367 EFFECT OF EYELINER ON THE PERCEIVED SCLERA COLOR Stimuli data set 3 consists of 2 popular eyeliner colors, blue and brown, and 2 eyeliner colors darker than these. The dark eyeliner has been set as an intermediate color between the original blue/brown eyeliner and black eyeliner in the eyeliner layer. PARTICIPANTS In all the psychophysical experiments, people working at the Odawara Research Center of Kao Corporation were recruited as participants. All the participants provided informed consent and participated in the experiments in exchange for a monetary reward. The experimental protocol was approved by the Ethical Committee of Kao Corporation. All participants reported normal or corrected-to-normal visual acuity and color vision. In Experiment 1, 15 participants (9 males, 6 females, 28–46 years old, mean =35.0) were recruited for sclera color measurement, and 12 participants (6 males, 6 females, 25–54 years old, mean =34.3) were recruited for eye size measurement. The experiment was conducted between April and May 2019. In Experiment 2, 16 participants (8 males, 8 females, 20–46 years old, mean =32.2) were recruited for sclera color measurement, and 15 participants (7 males, 8 females, 26–50 years old, mean =35.7) were recruited for eye size measurement. The experiment was conducted from November to December 2019. In Experiment 3, 27 participants (12 males, 15 females, 21–52 years old, mean =36.2) were recruited for sclera color measurement, and 23 participants (12 males, 11 females, 22–53 years old, mean =34.7) were recruited for eye size measurement. This test was conducted in two parts: July 2020 (15 participants for sclera color and 14 participants for eye size) and March 2021 (12 participants for sclera color and 9 participants for eye size). PROCEDURE In the experiment, a standard stimulus and a comparison stimulus were displayed side by side. The standard stimulus was a simulated eyeliner-applied image, and the comparison stimulus was a sclera-color–modulated image or an eye-size–modulated image. In the perceptual sclera color measurement, the sclera-color–modulated images were used as the comparison stimuli, and the test subject answered which sclera of the stimuli was whiter. In the case of perceptual eye size measurement, comparison stimuli were eye-size–modulated images, and the one with the larger eyes was chosen among the two depicted forced-choice options (Figure 4). Figure 4. Experimental paradigm employed in this study.

368 JOURNAL OF COSMETIC SCIENCE The experimental task started when a participant clicked the start button displayed on the monitor. Two white-noise images were displayed for 1,000 ms, and standard and comparison stimuli were displayed after the noise. The background of the images was grey. The participant selected one of the images by clicking a button located at the bottom of the stimuli. After the choice, the stimuli disappeared, white-noise images were presented at the stimuli position, and the next pair of stimuli selected using the staircase method were displayed side by side. For each standard stimulus, there were two initially ascending staircases and two initially descending staircases. The standard and comparison stimuli were presented twice, with the left and right positions reversed. Hence, each stimulus image was evaluated four times during staircase sessions. In each staircase session, the first trial started from the end of the comparison stimuli set, i.e., darkest/whitest sclera color or largest/smallest eye size. Each staircase session was terminated when the staircase direction was reversed eight times. EXPERIMENTAL CONDITIONS All the experiments were conducted at the Odawara Research Center of Kao Corporation. The monitor used for the image presentation was an Eizo Color Edge CG243W (1,920 × 1,200 pixels) (Eizo Inc., Hakusan, Japan). The monitor was set up under room illumination of approximately 400 lx and calibrated with ColorMunki (X-Rite) such that the white point was 100 cd m−2 at D65, considering the room illumination conditions. The resolution of the stimulus image was 1,700 × 2,129 pixels, but the image was resized on the monitor such that the width from the chin to the top of the head was approximately 21 cm. The observation distance was set at 1 m, and a chin rest was set at 1 m from the monitor to control the observation conditions. ANALYSIS The PSE of each staircase session was computed by averaging the stimulus value (eye size or lightness L* of the sclera) of the comparison stimuli, where the staircase direction was reversed from upward to downward or from downward to upward. The first two responses were excluded because of large errors caused by missed clicks, and only the latter six points were used. The PSE of each standard stimulus was determined by averaging the PSE of four sessions conducted with each standard stimulus. In Experiment 1, the PSEs for each participant were measured, and average PSEs of different eyeliner thicknesses were compared using a paired t-test with Holm’s correction. In Experiment 2, the average PSEs of different eyeliner colors were compared to baseline (eye size 100% /original L*) by a one-sample t-test with Holm’s correction. Here, the average PSEs of different colors of the eyeliners were not compared. In contrast, in Experiment 3, the average PSEs of different eyeliner colors were compared using a paired t-test with Holm’s correction. RESULTS MEASUREMENT OF THE SCLERA COLOR CHANGE WITH AGE The measured L*a*b* values of the sclera were averaged and plotted for each age band, as shown in Figure 5. All the color channels, particularly L* and b*, exhibit statistically