Color lab

In the colour laboratory, we have four different ways of presenting colour (fig. 1). These include coloured surfaces and materials on the one hand and coloured lighting on the other.

The dark side of the colour lab

At operating times, however, the colour laboratory is almost completely colourless (Fig. 2). This is because during a colour experiment, it is precisely controlled which light signals reach the eye of the observer. This is especially important because colours are always perceived relative to their surroundings. In order to exclude disturbing light sources, the walls in the colour laboratory are painted black and the windows can be closed light-proof. Furthermore, the computer screens are equipped with black-lined viewing tunnels through which the test participants look at the screen. Finally, the distance to the screen is controlled by chin rests. This is because the perceived size of the images presented on the screen changes with the distance. And this size also has an influence on the colours perceived due to the structure of the retina, i.e. the light-sensitive receptors in the eye.

Experiments on the computer

The monitors form the basis of many examinations in the colour laboratory (Fig. 2). They are used to present coloured stimuli in a scientifically controlled manner. To ensure this scientific control, special equipment as well as special calibration procedures are used. The colours displayed on a monitor are determined by setting three colour values, R, G and B. However, these RGB values are specific. However, these RGB values are specific to a monitor. This means that the same RGB settings on one monitor will produce different colours on another monitor. Therefore, for the purpose of calibration, both the exact chromaticity of the three values (R, G and B) and the brightness contributions of the individual levels of these three values are measured. Therefore, the spectrometer and the photometer are necessary accessories of the colour laboratory. With the spectrometer, one can measure the exact wavelength composition of the light. In this way, the chromaticity of the monitor and especially of the individual R, G and B light sources can be determined. The photometer measures with high accuracy the brightness of the colours displayed on the screen. It can be used to determine the brightness contributions of the individual levels of the RGB values. With the help of this information, it is possible to calculate which RGB values must be used for a particular screen in order to display a precisely defined colour on the screen. However, these calibration measurements are only useful if the screen in question emits light signals as uniformly as possible over the entire screen surface and, moreover, its performance remains as stable as possible over the time of the experiments. For this reason, the colour laboratory is equipped with the highest quality CRT monitors and corresponding graphics cards to control them. For experiments involving particularly small colour differences, additional equipment specially developed for colour research is used to increase the colour resolution. Each of the three RGB light sources can then display up to 16384 (14 bit) different levels instead of 256 (8 bit). In this way, the colour laboratory determines, for example, exactly those colour differences that are just perceived by the observer (so-called discrimination thresholds). Depending on the experiment, the colour lab also regularly uses, for example, a high-quality camera or special input devices, which are usually built by our technician. The latter include, for example, a high-precision, analogue reaction time measuring device or also an illuminated input device.

Example publications with experiments on the computer are:

Hansen, T., & Gegenfurtner, K. R. (2013). Higher order color mechanisms: evidence from noise-masking experiments in cone contrast space. Journal of Vision, 13(1). doi:10.1167/13.1.26

Hansen, T., & Gegenfurtner, K. R. (2017). Color contributes to object-contour perception in natural scenes. Journal of Vision, 17(3), 14-14. doi:10.1167/17.3.14
Witzel, C., & Gegenfurtner, K. R. (2018). Are red, yellow, green, and blue perceptual categories? Vision Research, in press. doi:https://doi.org/10.1016/j.visres.2018.04.002

Control of lighting colours

In addition to computer experiments, there is also the possibility in the colour laboratory to specifically measure and control the illumination of materials (Fig. 1). With a spectroradiometer we can precisely measure the spectra of the incident and reflected light. A hyperspectral camera even allows us to take pictures with spectral information. This means that information about the spectral composition of the incident light is stored at each pixel.

Example publications with this experimental setup are:

Ennis, R., Toscani, M., & Gegenfurtner, K. R. (2017). Seeing lightness in the dark. Current Biology, 27(12), R586-R588.

Ennis, R., Schiller, F., Toscani, M., & Gegenfurtner, K. R. (2018). Hyperspectral database of fruits and vegetables. Journal of the Optical Society of America A, 35(4), B256-B266.

DThe "room within the room"

For some experiments, we had set up a large cube accessible from one side - a room within a room, so to speak (Fig. 3). This room was used to control stimulation (and adaptation) across the entire visual field. Opposite the opening of the room, a screen is embedded in the wall. The opening of the room is equipped with input devices so that the test participant can work on the experimental tasks from here. To the left and right of the opening, invisible to the participant, are fluorescent tubes. These can be controlled by the computer and calibrated by the procedure described above. In this way, this small room can be illuminated in a controlled manner. In studies on the memory colour effect, for example, the test subjects have to adjust the colours of objects. The room must be illuminated in the exact shade that corresponds to the background of the monitor, because there should be as little difference as possible between the monitor and the wall. With the help of this experimental set-up, it could be shown that an object that normally has a certain colour can still be seen shimmering in this colour even if it is objectively colourless, i.e. grey.

Example publications with this experimental setup are:

Hansen, T., Olkkonen, M., Walter, S., & Gegenfurtner, K. R. (2006). Memory modulates color appearance. Nature Neuroscience, 9(11), 1367-1368.

Hansen, T., Walter, S., & Gegenfurtner, K. R. (2007). Effects of spatial and temporal context on color categories and color constancy. Journal of Vision, 7(4), 1-15.

The outsourced colour laboratory

Sometimes, however, the colour lab is also "outsourced". This is the case when colour perception is to be studied in an everyday environment. For example, the windows of the colloquium room were equipped with special colour filters to measure the stability of colour categorisation under different lighting (Fig. 4). In the process, the test subjects had to assign hundreds of coloured chips to the basic colour categories, such as red, green, purple, etc., under the respective illumination. At times, you could really see the world in pink as soon as you entered the colloquium room. But beware! After a short time, staying in this environment leads to adaptation. This means that the eyes get used to the rosy lighting. Then the pink light appears completely normal and colourless. Instead, you see everything in green as soon as you leave the room again. This lasts until you have adapted to the normal lighting again.

Example publications with realistic experimental set-ups are:

Bloj, M., Weiss, D., & Gegenfurtner, K. R. (2016). Bias effects of short- and long-term color memory for unique objects. Journal of the Optical Society of America. A, Optics, image science, and vision, 33(4), 492-500. doi:10.1364/JOSAA.33.000492

Olkkonen, M., Witzel, C., Hansen, T., & Gegenfurtner, K. R. (2010). Categorical color constancy for real surfaces. Journal of Vision, 10(9), 1-22. doi:10.9.16 [pii] 10.1167/10.9.16

_{Fig. 1: The coloured materials of the Farblaor. In the background on the right, you can see an apparatus for controlled illumination, here with green light. On the left, under a different illumination, one can compare colours with those in the illumination box, such as in experiments on colour constancy. In the foreground you can see a dress, as seen in the famous photo #theDress.}

_{Fig. 2: A typical experimental set-up for an experiment on colour perception. Almost everything is black. Only in the background can one see calibrated colours presented on the screen. In the foreground, one can faintly make out a chin rest and, to the side, the viewing tunnel. In this example, colour distributions are shown on the screen to investigate the cortical mechanisms of colour perception (Hansen & Gegenfurtner, 2013).}

_{Fig. 3: The "room within a room". The subject looks into an illuminated chamber. On the opposite wall, a screen is embedded on which the images for the experiment are displayed. The lighting of this chamber is controlled by the computer so that the screen and the lighting can be coordinated. In the foreground, you can again see a chin rest as well as the input devices needed to complete the task at hand. In this example, a banana is presented whose colour can be adjusted by the test subjects via the keyboard to demonstrate the effect of memory colours (cf. Hansen et al., 2006).}

_{Fig. 4. The "outsourced colour lab". The windows of the colloquium room are fitted with pink foils (in the background on the left) so that the room is illuminated in pink daylight (see Olkkonen et al., 2010). The camera is set in such a way that it does not compensate for the lighting completely automatically. As a result, a slight pink shimmer is seen, e.g. on the white wall in the background. In the foreground are standardised colour plates, which were to be assigned to different colour categories by the participants in the experiment. A white surface is shown on the monitor. Since the monitor is a light source, it emits a white light unaffected by the surrounding lighting. This appears greenish to the observer here and not white at all. Instead, the pink illuminated sheet of paper appears white. The reason for this is that the observer (here the camera) adapts to the pink illumination.}

For more information on colour perception and current research results, please click here.

Navigation

Direct Links