Reproducibility analysis in sensorimotor and language tasks: implications for clinical functional magnetic resonance imaging.

• Main Author: Law, Amos Li Xiong.
• Other Authors: School of Humanities and Social Sciences
• Format: Final Year Project
• Language: English
• Published: 2010
• Subjects: DRNTU::Social sciences::Psychology::Psychological testing
• Online Access: http://hdl.handle.net/10356/38767
• Institution: Nanyang Technological University

In clinical functional magnetic resonance imaging (fMRI), p-value thresholds are typically applied to activation maps to investigate activation intensity. Problems with this methodology include: 1) averaging conceals richness of data, 2) appropriate p-values, 3) emphasis on small p-values, and 4) moderately activated functional regions cannot surpass rigorous p-value thresholds. Similar challenges arise when applying clinical fMRI to evaluate a single patient, in addition to a lack of normative standards for comparison. We propose using reproducibility analysis (RA) with the thresholding method to provide an added dimension of information: reliability of a brain region’s activation over time. We found that sensorimotor tasks involving primary cortical areas produced more robust and reproducible activations than language tasks involving higher cognition and association cortex.