Palmer Awards Meeting

These awards are made in memory of David Palmer who was a vision scientist mainly known for his work on mesopic vision. In his later years he was appalled by the way successive Governments treated science, scientists and those being educated in science. He devoted much time to ways to improve this and fully supported the programmes of, for example, the Royal Institution and the Institute of Physics, to bring an understanding of science to the general population. He left substantial sums to several organisations to contiue this work and one beneficiary was the Colour Group and this has been used to endow an annual lecture and this set of awards, and to support a Teaching Fellows to work in schools. A short biography may be found here.

Programme:
Time	Speaker	Title
14.00 hrs	Mike Pointer	Introduction
14.10 hrs	W. Bi	The contribution of the rod/melanopsin driven ganglion cells to the dynamic pupil light reflex response
14.40 hrs	Evgenia Konstantakopoulou	Assessment of Chromatic Sensitivity in Female Carriers of Colour Vision Deficiency
15.10 hrs	TEA
15.30 hrs	Kaida Xiao	Evaluation of Unique Hue Predictions in CIECAM02 by using Unique Hue Data
16.00 hrs	Maria Georgoula	The Impact of Luminance Level on the Assessments of Colour Appearance and Difference
16.30 hrs	Questions and finale

Venue: Room CG04, Tait Building, City University, Northampton Square, London
Entry: The meeting is free and open to all

---

ABSTRACTS

The contribution of the rod/melanopsin driven ganglion cells to the dynamic pupil light reflex response
Wei Bi 1, S.Tsujimura2, G.T. Plant3 and J.L. Barbur
1 Applied Vision Research Centre, City University, Northampton Square, London, UK
2 Department of Information Science and Biomedical Engineering, Kagoshima University, Kagoshima, JAPAN.
3 The National Hospital for Neurology and Neurosurgery, Queen Square, London, UK

Recent studies have shown that the steady-state size of the pupil during long exposure to intense stimuli, and the corresponding sustained constriction in darkness following the offset of the stimulus, are likely to involve melanopsin signals through intrinsically photosensitive retinal ganglion cells. These cells also receive spatially pooled signals from rod and cone photoreceptors and the extent to which melanopsin contributes to rapid pupil responses elicited with brief stimuli remains less clear. The purpose of this study was to examine how the dynamic pupil light reflex response changes with the level of light adaptation for stimuli that isolate luminance, colour or combined rod and melanopsin.
An integrating sphere was used to homogenise the light output of four LED sources (selected to span the visual spectrum) which were modulated temporally to produce a 2 s sinusoidal envelope of luminance, colour or combined rod and melanopsin. Pupil responses were measured at three light levels: 456, 74.4 and 4.8 cd m-2. The study involved six normal subjects and two patients with congenital stationary night blindness.
In normal subjects, each of the three stimuli produced brief constrictions of the pupil at stimulus onset, with larger response amplitudes at lower light levels. Consistent with previous studies, the colour modulation also produced a large response at stimulus offset. The rod / melanopsin isolation condition yields larger response latencies, with the largest difference observed at the lowest light level. The patients with CSNB yield normal responses to colour and luminance modulation, but much reduced or absent pupil constriction to rod/melanopsin modulation at each light level.
When carefully interpreted, the findings from this study suggest that pupil responses to briefly presented stimuli are mediated largely by rod and cone signals with little or no contribution from melanopsin. The results also suggest that, although more sluggish, rod signals remain unsaturated and contribute to dynamic pupil responses at much higher light levels.

Assessment of Chromatic Sensitivity in Female Carriers of Colour Vision Deficiency
Evgenia Konstantakopoulou
Department of Optometry and Visual Science, City University, London

In spite of numerous studies, colour vision in female carriers remains controversial with some studies reporting improvement and others a worsening of red/green (RG) chromatic sensitivity. The purpose of this study is to investigate the extent to which the additional photopigment that can be expressed in female carriers of colour deficiency affects their RG chromatic sensitivity.
Thirty heterozygote females, aged 22 to 73 years, were examined. RG and yellow/blue (YB) chromatic sensitivity were measured using the Colour Assessment Diagnosis test. Four background light adaptation levels (i.e., 65, 26, 7.8 and 2.6 cd m-2) were employed and measurement of pupil size at each level provided an average measure of retinal illuminance. The optical density of the crystalline lens, i.e. for the short wavelength region, and the macular pigment optical density were measured using the Macular Assessment Profile test. Forty-one male subjects, aged 23 to 71 years, were used as controls.
The carriers were divided into four groups: carriers of deuteranomaly (DA), deuteranopia (D), protanomaly (PA) and protanopia (P). Carriers of DA and D showed lower RG chromatic sensitivity than normal trichromats (p<0.05), whereas carriers of PA or P showed similar RG sensitivity to controls (p>0.05).
Differences in RG colour vision between controls and P and PA carriers are not statistically significant. This is not the case for D and DA carriers, who show significantly reduced RG chromatic sensitivity. The results suggest that a relative increase in the number of LL’ cones in the retina, as expected in the D and DA groups, can cause a significant worsening of RG colour vision, whilst an increase in the number of MM’ cones, as expected in the P and PA groups, does not cause such a decrease. The processing of the fourth photopigment could play a significant role in carriers of anomalous trichromacies.

Evaluation of Unique Hue Predictions in CIECAM02 by using Unique Hue Data
Kaida Xiao
School of Clinical Dentistry, The University of Sheffield

In this paper, we present the results of the second part of our behavioural experiments, in which the same 185 observers repeated the same unique hues experiment using the same CRT display but under a daylight simulator (D65) and typical office (CWF) ambient illumination conditions respectively. At the end of each task, the selected unique hue stimuli were re-displayed on the CRT and measured with a tele-spectroradiometer, under identical illumination conditions. The unique hue stimuli were recorded in CIE XYZ tristimulus values in the unit of cd m-2 based on a 2 standard observer. In total, 59 940 unique hue data for three viewing conditions were collected. These data were then used to evaluate CIECAM02 by predicting colour appearance attributes for unique stimuli under these three illumination conditions.
Comparing the intra-observer variability between the dark condition and the mixed illumination conditions (D65 or CWF) showed that observers were more consistent for all four hues in mixed illumination conditions. Equally, the variability across observers was also found to be lower under D65 and CWF compared to the dark condition. The lowest observer variability was found for the unique green hue. The intra-observer variability was approximately 50% of the inter-observer variability.
The unique hue data were then transformed to colour appearance attributes using the CIECAM02 colour appearance model and were plotted in CIECAM02 chromatic diagram. Based on these unique hue stimuli, the mean predicted hue angles (h) for each unique hue in CIECAM02, in three different viewing conditions are summarized; a linear model to represent unique hue lines for each viewing condition in the CIECAM02 chromatic diagram is developed; the performance of hue uniformity for unique hue stimuli are also evaluated.

The Impact of Luminance Level on the Assessments of Colour Appearance and Difference
Maria Georgoula
Department of Colour Science , University of Leeds

Luminance is one of the major parametric factors in the viewing field. The objectives of this project were to model the visual effect, to evaluate the measured and perceived colour difference and to investigate the performance of various colour difference formulae and colour appearance models under different luminance levels.
Twenty observers participated in a unique experimental setup including two viewing cabinets and a combined psychophysical method using short-term memory matching and a grey scale methods for the investigation. Five luminance levels (about 1258, 445, 33, 3 and 0.8 cd m-2) were adjusted by using optical density filters. Forty coloured textile hue-difference pairs with a mean ΔEab* of 7.8 were measured with a tele-spectroradio-meter under all luminance levels.
The data analysis revealed inter- and intra-observer variability of 28% and 30% respectively. Furthermore, perceived colour differences decrease by 3%, 21%, 31% and 40% from the brightest luminance level (1258 cd m-2) against the other levels. The analysis on performance of CIELAB and CIEDE2000 colour-difference formulae, plus the CIECAM02 and CAM02-UCS colour appearance models, exposed that each formula gave more accurate prediction than the inter-observer variability. In addition, it was found that all formulae and spaces gave similar performances with CAM02-UCS marginally better, when an individual scaling factor was applied to each experimental phase. When applying only one scaling factor to the visual results from all five luminance phases however, CAM02-UCS and CIECAM02 performed distinctly better among all the formulae and spaces studied. This suggests that CAM02-USC can effectively estimate the impact of luminance on colour appearance.