Balancing sample size and scan duration in functional MRI studies: A reflection on Ooi et al.

In the era of ‘big data’, there is a push for neuroimaging datasets with more subjects and/or longer scans. At the same time, funding pressures interact with the public perception of science to demand increased transparency and return on investment for research spending. In a recent paper, Ooi et al. developed a timely cost-benefit analysis to address this tension between cost-effectiveness and big data. The main takeaways from the Ooi et al. paper are that sample size and scan duration are interchangeable and for many studies the most cost-effective approach is to ensure that functional magnetic resonance imaging (fMRI) scans are at least 30 minutes long. This commentary discusses when and why increased scan duration is helpful, and how the balance between sample size and scan duration interacts with broader goals of external validation, representation, and individual differences in brain-behavior modeling.