Function to compute thigmotaxis index

[niksirbi]

In rodent behavioural neuroscience, thigmotaxis (the tendency to stay close to the walls of an open arena) is a classic measure of anxiety-like behaviour in the open field test (OFT).

The thigmotaxis index (sometimes called thigmotaxis ratio or wall-hugging index) is usually quantified by comparing the animal’s time or distance spent in the periphery of the arena vs. the centre.

A common approach is:

1. Define "periphery" and "centre" zones. For example, in a 50x50cm circular open field arena the periphery may be a 10cm band along the exterior wall. Arena size matters: a larger arena makes centre entries rarer (because mice naturally avoid crossing large open spaces), so the periphery band is often proportionally defined (e.g. outer 20–25% of the arena width). For a more general approach in movement, we could ask the user to supply a PolygonOfInterest (the arena) or a LineOfInterest (the exterior wall) and a given distance from the wall.
2. Take the ratio of time spent in periphery vs total time. Sometime distance travelled is used instead of time spent, so maybe it should be an option?

Once we have something like this implemented in the kinematics module, we should write and example.
In the future we should also think of a convenient way to do this in napari as it's quite a common analysis.

[niksirbi]

Note

This issue is being offered as a potential project for a hackathon (https://github.com/neuroinformatics-unit/osw25-hackday), so please refrain from working on it till Aug 15th.