Ambient lighting for rodent behavioral tests

What are people’s opinions about controlling ambient lighting for behavioral testing? Our behavior rooms have standard fluorescent lighting. It’s pretty bright. Some areas of our rooms are guarded from the light by shelving so there can be gradients of light.
I’ve always thought about ambient light as a factor that should be controlled, but I don’t have much control over it. I just keep it constant between tests, so that at least I’m internally consistent in my own experiments.

But has anyone systematically studied the effect of different strengths of lighting on behavioral tests (nociceptive)? Would less light be superior?

@jmogil @tonellor @tberta @runDRG @rpseal @thicunha @TSheahan

I think the most practical solution is to try to keep things standardized, and report your conditions, like you’ve done. I haven’t seen anyone report on it, but since we know that testing during the light cycle vs dark cycle can impact evoked pain measurements, I wouldn’t be surprised if light intensity also had an effect.

Ideally, I’d do all my behavioral testing under red light during the dark cycle…


I haven’t studied it on nociception specifically/exactly but have had standard overhead lighting screw up behavior in two different labs. In my previous lab it was very noticeably affecting our morris water maze data. We stopped using the overhead lights and went to floor lamps. It was a huge pain in the ass to set them up such that there was equal lighting throughout the room but worth it. Nothing worse than after doing 12 days of water maze studying the tracks and seeing that all animals swim to a certain part of the tank because it’s least bright there. After a few years of begging they finally switched our rooms to a reverse light cycle which was wonderful.

In my current lab lighting became an issue when doing conditioned place preference (cpp) testing. Our pain model is chronic constriction injury of the infraorbital nerve. In these experiments during cpp we pair a chamber in the cpp apparatus with 30 minutes optogentic stimulation of the Sp5C and then 30 minutes in the other chamber with nothing. After 3 days of pairing we do a testing session where the animals can move freely between both chambers and measure time spent in each chamber.

We got reproducible results across a number of experiments and then didn’t do any cpp for close to a year.

Upon resuming cpp testing the animals always spent more time on one side of the apparatus. It didn’t matter which chamber that was or which chamber they received stimulation in. The only thing I came up with it that now there was a lighting difference between the chambers. I couldn’t get rid of it so back to no overhead room lights. Took more drastic measures too (probably overkill), all door cracks sealed so no outside light gets in, computer screens turned off during testing, and the only light used is a dual gooseneck microscope illuminator. We picked that because it’s the smallest light source we had where anymaze could track the animals and we can position it to provide equal lighting over both sides of the apparatus. 3 experiments have been done with this setup and results are back to what we previously had seen.

I will be using this setup when I do von Frey testing next and can add some ambient light groups in to that study and see if there any differences. Not sure exactly when I’ll be doing it but will report back with the results.


Thanks @jac
Awesome reply! I hope that effort is improving your experiments. For chamber-based assays, I too have noticed that the mice prefer one or the other chamber, and it seems to be related to lighting, even subtle differences. I think this is a factor we should look more closely at and report.

I look forward to hearing about your results with VF. Thanks for the great post!