Mar 31, 2015

Improving effect size and power with multi-echo fMRI and its impact on understanding the neural systems supporting mentalizing

BioRxiv : the Preprint Server for Biology
Michael V. LombardoPrantik Kundu


Functional magnetic resonance imaging (fMRI) research is routinely criticized for being underpowered due to characteristically small sample sizes. Additionally, fMRI signals inherently possess various sources of non-BOLD noise that further hampers ability to detect subtle effects. Here we demonstrate that multi-echo fMRI data acquisition and denoising can increase effect size and statistical power for block-design experiments, allowing for novel insights by detecting effects that are typically obscured in small sample size/underpowered studies. Application of this method on two different tasks within the social cognitive domain of mentalizing/theory of mind demonstrates that effect sizes are enhanced at a median rate of 25-32% in regions canonically associated with mentalizing. For non-canonical cerebellar areas that have been largely less focused on by the field, effect sizes boosts were much more substantial in the range of 43-108%. These cerebellar areas are highly functionally connected at rest with neural systems typically associated with mentalizing and the resting state connectivity maps largely recapitulate the topology observed in activation maps for mentalizing. Power simulations show that boosts in effect size enable...Continue Reading

  • References
  • Citations


  • We're still populating references for this paper, please check back later.
  • References
  • Citations


  • This paper may not have been cited yet.

Mentioned in this Paper

Magnetic Resonance Imaging
Cortical Cell Layer of the Cerebellum
Research Design

About this Paper

Related Feeds

BioRxiv & MedRxiv Preprints

BioRxiv and MedRxiv are the preprint servers for biology and health sciences respectively, operated by Cold Spring Harbor Laboratory. Here are the latest preprint articles (which are not peer-reviewed) from BioRxiv and MedRxiv.