Physiological noise often dominates the blood-oxygen level-dependent (BOLD) signal fluctuations in high-field functional MRI (fMRI) data. Therefore, to optimize fMRI protocols, it becomes crucial to investigate how physiological signal fluctuations impact various acquisition and reconstruction schemes at different acquisition speeds. In particular, further differences can arise between 2D and 3D fMRI acquisitions due to different encoding strategies, thereby impacting fMRI sensitivity in potentially significant ways.