Functional magnetic resonance imaging of language processing and its pharmacological modulation

by 1974- Tivarus, Madalina E

Functional Magnetic Resonance Imaging (fMRI), a non-invasive technique extensively employed for mapping brain function, was used to examine brain activation during language processing and the effect of dopaminergic agents on brain hemodynamics and language. Administration of Levo-Dopa, a drug known to increase levels of dopamine in the brain and used to treat Parkinson’s disease, has been associated in behavioral studies with restriction of the semantic network. The current project attempts to identify how these behavioral findings are reflected in brain activation as determined with fMRI. These investigations are summarized in the following. In the first part of the project, the goal was to determine the effect of L-Dopa on brain hemodynamics. This was motivated by the fact that the fMRI signal is based on cerebral blood flow, oxygenation and cerebral blood volume changes, and any drug administration could interfere with the coupling of neural activation with these parameters, independent of neuronal activity. In order to obtain information about possible global changes in cerebral blood flow (CBF) due to drug administration, a ii theoretical model of a relationship between the blood oxygenation level dependent (BOLD) signal and CBF was used. The BOLD signal was measured in two brain regions, motor and visual cortices, in two treatment conditions, placebo and L-Dopa, for each participant. Within subject comparisons revealed no significant differences between the measured BOLD signal in the two conditions, and the calculated changes in baseline CBF were less than 1% for both motor and visual cortices. We concluded that the changes in global CBF due to the drug administration were not significant and as a consequence, this was not used as a covariate in subsequent studies of drug effect on specific cognitive functions. The second part of the project examined the effect of semantic priming on brain activation, and the modulatory effect of dopamine on this type of language processing. The aims of this part of the project were: to implement a protocol for language function imaging, to explore different types of paradigm design in fMRI, and finally to examine the effect of semantic priming on brain activation and the effect of L-Dopa. Behavioral measurements were recorded and demonstrated a significant priming effect for all semantic distances. Imaging results showed activation in a cerebral network known to be involved in language processing and attention to task, similar to other studies previously reported in literature, thus confirming the successful implementation of our language imaging protocol. Two types of paradigm design, block and event related, were explored and the results were compared. They revealed different patterns of activation emphasizing the importance of careful selection of design in functional neuroimaging studies of cognitive functions. For the pharmacological part of the study a within subject design was used, each participant undergoing two scanning sessions, after ingesting L- iii Dopa and placebo. During each testing session, each participant performed two scan runs, targeted to examine automatic versus controlled processing. This way we were able to study both the effect of temporal characteristics of the priming process and the effect of L-Dopa on brain activation. No drug or temporal effects were found on the activation maps, suggesting that more sensitive techniques must be used to detect these changes. Our findings support the network model for the organization of semantic lexicon. Lastly, fMRI was used to study functional connectivity associated with semantic and phonological processing. The goal was to explore the interaction between language network components and to determine if administration of L-dopa would affect this interaction or these types of language processes. During two test sessions (placebo and L- Dopa) each participant performed two fMRI runs, involving phonological and semantic processing of visually presented words. These were used to generate maps of semantic and phonological networks. Activation patterns for the two language processes were obtained and compared to previous findings. A number of regions of interest (ROI) commonly activated by the two tasks, were chosen based on these activation maps and the functional connectivity was calculated as the degree of correlation between the activation time series data of two brain areas. The functional connectivity analysis revealed that language areas were activated in a more synchronous manner (i.e. higher correlation coefficients) for phonological tasks than for semantic tasks. No drug effect was found on either the activation maps or the functional connectivity results. Our findings could be of significance for patient populations showing atypical levels of dopamine in their brain, such as cocaine withdrawal, Parkinson’s disease or schizophrenic patients. iv