The brain is a visually biased computer,
and your eyes seldom tell you the truth,
especially when looking at a 2 colour merge image.
We are evolved to find bright coloured fruit
with high contrast against the green of leaves.
We are very bad at asessing quantitatively a whole grey scale range,
as our brains only really pick out very contrasty objects,
and we miss much information.
I'd like to reiterate the point that the human visual system is terrible at accurately estimating absolute values in data displayed with a color map. Psychophysical experiments have shown that simultaneous contrast effects in the gray scale map, in particular, can cause value estimation errors of up to 20% of the data value range displayed by the scale (Colin Ware, Color sequences for univariate maps: Theory, experiments, and principles, IEEE Computer Graphics and Applications, 8(5), pp. 41-49, 1988). See
http://web.mit.edu/persci/people/adelson/checkershadow_illusion.html for a good demonstration of simultaneous contrast. What the human visual system IS good at is detecting areas of high contrast, which lets us do things like detect outlines of objects so we can avoid them while walking. Unfortunately, for the purpose of data estimation from a color map, the same mechanisms that enhance edge detection prevent accurate absolute estimation of data values from the gray scale map.
I think colour merge images are a very bad idea,
as they are misleading for the reader.
Much better to get the reader to understand
what they see in a 2D hisotgram (its just like a FACS data presentation,
which everyone has no trouble understanding... at 2D dot plot if you like)
Here you get a much better visualisation of the corellation between 2 images
which is not intensity dependent (in the actual visualisation)
Absolutely. Moreover, showing merged color images where one of the chosen colors is blue is a particularly bad choice. Our sensitivity to contrast changes in blue is far less than to that of green or red, so using the blue channel will almost guarantee that any visual analysis, even qualitative, will be confounded.
For single channel images, I much preffer to use
at the very least grey scalem but much better
a rainbow colour look up table, where each intensity iod a different bright colour.
Our eyes and brain are much better at interpreting that than greyscale
where the lower intensitites just look like black, and the higher intensities
all look nearly white, and you cant tell the difference. With different colours it is much clearer.
The rainbow color map is almost always a poor choice for displaying single channel images, despite its ubiquity in the display of scientific data. While it is true that the gray scale map can cause up to 20% errors in value estimation, it is also true that the human visual system is much more sensitive to luminance changes than to changes in color. Therefore, if you want to convey the most information from your data, particularly information with a high spatial frequency, you need to use a color map with luminance perceptually varying linearly from dark to light. In the rainbow color map, luminance varies non-linearly along the spectrum. Furthermore, our sensitivity to differences in color varies along the spectrum (note the banding you see in a rainbow color map when it is displayed in a legend). The unfortunate consequence is that the rainbow color map introduces perceived sharp gradients in the data where there are none, and potentially reduces the perception of gradients where they exist. For some great examples of how this property can drastically change your interpretation of the data, see Borland and Taylor, Rainbow color map (still) considered harmful, IEEE Computer Graphics and Applications, 27(2), pp. 14-17, 2007.
For univariate data, your best bet is to use a color map that varies in hue while linearly varying in illumination as well. The blackbody radiation color map (black-red-orange-yellow-white) is a good example. Inclusion of hue variation will reduce the simultaneous contrast effects that cause severe value estimation errors.
Regardless of your color map choice, never rely on the human visual system for absolute value estimation!