Coding Effort Discounting Task

23 views
Skip to first unread message

Rafael Cazuza (Cazuza)

unread,
Apr 23, 2026, 1:24:47 PMApr 23
to FEDforum
Hi FED community!
I'm thinking of using the FEDs for effort discounting task (left poke = FR1 > 1 reward/ right poke = FR10 > 5 rewards).
Has anyone ever tried this before? I have many questions about it.

1)The main one is how did you program the disc to stop rotating unless all the pellets from previous "hard task-high reward" has been retrieved to not jam the device.
2) How to program the device to dispense exactly 5 pellets if the inhibition signal to stop delivering is the sensor block at the dispenser?
3) How to analyze the data? Would we just compare total number of pokes each side?

If nobody has ever done I can try a couple different strategies and post it later.

All the Best Y'all

Rafael

Alex Legaria

unread,
Apr 23, 2026, 4:21:37 PMApr 23
to FEDforum
Hi Rafael,

While I haven't tried that specific task, I have run paradigms where more than one pellet is dispensed. I think the answer for questions 1 and 2 is pretty straightforward: Just repeat the fed3.Feed() function as many times as desired. To make it prettier you can put it inside a for loop. E.g:

for (int i=0; i<5 ; i++) {
    fed3.Feed();
}

that should solve all the servo issues and avoids messing with the source code.

As for question 3. It depends on what your scientific question is, but I'd say a good place to start is number of pellets, pokes, trials completed. My guess is that relative preference for these metrics will be more useful e.g: pokes_hard/total_pokes.

Good luck,

Alex Legaria
Reply all
Reply to author
Forward
0 new messages