Training mice to eat from FED3

[Kathryn Brewer]

Hi Lex,

I was wondering what the best method is for training the mice to poke and eat from the FED3.

Thanks!

Kathryn

[Lex Kravitz]

Hi Kathryn,

Thanks for your question!  I have a quickstart video on starting up the device but guess I should make some videos that talk more about training mice too! 

To train mice to nosepoke from FED3, we first set the FED3 in "free feeding" mode (no pokes needed to get pellets), and place it in the cage with a single mouse with its regular chow still in the cage.  Usually one night of free feeding is enough to train mice to eat from FED3. They should earn ~75-100 pellets a night while regular chow is still available, so I just check the screen and confirm that they earn about this many pellets before moving to FR1.  If they earn <50 pellets their first night I'd give them another night with free feeding mode, but >90% of mice only need one night of this. 

Next we set the FED3 to FR1 mode and put it back in the cage, again with regular food still available.  They should figure the FR1 task out in one night too, meaning they should earn ~75-100 pellets with the FR1 schedule.  Once a mouse has done this they are trained that: 1) pellets come from the pellet well, and 2) a nose poke triggers a pellet delivery. 

You can then move into teaching them more complicated operant paradigms, and/or remove the regular chow and have them earn all of their calories from working with FED3, a situation known as a "closed economy".  One note on closed economy tasks: in our hands mice prefer the 20mg grain pellets to their colony chow.  I'm not sure why that is, but it is helpful in that you can train them to do operant tasks without removing their regular colony chow.  This means that mice won't starve if the FED3 device jams or if you try training them on a task that's too difficult for them to obtain their caloric need.  So whenever possible we train mice with regular chow still in the cage, and only use a closed economy situation when we need to account for calories or if the experimental question requires it.

Feel free to write back with any other questions,

-Lex

[Juan Matute]

Hi Lex!

Thank you very much for developing FED3 devices. My group is starting to use them. I am particularly interested in early-life phenotypes. At what age do you usually train your mice to use the FED3 device? What are the youngest mice you have trained? 

Best,

Juan

[Lex Kravitz]

Thanks Juan!  The earliest we have tried is ~6 weeks.  We don't ween our mice until 3-4 weeks of age and I think 3-4 week old mice would be too little to try FED3.  Best, -Lex