Submission category :
Year : 2005
Submission ID : 5119
Source : www.aesnet.org
Presentation date : 12/3/2005 12:00:00 AM
Published date : Dec 2, 2005, 06:00 AM

Authors :
1,2M. R. Asato, 2K. E. Garver, 2C. F. Geier, 2M. C. Costello, 2K. S. Scherf, and 2B. Luna

Previous studies of brain function in epilepsy patients using fMRI have focused on language and memory localization. Given the common involvement of MTL in epilepsy and emerging evidence of a more global pattern of brain dysfunction extending beyond the putative seizure focus in TLE patients, we were interested in assessing networks of function using a learning task known to tap into a widely distributed circuitry including the MTL.
Animal single cell and human lesion studies using sequential learning tasks have suggested the involvement of the medial temporal lobe. Assessment of MTL function in more complex cognitive skills in surgical candidates may be of clinical value. Furthermore, using tasks requiring widely distributed functions may offer insights into effects of epilepsy on cognitive function. We describe findings from a sequential learning experiment using fMRI in healthy adults. We studied 32 healthy college students (18-34 years) recruited from the University of Pittsburgh who completed IQ testing and visuomotor testing using eyetracking. We obtained whole brain fMRI scans on 12 subjects. Both block and fast event related designs were used to distinguish networks of activation in explicit learning, and differentiation between different stages of learning. The block design compared random and sequence eye movements. In the event related design, subjects were visually prompted to make sequences of eye movements and then asked to generate this sequence on their own. The sequence task was presented three times so as to distinguish between early and late learning. Eye movement performance was obtained in the scanner. Behavioral testing revealed that learning across different stages demonstrated increased skill and accuracy. Results from the block design studies confirmed contributions of prefrontal, posterior parietal, middle temporal and cerebellar regions to learning. While parietal regions showed increased activation during learning in comparison to oculomotor control, the middle temporal gyrus, insula, angular gyrus and prefrontal cortex were specific to sequence learning. Event related studies demonstrated differences between encoding a novel sequence and executing a learned sequence. During encoding the frontal and parietal eye fields were predominant. During execution of a learned sequence, activation was uniquely present bilaterally in the parahippocampal gyrus, caudate, lateral cerebellum, dorsomedial thalamus, and anterior cingulate region. These results indicate a transition in the brain network supporting early and later learning including medial temporal lobe, posterior parietal, cerebellum, and subcortical regions. Such localization of function suggests that engagement of the MTL and associated networks may be altered in sequence learning in TLE patients. (Supported by MH01727, MH 18951, MR Research Center of the University of Pittsburgh.)