Notes on a generative model for reset/inhibition hypothesis (marmoset paper)

[fusaroli]

1. Individuals have fixed intervals between calls
2. How to define distribution of intervals for individuals? are they the same btw individuals?
3. What about duration call? if it's roughly 0 (as in Ravignani) individuals do not overlap. How to define distribution of durations?
4. As individual B hears A's call they reset the interval. OR NOT. As individual B hears A's call they ignore it (in case of non-overlap) or simply avoid vocalizing (if overlap).

[fusaroli]

Additional complexity: some animals do not simply have regular intervals of calls, they do have regular intervals of length x between bouts of n calls, with an interval of length y between calls within a bout.

[fusaroli]

Given what we've done through the hackathon:

• make arousal, inhibition and non-interacting baseline in the same data format
• make functions, so we can just add parameter values and get the data
• test the analysis on these data

[fusaroli]

Further, we (hopefully me, RF) should also think through the generative function in Stan, to make a proper model for the data.