Timeout issue - Free feeding mode

[Hamid Taghipour]

Hello!

I realized that since the most recent library update v 1.16.3 is released, we are facing an issue with FEDs and that is a long delay for pellet delivery. Once the FED releases the pellet it goes to a Timeout period which lasts 5 sec during which FED is in an absolute refractory period, once the Timeout is over, it takes another 5 sec to release the next pellet! so it takes more than 10 sec on free feeding mode to have a new pellet in the bin (pellet well). 
I believe It may interfere with the microstructure of feeding behaviour.

Has anybody experienced a similar issue?

Best.

Hamid Taghipourbibalan

Ph.D. Fellow in Behavioral Neuroscience, Mccutcheonlab, UiT, Tromsø, Norway

[Yuhui Sunny Luo]

Hi Hamid,

I also updated to v1.16.3 and use free feeding mode. The time out is 5 seconds, then about 3 seconds later the motor will start turning. For most times, one motor turn is enough to send a pellet down, so it takes ~8 seconds for a new pellet to be delivered after one is removed. In our lab, before updating to v1.16.3, the interpellet interval plot has one density peak at ~20 seconds, but almost zero density at ~10 seconds. So I think it’s probably fine for the dispensing time to be ~8 seconds. 

Best,
Sunny

--
You received this message because you are subscribed to the Google Groups "FEDforum" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fedforum+u...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/fedforum/7ef8463b-e292-4d93-a85a-5d7f8291de6dn%40googlegroups.com.

[Hamid Taghipour]

Hi,

Thanks for your reply,

By the way, I am pretty sure that after the 5-sec Timeout, it takes another 5 sec before the dispensing process starts, so a 10-sec gap plus a time gap of 1-2 sec which is required for delivering the pellet to the bin depending on how much food is available in the hooper. It may interfere with the patterns of intake microstructure, especially if the animal is hungry and eats quicker than usual, I have seen them restlessly looking at FEDs waiting for another pellet! and I believe it is in contradiction with the ad lib scheme of food availability in free feeding mode.

Best.

Hamid Taghipourbibalan

Ph.D. Fellow in Behavioral Neuroscience, Mccutcheonlab, UiT, Tromsø, Norway

Contact:+4746566050

Skype ID: htbibalan_1

LINKEDIN

[Lex Kravitz]

Hi all,

Thanks for bringing this to my attention!   You've identified an issue with the free-feeding programs where the Timeout is 5s longer than requested.  This issue only applies to the free feeding task and I have a quick fix below.  I'll integrate a permanent fix for this in the next library update. 

The reason the timeout is an extra 5s long in the free-feeding task is that after each code loop FED3 goes to sleep.  FED3 wakes up when either 1) a nosepoke happens; or 2) 5s elapses.  So in nosepoking tasks FED3 wakes up when any nosepoke happens so this 5s sleep period does not cause any delays.  But with the free-feeding task the nosepokes aren't monitored, so FED3 will not wake up until that 5s delay is over.  Once it does, FED3 realizes the well is empty and dispenses the next pellet.  This results in your "timeout" being 5s too long.  The attached modified free-feeding sketch has a function to skip this sleep period after a pellet dispenses, so your timeout will be the correct length.  If you want to integrate this into other free-feeding variants you can copy the controlSleep() function from this to your code and simply call it once per loop.  

What this extra 5s delay means for mice is harder to answer conclusively. I agree with Sunny (and our experience with LOTS of inter-pellet-interval plots) that mice do not consume pellets quicker than 10s apart.  It's possible they can take a pellet, drop it, and take another one within 10s, but not consume 2 pellets in 10s.   Thanks again, and please keep reporting anything weird or potential bugs!  

-Lex