Identification of multiple timescales of adaptation using Expectation Maximization (EM)
EM is a statistical algorithm that can estimate model parameters for a data set containing latent variables.
Here we provide code that uses EM to fit two state models of learning to sensorimotor data. This tool uncovers the hidden fast and slow states of learning from observed behavior.
We provide two packages, one that can be used to fit sensorimotor data with set breaks (Version 2.1), and one which assumes no set breaks occurred (Version 1.1).
To get started with the package, download the zip file and refer to the README file.