Statistical Parametric Mapping: An Open Science adventure.

In 1988 the first brain activation studies, relying on Positron Emission Tomography (PET) and the availability of short half-life radiotracers, were reported. Back then regional differences between brain images were simply characterised over hand-drawn region of interests. The idea of producing a "statistical parametric map" (SPM), i.e. providing valid inferences about brain signals over all the voxels, was formally introduced in 1990 by Friston et al. for PET images. The SPM framework was further extended over the years to handle functional and structural MRI, along with EEG and MEG data. Since its first instance in 1991, the SPM software is openly and freely available to the neuroimaging community, to promote collaboration and a common analysis scheme across laboratories, as well as to allow the methods to be closely scrutinised by others.

There have been eight further major releases of SPM and many intermediate updates over the last 32 years, the latest in January 2020. Each release represents ongoing theoretical advances and technical improvements to the preceding version, most notably functional MRI analysis, Voxel-Based Morphometry (VBM) and Dynamic Causal Modelling (DCM). Nowadays SPM remains the most used software for the analysis of neuroimaging data. Its openness lead to the development of a whole ecosystem where extensions (more than 60!) were developed to address other problems and data types.

Beyond the specific SPM software, Prof. Karl Friston and the SPM team also contribute to the dissemination of data analysis methods and good practice. The "SPM Short Course" has been taking place for almost 30 years at the Wellcome Centre for Human Neuroimaging, University College London, UK. All of the educational material (including published papers) is available online and videos of past courses can also be freely viewed. Numerous other courses continue to be regularly organised around the world.