Trial-Level Regressor Modulation for Functional Magnetic Resonance Imaging Designs Requiring Strict Periodicity of...

Area of Science:

• Neuroimaging
• Cognitive Neuroscience
• Psychology

Background:

• Computer-based cognitive assessments often require fixed stimulus display times (SDT) and intertrial intervals (ITI).
• Invariant SDTs and ITIs in functional magnetic resonance imaging (fMRI) studies can lead to collinearity issues, complicating data analysis.
• Detecting brain activity related to cognitive processes like response readiness is crucial in fMRI research.

Purpose of the Study:

• To investigate regressor modulation using trial-level reaction times (RT) as a method to enhance signal detection in fMRI.
• To evaluate the impact of modulating the 'go' regressor on the detection of Blood-Oxygen-Level-Dependent (BOLD) signal change for the 'no-go' condition.
• To compare different RT-based modulation techniques (amplitude modulation and amplitude & duration modulation) for their effectiveness in reducing regressor collinearity.

Main Methods:

• Utilized a go/no-go task with invariant SDTs and ITIs within an fMRI setting.
• Modulated the 'go' regressor using a canonical hemodynamic response function (HRF), an RT-based amplitude-modulated (AM) HRF, and an RT-based amplitude and duration modulated (A&DM) HRF.
• Assessed the BOLD signal change for the 'no-go' stimuli under each modulation condition.

Main Results:

• Reaction time-based modulation successfully reduced collinearity between 'go' and 'no-go' regressors.
• The A&DM approach yielded the greatest reduction in regressor correlations.
• Increased reductions in collinearity correlated with longer mean RTs and greater RT variability.
• RT-based modulation significantly increased statistical power for detecting group-level 'no-go' BOLD signal change across numerous brain regions.

Conclusions:

• Regressor modulation with RT is an effective strategy for improving BOLD signal detection in fMRI studies with invariant stimulus timing.
• The A&DM HRF provides the most substantial benefit in reducing collinearity and enhancing statistical power.
• This technique is particularly valuable for fMRI research where fixed stimulus presentation is necessary, offering a robust solution to collinearity challenges.