Readout of behavior

[emirvine]

The traditional means of getting a readout of what the rat understands from the task is by offering probe trials, where the barriers are lifted and the rat has the choice of feeders.

This mostly poses technical challenges:
(1) Raising and lowering all 5 barriers within a short timeframe?
(2) Accounting for that traversal in subsequent trials so the proportion of experiences is not disturbed.

[emirvine]

To address (1) I think we could offer a probe trial as the first trial in each running phase and move the rat to the pedestal once he completes the trial to introduce the barriers. We could do this again halfway through the running phase by placing the rat on the pedestal, removing the barriers, allowing him to complete the probe trial before moving him back to the pedestal to replace the barriers.

It would be time consuming and not very elegant, but may accomplish our desired outcome of having a better sense of what the rat understands from the task.

[mvdm]

AFAIK Travis dealt with this by only ever offering the rat a choice between 2 arms. Could that work here, too?