Colour & Vision Research laboratory

Background

The Colour & Vision Research laboratory database originated at the University of California at San Diego in 1995. In 2001, the database and CVRL moved to the Institite of Ophthalmology at University College London, where they continue today.


Overview

Professor Andrew Stockman (AS) is the Steers Chair of Investigative Eye Research at the UCL Institute of Ophthalmology. He gained a B.A. in Experimental Psychology at Oxford followed by a Ph.D. specializing in human vision research at Cambridge under the supervision of John Mollon.  His research has been funded by a succession of NSF, NIH, Wellcome Trust, Fight for Sight and BBSRC grants. His principal research area is visual psychophysics, within which his specializations include colour vision, rod vision, visual adaptation and temporal sensitivity. AS is well known for his work on human spectral sensitivities. The “Stockman & Sharpe” cone spectral sensitivities and the related luminous efficiency function, all based on measurements in observers of known photopigment opsin genotype [1,2], have now been adopted by the Commission Internationale de l'Éclairage (CIE) as a new international standard for colour definition and colour measurement [3]. These functions are central to modern work on colour vision. Other important work includes measurements of cone and rod temporal (flicker) sensitivity and delay that has resulted in the identification of “slow” and “fast” signals in photopic and mesopic vision, the discovery of an unexpected S-cone input to luminance [5, 6], work on rod and cone adaptation [e.g., 7]; and measurements of visible distortion that allows the performance of the visual system to be dissected into early and late stages [e.g., 8]. At UCL, AS has also been involved in the visual assessment of clinical patients in several important collaborations with Moorfields Eye Hospital, including experimental work on the effects of specific gene defects in rod and/or cone photo­receptors, and their implications for retinal processing of visual signals in normal vision. AS is a Fellow of the Optical Society of America. Dr G. Bruce Henning (GBH) is an internationally-respected figure in the fields of visual and auditory psychophysics. He obtained a B.A. at the University of Toronto, and his Ph.D. was supervised by David M. Green at the University of Pennsylvania, where he also benefited from the teaching of Jack Nachmias, Saul Sternberg, and Duncan Luce. GBH became a lecturer in the Department of Experimental Psychology in Oxford where his research was supported principally by the Wellcome Trust. He is a Fellow of the Acoustical Society of America. GBH is best known in vision science for masking experiments in spatial vision that demonstrated failings in the linear channel model of spatial vision, for his studies of spatial detection and discrimination, and for insights into the mechanisms underlying Mach bands. His publications on visual topics are concerned mainly with spatial vision, but his work has dealt extensively with visual processing and thus has close links with the proposed research [9-12].

Selected projects. Please refer to our publications for papers related to other projects.

1.1 A psychophysical dissection of the common pathway that encodes colour and brightness. Flickering a light can change its appearance: lights near 560 nm appear brighter, and those near 650 nm appear yellower. Both effects are consistent with distortion within normal visual pathways: brightness enhancement at an expansive nonlinearity and hue change at a compressive one.  We have used the distortion products generated by each nonlinearity to extract the temporal properties of the early (pre-nonlinearity) and late (post-nonlinearity) stages of the pathways signalling brightness or colour. A surprising and important finding was that the attenuation characteristics of the two pathways are virtually identical both before and after the nonlinearity: in both brightness and hue pathways, the early temporal stage is a band-pass filter peaking at 10-15 Hz; the late stage is a two-stage low-pass filter with a cutting off near 3 Hz.
We proposed a novel physiologically-relevant model that not only accounts for both filter shapes but incorporates the expansive and compressive nonlinearities in a common, probably parvocellular, pathway that signals hue and brightness. This BBSRC-supported work has led to three substantial recent publications [13-15].

1.2 Red-green flicker is encoded by a peak detector but limited by slew rate.The appearance of rapid-on (slowly-off) or rapid-off (slowly-on) sawtooth-modulated flicker depends on temporal frequency. At low frequencies (< 4 Hz), both L- and M-cone isolating flicker is seen as a red-green hue change that approximately follows the waveform. However, at higher frequencies (5-13 Hz) the hue change becomes asymmetrical and the flicker takes on a steady hue appearance that depends on sawtooth direction: rapid-on-L-cone and rapid-off-M-cone sawtooth stimuli appear greener than the non-flickering average, while rapid-off-L-cone and rapid-on-M-cone stimuli appear redder, even though all have the same mean chromaticities. These changes can be explained by supposing that chromatic mechanisms are better able to track the slowly-changing part of the sawtooth than the quickly-changing part, because the hue mechanism is limited by a maximum rate of change or “slew” rate (thus skewing the mean output in the direction of the slow change).
To investigate this further, we decomposed the sawtooth stimuli into their sinusoidal components and then, presenting the important 1st and 2nd components alone, varied their relative phase. Our results show that red-green hue appearance is mediated by a mechanism that encodes peak excursions towards red or green, but exhibits a limiting slew rate when the frequency of the sawtooth exceeds about 4 Hz (so that its 2nd harmonic exceeds about 8 Hz). This BBSRC-supported work, which focuses on basic questions about how chromatic signals are processed, is in progress.

1.3 Light adaptation, molecular mechanisms, and gene defects. We have previously characterized and modelled sensitivity regulation in normal observers by measuring temporal sensitivities and phase delays as a function of adaptation level to allow a more complete characterization of the effects of light adaptation alone [7, 16] . We were able to link each stage of regulation to molecular processes within the photoreceptor: shortened time, increases in sensitivity to increased rates of molecular resynthesis and changes in channel sensitivity, and decreases in sensitivity at high levels, to photopigment bleaching.  We have extended this work by making measurements in patients with specific defects in genes that encode proteins in the photoreceptor or elsewhere in the retina: GNAT2, RGS9, GCAP1, RetGC1, KNCV2, RPE65 and NR2E3 [17-21] . By comparing psychophysical measurements for each disorder with data from normal observers, we characterized the losses associated with each defect, in each case linking the losses to the underlying molecular defect. This work has yielded new insights relating molecular mechanisms underlying normal visual functions.

1.4 Cone spectral sensitivities and luminous efficiency. The trichromacy of human colour vision depends initially on the long, middle- and short-wavelength-sensitive (L, M and S) cones and, in particular, on their spectral sensitivities. They are the physiological determinants of human colour matching, and thus of all other CMFs.
The cone fundamentals of Stockman & Sharpe (2000) have been adopted by the CIE as the “physiologically-relevant” international standard for colorimetry [3]. They rely upon psychophysical measurements made in both normal trichromats and colour deficient observers, and upon a direct analysis of the 10-deg CMF data of Stiles & Burch [22]. As a subsidiary function, the committee has also adopted the luminous efficiency [V(lambda)] proposed by Sharpe et al. [23], which is a linear combination the L- and M-cone spectral sensitivities.
By making a few simple assumptions and using Grassman’s Laws, the cone fundamental CMFs can be linearly transformed to the more familiar color­imetric variants: the X, Y and Z CMFs, a form still in common use (e.g., by assuming that the Y CMF is the V(lambda) function of Sharpe et al. [23], and that the Z CMF is the S-cone spectral sensitivity).

1.5 Sluggish, spectrally-opponent inputs to the luminance channel. We have published clear evidence  that the luminance pathway has slow, spectrally-opponent inputs in addition to its expected fast, additive L- and M-cone inputs [5, 6, 24-26] . Funded originally by NIH and the Wellcome Trust, this work is the starting point for this current proposal. The project should reveal new and fundamental insights into signal processing in the normal human visual system by building on a variety of strong preliminary evidence that flicker perception depends on multiple signals from photo­receptors.

1.6 Colour & Vision database on the World Wide Web.  We maintain a web-based database that provides an annotated and easily downloaded library of standard data sets relevant to colour and vision research (http://www.cvrl.org). The database collects and centralizes many basic data sets that have been difficult to obtain and are now widely used in the vision science community, simplifying the research of others.

References

1.      Stockman, A., L.T. Sharpe, and C.C. Fach, Vision Research, 1999. 39: 2901-2927.
2.      Stockman, A. and L.T. Sharpe, Vision Research, 2000. 40: 1711-1737.
3.      CIE, Fundamental chromaticity diagram with physiological axes – Part 1. Technical Report 170-1. 2006, Vienna: Central Bureau of the Commission Internationale de l' Éclairage.
4.      Smith, V.C., et al., Journal of Neurophysiology, 2001. 85: 545-558.
5.      Stockman, A., D.I.A. MacLeod, and D.D. DePriest, Vision Research, 1991. 31: 189-208.
6.      Stockman, A. and D.J. Plummer, Journal of Vision, 2005. 5: 702-716.
7.      Stockman, A., et al., Journal of Vision, 2006. 6: 1194-1213.
8.      Stockman, A. and D.J. Plummer, Vision Research, 1998. 38: 3703-3728.
9.      Henning, G.B., B.G. Hertz, and D.E. Broadbent, Vision Research, 1975. 15: 887-897.
10.   Henning, G.B., Journal of the Optical Society of America A, 2004. 21: 486-490.
11.   Curnow, T., et al., Journal of the Optical Society of America  A, 2007. 24: 3233-3241.
12.   Goris, R.L.T., F.A. Wichmann, and G.B. Henning, Journal of Vision, 2009. 9: 1.1-18.
13.   Stockman, A., D. Petrova, and G.B. Henning, Journal of Vision, 2014. 14: 1.1-32.
14.   Petrova, D., G.B. Henning, and A. Stockman, Journal of Vision, 2013. 13: 15.1-23.
15.   Petrova, D., G.B. Henning, and A. Stockman, Journal of Vision, 2013. 13: 2.1-26.
16.   Stockman, A., M. Langendörfer, and L.T. Sharpe, Journal of Vision, 2007. 7: 4.1-7.
17.   Stockman, A., et al., Investigative Ophthalmology and Visual Science, 2014. 55: 832-884.
18.   Ripamonti, C., et al., Investigative Ophthalmology and Visual Science, 2014. 55: 963-976
19.   Stockman, A., et al., Investigative Ophthalmology and Visual Science, 2014. 55: 1930-1940.
20.   Stockman, A., et al., Journal of Vision, 2008. 8: 10.1-10.
21.   Stockman, A., et al., Journal of Vision, 2007. 7: 8.1-13.
22.   Stiles, W.S. and J.M. Burch, Optica Acta, 1959. 6: 1-26.
23.   Sharpe, L.T., et al., Color Research & Application, 2011. 36: 42-46.
24.   Stockman, A., D.J. Plummer, and E.D. Montag, Journal of Physiology, 2005. 566: 61-76.
25.   Stockman, A. and D.J. Plummer, Journal of Physiology, 2005. 566: 77-91.
26.   Stockman, A., E.D. Montag, and D.J. Plummer, Visual Neuroscience, 2006. 23: 471-478.


Lab address

UCL Institute of Ophthalmology, 11-43 Bath Street, London EC1V 9EL, UK