LED lights as cues to initiate trials?

[Thomas Huff]

Hi all,

I've recently been implementing some new probabilistic bandit tasks in the FED3 -- basically different versions of a "restless bandit" task where the probabilities of the two poke holes very gradually change over time. 

Ideally, I would like to incorporate a light cue into the task to signal when a trial starts and ends. Something similar to what the researchers did in this 2024 paper in neuropsychopharmacology: 

Following that example, I've been attempting to use the two FED3 nosepoke holes LEDs as the cue. When the nosepoke lights illuminate, the machine is active and the animal is cued to engage. When the lights turn off (e.g., after a choice or timeout for non-engagement), the machine becomes inactive, thus requiring the animal to wait until the LEDs light up again to start the next trial.

I was just curious if anyone has implemented anything similar using the FED3. If so, I would love to learn more about any initial training protocols you used to teach the animals to respond to the light cues as an engagement signal (before moving on to the actual task). 

All the best,

Tom