Functional MRI of Language in Aphasia: A Review of the Literature and the Methodological Challenges

Crosson, B., McGregor, K., Gopinath, K., Conway, T., Benjamin, M., Chang, Y., Moore, A., Raymer, A., Briggs, R., Sherod, M., Wierenga, C., and K. White

Neuropsychology Review (2007) 17: 157-177

The functional imaging (fMRI, PET, MEG) literature relating to neuroplasticity, or the changes in brain organization, during recovery from aphasia and during rehabilitation is reviewed. Some researchers emphasize the right-hemisphere as the substrate for recovery in aphasia, while other researchers advocate that language recovery is due to left-hemisphere activity since the right-hemisphere does not normally contribute to language functions. However, it has become apparent that both hemispheres most likely contribute to recovery. Additionally, the size and location of the lesions seem to determine participation of the right-hemisphere. For example, large lesions make right-hemisphere participation more likely, and small lesions generally lead to recovery supported by the left-hemisphere. Rather than studying whether the right- or left-hemisphere is responsible for aphasia recovery, the more appropriate question is: When does each hemisphere play an important role? Unfortunately, the studies involving functional imaging of neuroplasticity during aphasia treatment have relied on a small number of patients, making it impossible to draw definitive conclusions regarding the substrates of successful treatment.

Functiontal imaging of language in aphasia involves many challenges, espcially when using fMRI. The second part of this article outlines six such challenges:

· Selection of a baseline task: Which cognitive function should be imaged and will the task/baseline combinations make images sensitive to neural mechanisms in aphasia and during recovery?

· Structure of language production trials: Consider motion artifacts and anticipate response latency in trials.

· Mitigation of motion-related artifacts

· Use of stimulus onset versus response onset in fMRI analyses

· Correct responses and errors in analyses: Does the investigator include correct responses or all responses?

· Reliability and stability of fMRI images across sessions: Studies on rats and monkeys indicate differences in sessions to be a function of the variability in the underlying map, not a function of the reliability of the mapping technique.

Before fMRI can be used clinically to evaluate aphasia and aphasia rehabilitation, these challenges must be adequately addressed.