The response time paradox in functional magnetic resonance imaging analyses

Area of Science:

• Cognitive Neuroscience
• Neuroimaging

Background:

• Response times (RTs) are crucial in cognitive psychology but often overlooked in functional MRI (fMRI) analyses.
• fMRI signal intensity is a proxy for neuronal activity, but it cannot distinguish between intensity and duration changes.
• Ignoring RTs can lead to misinterpretations of fMRI data, conflating signal intensity with neuronal duration.

Purpose of the Study:

• To investigate the confounding effects of ignoring RTs in fMRI analyses.
• To demonstrate how RTs can influence activation estimates and group-level findings.
• To introduce a novel time-series model for addressing RT confounds in fMRI.

Main Methods:

• Analysis of fMRI data, specifically examining the relationship between RTs and activation estimates.
• Simulation or statistical modeling to illustrate confounding effects at both individual and group levels.
• Development and proposal of a new time-series modeling approach.

Main Results:

• Longer RTs are associated with larger fMRI activation estimates, creating a direct confound.
• Ignoring RTs can lead to spurious group-level associations unrelated to experimental conditions.
• The proposed time-series model effectively accounts for RT variations.

Conclusions:

• RTs are a significant, often overlooked, factor in fMRI analysis that can distort findings.
• Failure to account for RTs can lead to incorrect conclusions about neuronal activity and group differences.
• Implementing RT-aware analytical methods, such as the proposed time-series model, is essential for accurate fMRI research.