Breaking point in Closed Economy (PR1) paradigm?

[Benoît Delignat-Lavaud]

Hello FED3 users,

I wanted to know if it's possible to determine the breaking point of food consumption during a Closed Economy paradigm ?

I used a PR1 protocol for this and the program is supposed to restart as a FR1 after 30 minutes of inactivity; but I wonder if by setting a break length shorter than that we can still evaluate the breaking point of the animals and use it as a measure of their motivation.

I use FEDviz for my analysis and apparently by setting a break lenth of 15 or 30 minutes the GUI doesn't allow me to create the plot (either breakpoint plot or group breakpoint plot)... 

Thanks in advance.

Benoît Delignat-Lavaud

Laboratoire Dr. Louis-Eric Trudeau,

Département de pharmacologie et physiologie, département de neurosciences

Local S-405, pavillon Roger-Gaudry

Université de Montréal
2900 Boulevard Édouard-Montpetit

Montréal, Québec, H4T 1J4

 

ᐧ

Garanti sans virus. www.avast.com

[Lex Kravitz]

Hi Benoît, I think you are bringing up a deep question: how should the break point be interpreted in a traditional progressive ratio session?  One of the interesting things the Closed Economy paradigm reveals is that the break point is highly dependent on the time of day.  So there isn't really just one "break point", but rather a continuously varying parameter that changes across the circadian cycle.  If you want to boil this behavior down to one number for statistical comparisons I would suggest you take the maximum break point, or the total work performed per day (total pokes).  Does this answer your first question?

Regarding your second question, you can certainly set the reset period to shorter than 30 minutes, you can change resetInterval at the top of the script to do this.  I haven't messed with this, I based the 30 minute reset on this Mourra et al paper.  I presume you will get more resets that way and will therefore see a different pattern of behavior, but hopefully if you are looking at a group of mice that are more/less thrifty it won't depend on this reset interval to see this.

FED3VIZ was written before we wrote the PR1 program, and the developer (Tom Earnest) is no longer in my lab - he went on to grad school.  Unfortunately I do not have anyone in the lab at the moment who can continue developing FED3VIZ so it's sort of stuck in the state it is for now. 

[Benoît Delignat-Lavaud]

Hello Lex,

You made a very good point regarding the variation of behaviors across the circadian cycle. I am not an expert in chronobiology but by looking at my results in ClosedEc it is pretty clear that their "motivation" for food is maximal at the beginning of the night.

I have a mouse model (black in the fig below, WT in green) in which phasic dopamine release is almost abrogated but performed well at the task (PR1), and seems to exhibit higher motivation for food, but at night #1 only (AUC analysis). I was curious to see if their motivation could be evaluated in this paradigm using a "breaking point approach" (hence my try to plot the data for this using FED3viz). 

But maybe looking at the total work for each day instead of breaking points would work better in my case indeed.

Best.

--
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/9c841e25-3795-437d-933a-6d782e1a95f2n%40googlegroups.com.

--

Benoît Delignat-Lavaud, candidat au doctorat

[Lex Kravitz]

That's very cool data!  I like how the mutant becomes more efficient at the PR1 task over the 6 days.  We've seen that before, some mice will poke a lot on the first couple of days and once they realize the ratio resets they start acting more thrifty/efficient (ie: fewer pokes to get the same number of pellets). Thanks for sharing!

[Ahsan Habib]

I was wondering what are the ways to normalize either nose pokes or earned pellets data where the animals do not have breakpoint or have multiple breakpoints or no breakpoints. Is there any ways to compare them by minimizing those factors which might biased your data?

Best

Ahsan

[Lex]

Hi!  I'm sorry but I'm not certain what you're asking. Would you be able to share any data or code that illustrates your question?  Thanks!  

To view this discussion on the web visit https://groups.google.com/d/msgid/fedforum/8ba741e2-68e7-42c4-8bda-3651c75891d8n%40googlegroups.com.

[Ahsan Habib]

Hi Lex and everyone,

Thank you for your response. I did use the FED3 device at home cage to study the motivation/efforts in food restricted (overnight food withdrawal) animals. Mice were trained in different fixed ratio (FR1, FR3, and FR5) schedules to receive food pellets. Then, we tested the animals in progressive ratio (PR3 and PR5) tasks for 5-8 sessions. The reset time was 8 -minutes and each session consisted of 2-hr. We only find break-point in 30-35% of the animals in both groups and some animals have multiple breakpoints. Almost 65-70% of animals do not have breakpoints. We cannot compare the total workout since the number of nose-pokes will be lower in those that have breakpoints compared to those that do not have breakpoints. The total number of pellets earned did not differ much in the two groups. I would like to normalize the data to see whether there is a difference in motivation or efforts in two groups.

Thanks

Ahsan

[Lex Kravitz]

You're welcome!  As your sessions are only 2 hours long, it makes sense that most of your food restricted mice never reached their first breakpoint.  We often run this task over multiple days, so all mice get multiple breakpoints. See Figure 7 in the FED3 paper.  

However, I think you can still generate some meaningful comparisons from mice without breakpoints.  I would look at the total pokes divided by the total pellets (pokes/pellet).  This will give you an idea of the "cost" of each poke (to use terminology from this paper from the Beeler lab), which I think is a good operational proxy for effort.  You can also look at the impact of increasing the cost (PR) on pellets earned by individual mice, to look at cost sensitivity of individual mice.

I hope this helps!
-Lex

[Paige Bensing]

Hi all,

I am reopening this thread because I have a few questions re: the built-in PR program and potential use for determining breakpoint. The first message here seems to indicate that the PR session can automatically switch to a different FED mode (FR1) after a period of inactivity (in this case 30 min). I was wondering if this is part of the base code, or if this is a modification that this author made themself? If this is standard, are there ways to adjust the period of inactivity (say to 15 minutes instead of 30) and to change the mode it switches to (i.e. to FR3 instead of FR1)? If the program cannot automatically switch modes, is there a way to create some kind of "notification" when the breakpoint is reached? An idea we had was if the device lights turn on after breakpoint is reached so we can manually reset them. In the best case scenario, we would also be able to set a maximum session length (e.g. 3 hours), where the device switches modes or "notifies" after this period of time.

Please advise and thank you in advance!

Paige Bensing

[Taaseen Rahman]

Hi Paige,

You can adjust the reset interval duration by changing this line of code from 1800 to 900

int resetInterval = 1800;                              //number of seconds without a poke to reset

To change it from FR1 to FR3, change this line of code from 1 to 3:

int pokes_required = 1;                                //current FR

Not sure how to help with the other aspects though. Hope this helps.

To view this discussion visit https://groups.google.com/d/msgid/fedforum/776c0faa-19a1-4d3f-9f48-7156badafc4bn%40googlegroups.com.

[Zane Andrews]

Hi Paige

If you look at line 23 in the ClosedEconomy_PR1 code in Arduino (from the FED3 library) you will notice the reset interval…you can adjust as you please.

int resetInterval = 1800; //number of seconds without a poke to reset

There might be some slight confusion about how the task works - I read Lex’s description and maybe its not that clear 😂, sorry Lex.

At the start, mice need to do 1 active poke to get a pellet….then for each subsequent pellet mice need to do 2 more active pokes until the reset interval is met 

For example, pokes required to get a pellet start at 1 then is n+2……ie 1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25, 27, 29 etc etc if mice dont do 29 pokes in 1800 secs (reset interval) then the next pellet only needs 1 active poke then 3, 5, 7 etc etc…..and the pattern continues. Whether you call this FR +2 or PR +2 or FR1, FR3, FR5, FR7, FR9, FR11, FR13, FR15, FR17, FR19, FR21, FR23, FR25, FR27, FR29 doesnt really matter - above is how it works 

Also, once the reset interval is met the FED automatically goes back to 1 active poke for 1 pellet……until the resetinterval is met again…..and it just keeps repeating this for a as long you run the experiment. 

Im not sure about the experiment you want to run - could you not just run a progressive ratio for 3 hours? Closed economy is addressing the balance between an animals motivational drive versus its response inhibition. Its assessing whether (or how long) mice can learn to flexibly adapt their the feeding behaviour to optimise pellet collection, ie after time mice should learn that its easier to get 10 pellets (pokes required = 1, 3, 5, 7, 9, 11, 13, 15, 17, 19) and then wait 30 mins and start again from 1. This way they get more pellets than just incessantly poking  - it becomes a more economical decision to collect some pellets and wait for the reset rather than just poking away. 

We typical find you need to run this task continuously for days to really see the changes in feeding flexibility…..You’ll see learning with time were the first block of pokes before a reset is longer than the last…

Hope that helps.

We have python GUI on Github  - https://github.com/Andrews-Lab/FED3_time_bins that will allow you to analyse your Closed_Economy data……here are the installation guides (they need updating on GitHub)…