Running Morris Water Maze under red light
I recently had an interesting conversation at Imperial College in London. about what the best lighting is to use when running Morris Water Maze. It seems that while most workers are using indirect incandescent lighting to prevent glare in the pool and specular reflection, some are illuminating the pool with red light presumably to avoid giving lighting cues to the animal. Since in classic MWM the animal learns to locate the platform using visual cues placed around the room this might be problematic so I thought this was worth discussing.
I think some people had suggested that rats are "colorblind" so this shouldn't make any difference, but of course they are simply di-chromats like most mammals rather than tri-chromats like humans. [Dichromats can match any color they see with a mixture of no more than two pure spectral lights. By comparison, trichromats require three pure spectral lights to match all colors that they can perceive.] This is, of course, not relevant unless we are asking the rats to do a color matching task. The relevant factor is actually the sensitivity to red light. I remembered that hunters in the states use read torches to hunt deer on the basis that the animals "can't see red" and that crepuscular rodents are reported to have exceptionally good UV sensitivity (I think this was verbally imparted to me by the late T.R.E. Southwood). Armed with this information I thought I'd look into exactly what those cutoff were - and I quickly turned up "Cone-based vision of rats for ultraviolet and visible lights" by Gerald H. Jacobs, John A. Fenwick and Gary A. Williams 2001. As you can see
I think some people had suggested that rats are "colorblind" so this shouldn't make any difference, but of course they are simply di-chromats like most mammals rather than tri-chromats like humans. [Dichromats can match any color they see with a mixture of no more than two pure spectral lights. By comparison, trichromats require three pure spectral lights to match all colors that they can perceive.] This is, of course, not relevant unless we are asking the rats to do a color matching task. The relevant factor is actually the sensitivity to red light. I remembered that hunters in the states use read torches to hunt deer on the basis that the animals "can't see red" and that crepuscular rodents are reported to have exceptionally good UV sensitivity (I think this was verbally imparted to me by the late T.R.E. Southwood). Armed with this information I thought I'd look into exactly what those cutoff were - and I quickly turned up "Cone-based vision of rats for ultraviolet and visible lights" by Gerald H. Jacobs, John A. Fenwick and Gary A. Williams 2001. As you can see
with permission from Jerry Jacobs
The cut-off of the l-cone type for rats is about 650nm - i.e. for long wavelengths the rates are essentially 'in the dark'. As you can see that's much closer to the human m-cone ('green receptors').
In other words rats (and presumably mice) don't have the specialized red cone type that we do - so yes when the lightsource is longer than 650nm (but less than about 700nm) and dim we can see and they basically can't.
Looking at available light sources we can see that it's easy to produce such a red light from a filtered incandescent source (which is rich in red and infra-red) but less so with other common sources. The ideal here would be a long pass dichroic filter with a transmission step at about 650nm.
The quest obviously arises that if you do that what is the effect on behavior - and that's something that we'll address here shortly.
No comments