Measuring neural representations
Do you want to know more about how to use MVPA to address the activation level of specific neural representations? 🧠 Check our new paper on Canonical Template Tracking (CTT) in Frontiers in Neuroimaging.
In this work, we described CTT: a multivariate approach that estimates canonical neural representations (or templates) from localizer tasks, to later track them in independent paradigms’ data using either RSA similarity measurements or machine learning classifiers.
Despite the previous use of CTT in neuroimaging research and its theoretical relevance for some cognitive domains, no methodological guidelines were yet available in the literature. We aimed to fill this gap!
In our paper, you will find recommendations on how to design localizer tasks from a multivariate perspective, a step-by-step tutorial on how to implement CTT analyses on spatial and temporally resolved datasets, and guidelines regarding results interpretation.
Do you want to give CTT a try? In the link below, you’ll find Matlab scripts to perform different versions of the analysis on fMRI data. Also, the functions can be easily integrated in The Decoding Toolbox flow (check the analysis template).
– GitHub repository with CTT code
Want to read the actual paper?
Coming soon