The article discusses the challenges in neuroimaging research, particularly in functional magnetic resonance imaging (fMRI), and proposes solutions to improve transparency and reproducibility. It highlights issues such as low statistical power, flexibility in data analysis, software errors, and insufficient study reporting. The authors emphasize the need for pre-registration of methods and analysis plans, proper statistical power analysis, and clear reporting of study details. They also stress the importance of independent replications and the use of standardized data formats. The article suggests that neuroimaging research should adopt practices similar to those in genetics, such as large-scale collaborations, rigorous statistical methods, and open data sharing. The authors argue that current neuroimaging studies often lack sufficient power to detect meaningful effects and that the field needs to address these issues to ensure reliable and reproducible results. They propose a future for neuroimaging research that includes detailed planning, automated analysis workflows, and open sharing of data and code to enhance transparency and reproducibility. The article concludes that the neuroimaging community must take steps to improve the reliability of its findings and ensure that public funds are used effectively to advance understanding of the human brain.The article discusses the challenges in neuroimaging research, particularly in functional magnetic resonance imaging (fMRI), and proposes solutions to improve transparency and reproducibility. It highlights issues such as low statistical power, flexibility in data analysis, software errors, and insufficient study reporting. The authors emphasize the need for pre-registration of methods and analysis plans, proper statistical power analysis, and clear reporting of study details. They also stress the importance of independent replications and the use of standardized data formats. The article suggests that neuroimaging research should adopt practices similar to those in genetics, such as large-scale collaborations, rigorous statistical methods, and open data sharing. The authors argue that current neuroimaging studies often lack sufficient power to detect meaningful effects and that the field needs to address these issues to ensure reliable and reproducible results. They propose a future for neuroimaging research that includes detailed planning, automated analysis workflows, and open sharing of data and code to enhance transparency and reproducibility. The article concludes that the neuroimaging community must take steps to improve the reliability of its findings and ensure that public funds are used effectively to advance understanding of the human brain.