1. INTRODUCTION

The role of melanopsin-dependent retinal ganglion cells in conscious vision has been supported by numerous studies in recent years [1–7]. Most of these studies focused on their role in brightness detection, but some studies have also focused on the ability to contribute chromatic information [8]. In a past study in our lab, we looked at the relative contribution of these cells to the brightness when other factors, such as chromaticity and luminance (both strong contributors to brightness perception), are also varied [1]. We used a four-channel projector apparatus to create stimuli with constant cone activity but variable melanopsin activity (silent substitution). Then we had subjects compare two targets that varied in either only melanopsin activity, or melanopsin activity and luminance, chromaticity, or both. We found that melanopsin signals contribute greatly to perceived brightness when the other factors are held constant (no variation in the luminance or chromaticity between targets). However, this contribution became minimal or disappeared as these other factors varied significantly.

Another longtime area of interest in our research has been the influence of the surround characteristics on the red/green balance point of a target. This is a relatively complex process with contributions from chromatic, brightness, and saturation contrasts [9,10]. These complexities can be analyzed by looking at the relative amount of S, M, and L cone activity in a given target compared to its surround. Chromatic contributions are largely simple and behave as expected from the long history of chromatic contrast studies: If a surround has greater L cone activity relative to a target, it will induce greenishness into that target––and vice versa for the induction of reddishness. However, simple changes in the luminance with stable ratios of L/M also contribute chromatic induction to a target. A surround that has relatively higher luminance does not just cause a target to appear darker, but also to appear greener (and vice versa for a low luminance surround and inducing reddishness). This is what is responsible for several well-known illusions and studies concerning brown, where a balanced yellow stimulus will look like a greenish brown when made dim or a balanced brown stimulus will appear reddish (orange) when made bright [11–13]. This same phenomenon also occurs for blue stimuli (which are also a red/green balance), but is not nearly as strong. This is due to the contribution of S-cone activity to these induction effects. This little-appreciated aspect has been well established by our past research. As S-cone activity in a target relative to its surround increases, the degree to which a surround induces reddishness or greenishness into a target is reduced [14]. This is a largely linear relationship.

However, one strange aspect of these induction effects is that they do not act as expected when a surround is equiluminant [10]. An equiluminant surround will cause an induction of reddishness greater than that of a dimmer surround. This equiluminant surround induction effect quickly disappears when the surround is made brighter or dimmer, which suggests some sort of additional induction effect that is only present when a surround is equiluminant. We have been unable to find an aspect that could explain this phenomenon, but its presence only when a surround is equiluminant suggests that melanopsin-dependent ipRGCs could play a role. As we showed in our earlier experiment [1], melanopsin signaling only showed a significant brightness induction effect in conditions where the targets were equiluminant. Therefore, due to the large role brightness has in these induction effects it is possible melanopsin could be contributing in these equiluminant viewing conditions. There is a major problem with this simple explanation, although the effect of the equiluminant surround is equivalent to that of a significantly dimmer surround. It is not inducing greenishness like a brighter surround would, but instead induces reddishness. This does not mean that there is no potential role for melanopsin in surround induction, but it does mean it cannot simply be a contributor to brightness induction if it is responsible for the previously observed equiluminant surround induction effect. Regardless of the bearing of these results on that past experiment, this is a constant question of psychophysics experiments: To what degree is it necessary to control for melanopsin activity? Most experiments ignore it entirely, but under some conditions, such as the equiluminant brightness induction discussed in our previous experiment, it is a significant contributor to conscious visual phenomena. Therefore, a major motivation in this experiment is to determine to what extent the melanopsin activity may affect psychophysics experiments that use equiluminant surrounds.

One very important thing to note about our experiment, which will be discussed in more depth in the Section 4, is that we cannot keep rod activity silenced on a four-channel apparatus; to do that would require a five-channel apparatus. The type of apparatus we have constructed allows the display and manipulation of relatively complex stimuli and surrounds, but at a greatly reduced amount of melanopsin variation compared to many other apparatuses. Research has shown interactions between rod and melanopsin signals even at photopic light levels, such as those used in the present study [15]. We also know from our past research that rods play a role in these induction effects [16], although the direction and consistency of the rod effects vary between the subjects and depend on the nature of the luminance contrast. We have attempted to control the rod activity to the degree to which we can using this apparatus, but it is also in keeping with our goal to try and stay somewhat close to the conditions under which the psychophysics experiments are normally run. Additionally, a recent experiment by Barrionuevo et al. [8] successfully controlled rod interference through bleaching, although it used a much stronger bleaching paradigm than we used here.

To test for the role of melanopsin activity in surround-induced red/green balance shifts, we presented subjects with a central yellow target surrounded by an equiluminant surround that could vary in the amount of melanopsin-activating light present. The central yellow stimulus had a randomly changing red/green balance that subjects then adjusted to a balance point, or null, where it appeared neither reddish nor greenish. Initially data was collected under normal (room adapted) conditions, but concerns of rod interference led to a new set of experiments. Appendix A provides the data from the initial experiment. It is recommended not to directly compare data from the initial experiment with the data in the main experiment due to the differences in the observers, the time, and adaptation procedure. (See Appendix A for further details.) In this new set of experiments, adjustments were completed under dark-adapted and light-adapted conditions to attempt to control for rod influence. We hypothesized that melanopsin would contribute to brightness perception as it had in our previous experiment. Therefore, it would induce greenishness into the test, which would require the subjects to add more red to achieve a balance. Our past experiments on rod induction effects showed different effects for dark and bright surrounds, so it was difficult to base our hypothesis on these data because we are using equiluminant surrounds in this experiment [16,17]. We instead hypothesized that rods would show a similar brightness induction effect and also induce greenishness into the test based on the literature showing a role for rods in brightness induction [18].

2. METHODS

A. Apparatus

The apparatus is identical to the one used in DeLawyer et al. [1], which was also used in 2018 by Yang et al. [19]. To quote from [1], “We used three NEC 5500-lumen Widescreen Advanced Professional Installation Projectors. Each of the projectors has a filter applied such that 4 primaries (red, blue, yellow, and green) are created that can be varied independently. The three filtered light sources are then projected onto a white diffuser plate. In order to align the projectors an identical circular central stimulus was projected by all 3 projectors and then they were manually adjusted until visible edge inhomogeneities of the circle stimuli were eliminated. Due to the nature of the apparatus visible inhomogeneities were created on the screen, especially near the outer edges. We used a Topcon SR-5000 2D Spectroradiometer to measure a central area of the screen that would be used for the experiment (20° visual angle in total). This area appeared homogenous and we confirmed the light in the 20° experimental area did not possess inhomogeneities greater than 1% in luminance or CIE xyz coordinates when a full field white stimulus was displayed. We then measured the combined output of the three projectors with a Minolta CS-1000 Spectroradiometer. In front of the diffuser a chinrest was utilized with a pinhole cut-out placed in front of it. The pinhole cut-out was set to 2 mm and a black square with a circular cutout that a corrective lens could be attached to the center was placed after it. The total visible area for the experiment was approximately 20° of visual angle.” The purpose of the pinhole cutout was to create a Newtonian view apparatus with a constrained visual field (20°) that could incorporate corrective lenses, not to control for pupil size. The apparatus is controlled by a single computer running Windows 10 and a Nvidia Quadro K 2200 controls the projectors with 8-bit color at 59 Hz.

Table 1. Properties of the Surrounds Used in the Experiment

View Table | View all tables in this article

 

Fig. 1. (a) and (b) Spectral distribution of the low and high melanopsin conditions, respectively. The four primaries have a peak wavelength of 445, 533, 578, and 598 nm. (c) Spectral distribution of the light adapting field.

Download Full Size | PDF

B. Stimulus Set

The color of the surround was selected based on the previous experiments’ orange surround, which allows for a great variation in the melanopsin activity. This surround covered everything except the central 2.5° foveal area where the test stimulus was presented. The surround could either randomly appear as a high or low melanopsin activity version, as detailed in Table 1, with the spectral distributions visible in Fig. 1. The values seen in Table 1 are an average of the values used during the six experimental sessions. Due to the nature of the apparatus (three separate projectors with lamps that are naturally susceptible to the loss of luminance over time and temperature-dependent fluctuation in the output), the stimulus values (the luminance, and therefore, the final stimulus chromaticity values) would vary over time. To compensate for this, measurements were taken before each session and adjustments were made if the values deviated more than 1% from the reported values in either the luminance, or the CIE x or CIE y values. Although the variation of the stimulus parameters between sessions could be substantial, the variation within a session was relatively small. More details can be seen in [1]. The reported CIE x and CIE y values are based on the 1964 10° standard observer (1964 values were used for ease of use with our spectrometer) [20]. The cone fundamentals are from Stockman and Sharpe [21]. It can be argued that the central 2.5° test stimulus is better represented by CIE 2° values because the macular pigment density can influence the luminance and especially the S-cone contribution. However, because we were not attempting silence substitution of the test with the surround outside of the luminance and 10° values were also used in our past experiment [10] we decided to use them here as well. Note that the nature of the experiment required the test stimulus to vary in every regard except luminance. The potential implications of this will be discussed in the Section 4. The calculation of melanopsin stimulation was done in accordance with what was reported in 2011 by Tsujimura and Tokuda [22]. Accordingly, we used a pigment template nomogram with a peak wavelength of 480 nm and a peak axial optical density of 0.4 to calculate a peak spectral sensitivity function for melanopsin stimulation of 489 nm. The total melanopsin activity in the high condition is 136% that of the low condition. Additionally, the calculation of cones, rods, and melanopsin using CIE S 026 can be found in Appendix B. The ratio of melanopsin activation is the same when using CIE S 026 (1.36, high to low) [23]. The test stimulus would flash on and off during the experiment (2 s on and 3 s off) to prevent chromatic adaptation [24–26]. The test stimulus was equiluminant (based on CIE 1964 10° values), but set to a random L/M ratio that observers could then vary the ratio of L to M cone activity, allowing it to appear more reddish or greenish (which could range from an ${\rm L}/({\rm L} + {\rm M})$ cone ratio of 0.676 to 0.754). The test stimulus was an average of ${493.4}\;{{\rm cd/m}^2}$ and CIE 1964: x 0.465, y 0.515, although the luminance was kept constant, the CIE coordinates could vary considerably (up to $+$ or − approximately 0.05. for x and 0.04 for y) as the ratio of L to M cone troland activity was adjusted by the observer. The stimuli were generated by a Dell computer running Windows 10 with MATLAB using the Psychtoolbox-3 [27]. Generated stimuli extend beyond the visible experiment area to reduce edge inhomogeneities.

C. Procedure

The procedures and experiments described here conform to the principles expressed in the Declaration of Helsinki and were approved by the Kochi University of Technology Research Ethics Committee. Written informed consent was obtained from each observer prior to testing. A total of six subjects between the ages of 21 and 29 participated in the experiment. They were native speakers of English, French, Japanese, Mandarin Chinese, or Indonesian. Although these subjects had extensive experience with setting red/green balances, they were not informed about the nature of the experiment beforehand and did not know the surrounds were varying in their melanopsin activity. To make sure they could accurately collect the maximum amount of data within the short time frame available they completed multiple training sessions (2–3) before collecting actual data. Within a session, subjects would collect 40 trials on both dark-adapted and light-adapted conditions (20 trials with a high melanopsin surround, and 20 trials with a low melanopsin surround). All sessions had to be completed within an hour because the display had to be recalibrated. The adaptation conditions meant that all data had to be collected within 12 min. Due to the training, all subjects were able to do this for the actual experiment.

 

Fig. 2. Results from the four conditions as box plots showing the 25–75% interquartile range in the boxes with the stems representing the maximum and minimum values of the data combined from the six subjects. Light and Dark refer to light and dark adaptation, respectively. High and Low, refer to high and low melanopsin activity surrounds, respectively. Subject data is given as the ratio of ${\rm L}/({\rm L} + {\rm M})$ cone trolands on average for their balanced yellow stimuli, a measure of the surround induction of redness or greenness. Error bars represent the SEM for the group average data.

Download Full Size | PDF

Dark adaptation was achieved by having each subject wear an eye patch in a pitch black room (${\lt}{0.01}\;{{\rm cd/m}^2}$) for 15 min and only moving the eye patch off their test eye and onto their other eye when the experiment began. Although 15 min may be insufficient for a full dark adaptation, but it was used to keep the experiment within a one h time frame. The chinrest and optics were adjusted before the experiment began to minimize interference with adaptation. Light adaptation was achieved using the XC-100 Artificial Sun. The manufacturer states the luminance is ${3000}\;{{\rm cd/m}^2}$ (2°), but we used an indirect presentation that was ${995.5}\;{{\rm cd/m}^2}$ (10°) with the spectral distribution shown in Fig. 1. This allowed for a bright adaptation field, without causing discomfort or persistent visual afterimages. Subjects would look at the bright light for 4 min before the experiment, with a 2 min recovery period for cones. The total amount of cones, rods, and melanopsin bleaching was calculated with the formulae used in 2021 by Pant et al. [28]. Using the constants there, along with an assumed pupil size of 2 mm to calculate a photopic troland amount of 3,177.4 (I), and a time course of 240 s ($\tau$), we calculated a 13% cone bleach, a 16% rod bleach and a 3% melanopsin bleach. The experiment was conducted in a dark room (${\lt}{0.01}\;{{\rm cd/m}^2}$) and room lights were not used during the experiment. The room lighting present during the preparation for the experiment was approximately ${31.3}\;{{\rm cd/m}^2}$ when viewing the display area with the projectors off, although the subjects were free to look as they pleased prior to the start of the experiment. Display calibration was completed immediately prior to the start of the experiment.

Subjects were instructed to set the test stimulus to a red/green balance, or null, where the target appeared neither reddish nor greenish. Subjects were asked to prioritize accuracy and make sure they viewed at least two full cycles (10 s) of stimulus presentation before accepting a balance. No technical terms or concepts such as cone trolands or unique yellow were discussed. Instructions were available in English, Japanese, and Mandarin Chinese. Subjects completed near identical conditions as training at least two or three times before the actual experimental condition (not on the same day as the experiment). All the data was analyzed using IBM’s SPSS statistical package.

3. RESULTS

A repeated measures ANOVA was performed on the data for the two levels of the melanopsin variable (high and low) and the two levels of the adaptation variable (light adapted and dark adapted). The ANOVA is shown in Table 2. The results show a significant (${\rm F}\;({1},{5}) = {13.249}$, ${\rm p} = .{015}$) main effect for melanopsin (high melanopsin surround compared to low melanopsin surround) with the high melanopsin condition causing subjects to set their balance points at significantly higher ${\rm L}/{\rm L} + {\rm M}$ cone ratios (in line with a surround induction of greenness) with a relatively large effect size (${{\boldsymbol \eta}^{\textbf 2}}\; = .{727}$). There was no significant main effect (${\rm F}\;({1},{5}) = {13.249}$, ${\rm p} = .{426}$) for adaptation (light versus dark) and the interaction between the melanopsin and adaptation was not significant (${\rm F}\;({1},{5}) = {13.249}$, ${\rm p} = .{843}$).

Table 2. Repeated Measures ANOVA for the Experimental Data

View Table | View all tables in this article

The results for the group data can be seen in Fig. 2. The data is spread over a wide range due to a high amount of variance between subjects’ perceptual null point for the yellow stimuli. Figure 3 shows the results for individual observers. The individual data shows very low SEMs because of the high internal consistency of the subjects in setting their perceptual null point for yellow.

 

Fig. 3. Results from the experiment for the four conditions and six subjects. Light and Dark refer to light and dark adaptation, respectively. High and Low, refer to high and low melanopsin activity surrounds, respectively. Subject data is given as the ratio of  ${\rm L}/({\rm L} + {\rm M})$ cone trolands on average for their balanced yellow stimuli, a measure of surround induction of redness or greenness. Error bars represent the SEM for individual subject data.

Download Full Size | PDF

4. DISCUSSION

The results of this experiment supported the hypothesis of a melanopsin-mediated induction of greenness by the surround into the yellow target. This is what also would be expected from a brightness induction, but it cannot be said that the only explanation for this result is due to an increased brightness in the surround. It is also possible the increased melanopsin activity preferentially activates the L-cone (or S-cone, as per an ${\rm S} + {\rm L}$ versus M color opponency model) pathways, which would require more L-cone activation in a central target to create a balance. This is because a surround that has more L-cone activity (reddish) would induce greenishness into a central stimulus, thus requiring greater L-cone activity in the center to overcome the green induction and achieve a balance.

The results for the adaptation condition show no significant effect for adaptation (light versus dark), although all but one observer had data that trended in the direction of light adaptation requiring greater amounts of L-cone excitation to achieve a balance. It is possible if more subjects completed these conditions it would be significant, as our past research showed some subjects have rod influences that go in the opposite direction of the majority. If we assume that the dark adaptation condition increases rod activity that would suggest that rods are inducing reddishness into the central test for most observers. This is not in line with a rod contribution to brightness nor is it in the direction of the observed effect here that is potentially attributable to melanopsin. However, it could reflect differential adaptation effects that both rods and melanopsin undergo following light or dark adaptation [28]. Because both rods and melanopsin contribute to the color appearance during cone-mediated vision [16,29], a differential state of adaptation between the two could drive the overall effect in one direction or the other. Individual variations in the adaptation time courses and spectral sensitivity could also contribute to the observed individual variations between observers here. Because the light adaptation paradigm we used does not bleach most rods, it is possible the adaptation effect is significantly understated in this experiment.

These results do not explain the equiluminant surround induction effects we have seen in the past (where no control or measurement of melanopsin excitation was used), as melanopsin activity goes in the direction of a brightness induction effect (inducing greenishness into the central target) rather than the induction of reddishness we observed in those experiments [10]. Therefore, this question remains open: Why do the equiluminant surrounds behave differently than brighter or dimmer surrounds (having an effect more in line with that of a significantly dimmer surround)? In [1], the melanopsin influence was the greatest at equiluminant conditions, providing a possible explanation, but the current data doesn’t support melanopsin as a potential explanation for the equiluminant surrounds’ significant induction of reddishness into the central stimuli.

Another goal of this experiment was to examine the potential influence of melanopsin on surrounds induction in general. Here, we demonstrated a robust and consistent melanopsin effect that statistically has a relatively strong effect size. However, the size of the effect in terms of how much it shifts the red/green balance points is not particularly large. Individual observers show far more variation between each other than the systematic variation the melanopsin contributes. Individual variation in the yellow null points is not unusual, and our past studies have shown similar variations [9]. These individual variations do not appear influenced by the relatively large variation in human cone ratios, as even studies that measure cone ratios show persistent uncorrelated variations between individual observers in their unique yellow points [29]. For practical purposes, it is unlikely that most researchers need to worry about the melanopsin interference from a surround. Although we did not use an overly large contrast in this study (136% melanopsin in the high condition compared to the low condition), it is quite large compared to things that would be equiluminant in nature but vary in their melanopsin activity as well as for a given display. Depending on the nature of the effect being investigated, you may or may not want to control for melanopsin activity in the surround. All displays produce some level of melanopsin stimulation, which normally cannot be manipulated on a standard display (whether CRT, LCD, or OLED). It is unlikely that the total amount of melanopsin activity varies by a large degree between the calibrated displays used within the field (at least not to the degree to which we have manipulated it within this experiment). However, if a very minute surround induction effect is being studied, it would be beneficial to control all possible interfering variables. Even if melanopsin is not varying systematically to a large degree between displays, it is still likely contributing to some effects that are reported using those displays. However, if a surround induction effect that is very powerful or does not involve equiluminant surrounds is being studied, it is unlikely that melanopsin activity is contributing significantly to a given effect. Additionally, rod effects have been consistently shown to influence the color appearance as well under a variety of lighting conditions and stimulus presentations, and therefore should also be considered in applicable experiments [30,31].

The effect of adaptation was not significant in this experiment. Although collecting more data could lead to significance (as only one subject showed an effect in the opposite direction of the group means), it would be better to study it on a normal apparatus. The amount of time and effort required to prepare and run an experiment on this apparatus doesn’t lend itself to such a relatively simple psychophysics experiment, which could instead be run on a standard calibrated CRT. Still, our past research left open this question: To what extent and direction do rods influence red/green balance for equiluminant surrounds? We believe this experiment provided some initial data in that regard.

The two biggest potential obstacles to explaining this result are penumbral cones and rods. Penumbral cones are cones lying in the shadows of blood vessels (the penumbra) that shift the wavelength of incoming light, preferentially biasing cone activation in what is believed to be a silence substitution experiment [32]. Penumbral cones can be controlled by using temporal white noise [33,34], but unfortunately that is not practical for this type of experiment. Penumbral cones are a potential confound, although they could also simply be extraneous. However, to explain the surround induction effects, a penumbral cone mediated effect would be just as relevant as one mediated by melanopsin. Additionally, because we used a relatively low-temporal frequency stimulus, the effect of penumbral cones is smaller than it would be for a higher frequency stimulus [32].

For rods, which can have a large effect on brightness perception, we attempted to control them with light and dark adaptation conditions. Initially, one may wonder why we would need to control the rods at all when our stimuli are so bright, but the spectral distribution of our apparatus is such that rods are not receiving much stimulation near their spectral peak. It cannot be assumed they are always bleached under these conditions. However, the adapting field we used was even brighter and fully covered the spectral sensitivity curve of the rods. We computed the bleach of rods under this condition as approximately 16%, which is significantly less than that in many other studies that had more than 85% rod bleach using brighter lights with shorter exposures [8,28]. However, except for one observer, when the subjects’ data was considered, the trend was that the rods bias the surround in the opposite direction of the effect we attribute to melanopsin. We did not see a significant interaction, so it cannot be said that rods mask the melanopsin effect, but they also do not enhance it or appear to confound it. That said, it would be worth investigating the effects of a more thorough rod bleaching on the present experimental paradigm.

It is also worth discussing the issue of the central foveal stimulus being measured with CIE 10° values. The difference in macular pigment density between the fovea and periphery is significant and also varies between individuals [35]. There is good reason to believe the central luminance values and especially S-cone trolands reported with 10° fundamentals are not fully accurate for this reason. Our previous study did use 10° values [10], but the results here do not suggest a role for melanopsin in the effect observed there. It is preferable for future studies to use the CIE 2° values or otherwise modify the 10° values to account for macular pigment density differences. As our central stimulus was being varied in the L/M ratio by observers, we did not seek to hold the center and surround constant in anything outside of luminance. Because the luminance measured will vary between CIE 2° and 10° and as equiluminant stimuli appear to enhance the influence of melanopsin [1], future studies should establish the potential role of luminance variation between the center and the surround on the effect described here.

Many aspects still remain to be explored concerning the melanopsin contribution to conscious vision and the factors involved in the surround induction left. Although we do not see a particularly strong role for melanopsin in the surround induction effects at this time, we have shown a potential contribution under some circumstances that may warrant consideration. It is possible that other surround conditions could be affected to a greater degree by melanopsin activity, but the conditions used here were not ideal to stimulate melanopsin; instead, they were a compromise between the external and internal validity. Still unexplored are the potential roles of luminance contrast and the degree to which a more thorough rod bleaching procedure may influence these induction effects. Additionally, some studies [4,34,36] suggest a chromatic aspect to melanopsin stimulation that could contribute to surround inductions through color contrast effects. Further research should continue to try to find roles for melanopsin in complex scenes and stimulus configurations.

APPENDIX A: ROOM ADAPTED MELANOPSIN DATA

The data presented here comes from eight subjects, five who are in the other experiment and one who is the author of this paper. It was collected under conditions identical to those in the first experiment but with no adaptation outside of standard room lighting (room adapted, approximately ${31.3}\;{{\rm cd/m}^2}$ when looking at the display area with the projectors off). However, no set time to adapt to the room lighting was used and there was a considerable time difference between when the two experiments were conducted (around four months, which can cause systematic changes in unique yellow perception [37]), so the data should not be compared directly to the experiment in the main paper due to the difference in the subjects, the adaptation procedure, and the time period when the experiment was performed. The graphed data and ANOVA are presented in Table 3 and Fig. 4.

Table 3. Repeated Measures ANOVA for the Data Presented in Appendix A (Fig. 4)

View Table | View all tables in this article

 

Fig. 4. Individual data for subjects completing the room adapted initial experiment. The high and low correspond to high and low melanopsin activity surrounds, respectively. Error bars are subject SEMs.

Download Full Size | PDF

APPENDIX B: CALCULATION OF CONE, ROD, AND MELANOPSIN RADIANCE USING CIE 026

Table 4 is the data from Table 1 using calculations from CIE 026.

Table 4. Properties of the Surrounds Used in the Experiment Using CIE 026

View Table | View all tables in this article

Funding

Japan Society for the Promotion of Science (18H03323).

Acknowledgment

This research was supported by Grant-in-Aid for Specific Research to Vision and Affective Science Integrated Research Laboratory, Research Institute from Kochi University of Technology to KS and by JSPS KAKENHI (B) to KS.

Disclosures

The authors declare no conflicts of interest.

Data availability

Raw data from this experiment as well as additional data from the author are available upon request.

REFERENCES

1. T. DeLawyer, S. Tsujimura, and K. Shinomori, “Relative contributions of melanopsin to brightness discrimination when hue and luminance also vary,” J. Opt. Soc. Am. A 37, A81–A88 (2020). [CrossRef]  

2. T. M. Brown, C. Gias, M. Hatori, S. R. Keding, M. Semo, P. J. Coffey, J. Gigg, H. D. Piggins, S. Panda, and R. J. Lucas, “Melanopsin contributions to irradiance coding in the Thalamo-cortical visual system,” PLoS Biol. 8, e1000558 (2010). [CrossRef]  

3. T. M. Brown, S. Tsujimura, A. E. Allen, J. Wynne, R. Bedford, G. Vickery, A. Vugler, and R. J. Lucas, “Melanopsin-based brightness discrimination in mice and humans,” Curr. Biol. 22, 1134–1141 (2012). [CrossRef]  

4. D. Cao, A. Chang, and S. Gai, “Evidence for an impact of melanopsin activation on unique white perception,” J. Opt. Soc. Am. A 35, B287–B291 (2018). [CrossRef]  

5. A. J. Zele, P. Adhikari, B. Feigl, and D. Cao, “Cone and melanopsin contributions to human brightness estimation,” J. Opt. Soc. Am. A 35, B19–B25 (2018). [CrossRef]  

6. M. Yamakawa, S. Tsujimura, and K. Okajima, “A quantitative analysis of the contribution of melanopsin to brightness perception,” Sci. Rep. 9, 7568 (2019). [CrossRef]  

7. A. J. Zele, A. Dey, P. Adhikari, and B. Feigl, “Melanopsin hypersensitivity dominates interictal photophobia in migraine,” Cephalalgia 41, 217–226 (2021). [CrossRef]  

8. P. A. Barrionuevo, C. P. Filgueira, and D. Cao, “Is melanopsin activation affecting large field color-matching functions?” J. Opt. Soc. Am. A 39, 1104–1110 (2022). [CrossRef]  

9. S. L. Buck, F. Rieke, and T. DeLawyer, “Contrast-dependent red-green hue shift,” J. Opt. Soc. Am. A 35, B136–B143 (2018). [CrossRef]  

10. T. De Lawyer, “Brown induction and red-green hue shifts,” Ph.d. dissertation (University of Washington, 2017).

11. R. B. Lotto and D. Purves, “The empirical basis of color perception,” Conscious. Cogn. 11, 609–629 (2002). [CrossRef]  

12. C. J. Bartleson, “Brown,” Color Res. Appl. 1, 181–191 (1976).

13. H. Uchikawa, K. Uchikawa, and R. M. Boynton, “Influence of achromatic surrounds on categorical color perception of surface colors,” Vis. Res. 29, 881–890 (1989). [CrossRef]  

14. T. DeLawyer, M. Tayon, C. Yu, and S. L. Buck, “Contrast-dependent red-green balance shifts depend on S-cone activity,” J. Opt. Soc. Am. A 35, B114–B121 (2018). [CrossRef]  

15. S. Uprety, P. Adhikari, B. Feigl, and A. J. Zele, “Melanopsin photoreception differentially modulates rod-mediated and cone-mediated human temporal vision,” Iscience 25, 104529 (2022). [CrossRef]  

16. S. L. Buck and T. DeLawyer, “Dark versus bright equilibrium hues: rod and cone biases,” J. Opt. Soc. Am. A 31, A75–A81 (2014). [CrossRef]  

17. S. L. Buck and T. DeLawyer, “A new comparison of brown and yellow,” J. Vis. 12(14): 9 (2012). [CrossRef]  

18. H. Sun, J. Pokorny, and V. C. Smith, “Brightness induction from rods,” J. Vis. 1(1): 4 (2001). [CrossRef]  

19. P. L. Yang, S. Tsujimura, A. Matsumoto, W. Yamashita, and S. L. Yeh, “Subjective time expansion with increased stimulation of intrinsically photosensitive retinal ganglion cells,” Sci. Rep. 8, 11693 (2018). [CrossRef]  

20. CIE, Proceedings, Vienna Session, 1963 (Bureau Central de la CIE, 1964), Vol. B.

21. A. Stockman and L. T. Sharpe, “The spectral sensitivities of the middle-and long-wavelength-sensitive cones derived from measurements in observers of known genotype,” Vis. Res. 40, 1711–1737 (2000). [CrossRef]  

22. S. Tsujimura and Y. Tokuda, “Delayed response of human melanopsin retinal ganglion cells on the pupillary light reflex,” Ophthalmic Physiol. Opt. 31, 469–479 (2011). [CrossRef]  

23. CIE, CIE System for Metrology of Optical Radiation for ipRGC-Influenced Responses to Light,” CIE S 026:2018.

24. J. M. Loomis and T. Berger, “Effects of chromatic adaptation on color discrimination and color appearance,” Vis. Res. 19, 891–901 (1979). [CrossRef]  

25. M. D. Fairchild and L. Reniff, “Time course of chromatic adaptation for color-appearance judgments,” J. Opt. Soc. Am. A 12, 824–833 (1995). [CrossRef]  

26. O. Rinner and K. R. Gegenfurtner, “Time course of chromatic adaptation for color appearance and discrimination,” Vis. Res. 40, 1813–1826 (2000). [CrossRef]  

27. M. Kleiner, D. Brainard, and D. Pelli, “What’s new in Psychtoolbox-3?” Perception 36, 14 (2007).

28. M. Pant, A. J. Zele, B. Feigl, and P. Adhikari, “Light adaptation characteristics of melanopsin,” Vis. Res. 188, 126–138 (2021). [CrossRef]  

29. J. Neitz, J. Carroll, Y. Yamauchi, M. Neitz, and D. R. Williams, “Color perception is mediated by a plastic neural mechanism that is adjustable in adults,” Neuron 35, 783–792 (2002). [CrossRef]  

30. S. L. Buck, R. Knight, G. Fowler, and B. Hunt, “Rod influence on hue-scaling functions,” Vis. Res. 38, 3259–3263 (1998). [CrossRef]  

31. W. B. Thoreson and D. M. Dacey, “Diverse cell types, circuits, and mechanisms for color vision in the vertebrate retina,” Physiol. Rev. 9, 1527–1573 (2019). [CrossRef]  

32. M. Spitschan, G. K. Aguirre, and D. H. Brainard, “Selective stimulation of penumbral cones reveals perception in the shadow of retinal blood vessels,” PLoS One 10, e0124328 (2015). [CrossRef]  

33. A. J. Zele, P. Adhikari, D. Cao, and B. Feigl, “Melanopsin driven enhancement of cone-mediated visual processing,” Vis. Res. 160, 72–81 (2019). [CrossRef]  

34. A. J. Zele, P. Adhikari, D. Cao, and B. Feigl, “Melanopsin and cone photoreceptor inputs to the afferent pupil light response,” Fron. Neurol. 10, 529 (2019). [CrossRef]  

35. A. G. Robson, J. D. Moreland, D. Pauleikhoff, T. Morrissey, G. E. Holder, F. W. Fitzke, A. C. Bird, and F. J. G. M. van Kuijk, “Macular pigment density and distribution: comparison of fundus autofluorescence with minimum motion photometry,” Vis. Res. 43, 1765–1775 (2003). [CrossRef]  

36. M. Spitschan, A. S. Bock, J. Ryan, G. Frazzetta, D. H. Brainard, and G. K. Aguirre, “The human visual cortex response to melanopsin-directed stimulation is accompanied by a distinct perceptual experience,” Proc. Natl. Acad. Sci. USA 114, 12291–12296 (2017). [CrossRef]  

37. L. E. Welbourne, A. B. Morland, and A. R. Wade, “Human colour perception changes between seasons,” Curr. Biol. 25, R646–R647 (2015). [CrossRef]