Seeing Lightness In The Dark

These are data files for a manuscript about lightness perception under scotopic viewing conditions. The main finding was that observers do not perceive white during scotopic adaptation, implicating that the cones must be activated to produce the appearance of white. The publication will be published in Current Biology as a Correspondence under the title, "Seeing lightness in the dark," on June 19th, 2017. The data file contents are as follows:

chipBlocks.csv - This CSV file contains the reflectances of our paper stimuli, relative to the reflectance of our PR650 white reference. Those numbers are in the first column. The remaining columns are for plotting purposes (third column = y-coordinate of squares along bottom of Fig. 1A in the associated paper, remaining columns = RGB values for coloring those squares).

chipsVSA4.tsv - A TSV file containing the relevant ratings for Fig. 1B in the associated paper. The first column is the x-axis position for plotting, the second column is population mean white rating, and the third column is standard error of the mean for that rating. The first set of two rows are for peripheral viewing of the chips, the second set of two rows are for foveal viewing of the chips, and the third set of two rows are for ratings given to the larger A4 papers. The ratings are grouped in sets of two because each x-position is repeated twice: once for the rating of the darkest stimulus and a second time for the lightest stimulus. The gray values given to the data points in Fig. 1B were computed by taking the average white rating for that point and dividing it by 100, giving a value in the range of 0-1, which can be used as a normalized grayscale RGB value.

obsDataChips.csv - This is a CSV file containing each individual white rating (from 0%-100% in steps of 10%) that observers gave to the chips, with two trials per chip across four different lightness levels and at two different positions of fixation. In the final report, fixation position was only considered for the scotopic condition. A header is in the file labelling the columns.

fovealPopDataChips.csv - A CSV file containing white ratings of the chip stimuli when viewed with the fovea. First column is the reflectance of our paper stimuli, as in chipBlocks.csv. The remaining columns are the population mean and standard error of the mean (SEM) for the different paper stimuli under the different adaptation conditions, as follows: 2nd/3rd column = population mean and SEM for scotopic adaptation, 4th/5th column = population mean and SEM for mesopic adaptation, 6th/7th column = population mean and SEM for dim photopic adaptation, 8th/9th column = population mean and SEM for bright photopic adaptation.

obsDataA4.csv - Same as obsDataChips.csv, but for the A4 sized papers. Observers were not given explicit instructions about different fixation positions during this experiment, so fixation was always listed as "center" and this column was ignored in later analysis.

popDataA4.tsv - Same as fovealPopDataChips.csv, but for the A4 sized papers.

whiteOLED.zip - A collection of MATLAB MAT files with data for each of the 17 subjects that performed the white patch adjustment task on an OLED, as described in the Supplementary Info of the associated paper. Each MAT file contains a 2x3 matrix variable named, "v", which holds the "luminance (v)alue" that each observer choose as their impression of white (see Supplementary Info for more details). The first row contains values for the larger patch and the second row contains the values for the smaller patch (each tested 3 times). The values are in normalized units of 0-1 for our monitor. To convert to luminance, use the following transformation: 2.04698e-5*((v*1023)^2.28231).