Available in English, German, Spanish, French (see below for information on implementing the task in other languages).
The Chronometric Radial Fitts' Task (CRFT) Czilczer et al., 2025 is a behavioural paradigm designed to assess the ability to manipulate movement imagery, approximated via imagery duration. It specifically measures the extent to which the relationship between movement difficulty and duration (through varying target size according to Fitts' Law) is preserved in imagery compared to execution. The task was developed to address limitations associated with using the deviation between execution and imagery durations as an indicator of movement imagery ability.
If you are interested in assessing broader movement imagery ability, visit the Movement Imagery Ability Platform for an overview of open-source behavioural tasks.
Example imagery trial:
Example execution trial:
Refresh to replay animation.
This repository contains open-source materials for the most up-to-date versions of the CRFT. It links two folders representing PsychoPy (.psyexp) and OpenSesame (.osexp) experiments, together with the associated files required to run them locally (lab/desktop experiments).
Please consult the accompanying manuscript (Czilczer et al., 2026) on the Movement Imagery Ability Platform for guidance on the necessary steps to run the task locally in PsychoPy and OpenSesame. Furthermore, please read the version-specific README provided in the respective folders:
Subsequent updates to the native software (PsychoPy, OpenSesame) may require adjustments. As developers, we are not responsible for implementing these changes in every use case.
If you want to contribute to this repository by providing a language translation, or want to run the task in your own language, expansions can be done relatively easily thanks to the implementation of language localisations (please read each README to understand how to implement these). You can also see these demos showing how to implement a language localisation in PsychoPy and OpenSesame with virtually no code.