Data Downloads
Hillis, J. M. & Brainard, D. H. (2005), Do common mechanisms of adaptation mediate color discrimination and appearance? Uniform backgrounds. Journal of the Optical Society of America A, 22, 2090-2106.
The data available on this web site were originally published in Hillis &
Brainard (2005). This paper investigated the effect of background luminance
and chromaticity on color appearance and discriminability. There are two types
of data files available on this web page. One type is discrimination data from
a 2-interval forced choice (2-IFC). The other data are from an asymmetric matching
task. This following describes how the data are formatted in the downloadable
text files.
In the paper, there were 3 adapting conditions for L- and M- cone isolating
tests and 3 adapting conditions for S-cone isolating tests. Data for LM test
and S tests are available in separates data files for each of the two observers.
Each data file contains 11 columns of tab delimited text. The first 9 columns
provide L, M, S coordinates the background, pedestal increment/decrement and
test increment/decrement respectively. Specifically, the first 3 columns correspond
to the L, M, S coordinates of the background, the next three columns correspond
to the L, M, S coordinates of the pedestal and the next three columns correspond
to the L, M, S coordinates of the pedestal. The last 2 columns correspond to
the number correct and number of trials in the 2-IFC. Thus the column organization
is as follows:
L_bg | M_bg | S_bg | L_ped | M_ped | S_ped | L_test | M_test | S_test | #Correct | #Trials |
These data are available in 3 formats. To compute the Smith-Porkorny cone coordinates,
cone spectral sensitivities were normalized to one and cone coordinates were
computed for the light reaching the cornea. Isomerization rates and totals are
computed for the light reaching the photoreceptors (i.e. light losses due to
optical factors in the eye are accounted for. Left click on the links below
to open data in web browser or right click to download (email Jamie, jmhillis@psych.upenn.edu,
if you have any problems obtaining the data).
(1) LMS Smith-Porkorny
JMH: LM cone data, S cone Data
QRS: LM cone data, S cone data
(2) LMS Isomerization rates
JMH: LM cone data, S cone Data
QRS: LM cone data, S cone data
(3) LMS Isomerization totals.
JMH: LM cone data, S cone Data
QRS: LM cone data, S cone data
Matching
Data
Similar to the discrimination data,
there were 3 adapting conditions for L- and M- cone isolating tests and 3 adapting
conditions for S-cone isolating tests. Each data file contains 13 columns of
tab delimited text. The first 6 columns provide L, M, S coordinates the background
againts which the test was presented and the test. The next 6 columns provide
L, M, S coordinates the background againts which the adjustable match was presented
and the observer's settings for those matches. The last column is the rating
(between 0 ans 3) the observer gave to the quality of the match they set. Thus
the column organization is as follows:
L_test_bg | M_test_bg | S_test_bg | L_test | M_test | S_test | L_match_bg | M_match_bg | S_match_bg | L_match | M_match | S_match | Rating |
Note that there is variation in the test values because the numbers reported here are values measured from the display. That is, after each block of trials a Photo Research 650 spectra-radiometer was used to measure the spectral density function of the test and match. The values of the test and match contained in the data tables are these measured values. These data are also available in 3 formats (left click to open data in web browser, right click to download):
(1) LMS Smith-Porkorny
JMH: LM cone data, S cone Data
QRS: LM cone data, S cone data
(2) LMS Isomerization rates
JMH: LM cone data, S cone Data
QRS: LM cone data, S cone data
(3) LMS Isomerization totals.
JMH: LM cone data, S cone Data
QRS: LM cone data, S cone data
Last Modified: January 11, 2007. DHB. Updated publication info for paper from in press to actual citation.