Color Brightness Recognition of Extremely Severe Amblyopia Children in an Indoor Environment

Abstract

This study aims to investigate how indoor lighting (natural and artificial) and distances (3 m and 5 m) affect color recognition in visually impaired children. Ten participants from a special education school were selected to identify the brightness of five colors at varying lighting and distance circumstances. Each color was presented at six different brightness levels, classified into the low-brightness, the standard-color, and the high-brightness groups. Participants were directed to assess the top three brightness levels they considered most attractive, and each rating was assigned a weighted score. The findings revealed that: (1) Visually impaired children can recognize color brightness in both natural and artificial lighting situations. In indoor conditions, the low-brightness group exhibited greater recognition ability compared to the high-brightness group. Purple did not exhibit a clear pattern, as colors from the high-brightness, the low-brightness, and the standard-color groups were all preferred. (2) Significant differences were observed in the brightness recognition among visually impaired children at distances of 3 m and 5 m. Recognition for low-brightness colors improved with distance, contrasting high-brightness scores that declined. However, there was no significant variation in the perception of green with distance changes.

1. Introduction

According to the survey and estimate conducted by the World Health Organization in 2010, there are approximately 285 million visually impaired people worldwide, accounting for about 4.24% of the total global population. Among them, 39 million are blind, while 246 million individuals have impaired vision [1]. In China, there is a population of over 40 million individuals who are visually impaired, including 12 million children, with this figure continuing to rise [2]. Visual impairment encompasses blindness and low vision [3], which are classified into five levels: mild (0.5–0.8), moderate (0.3–0.5), severe (0.1–0.3), very severe (0.05–0.1), and complete blindness (below 0.05) [4]. Compared with sighted children, visually impaired children face challenges in visual acuity or field, either due to congenital or non-congenital factors. These challenges restrict their involvement in education, work, and social activities and seriously affect their quality of life as well as their physical and mental health [5,6,7]. It is important to note that many visually impaired children are not completely blind [8]. Instead, they have poor vision and heavily rely on color perception compared to others with normal vision [9]. Many studies have shown that color design can improve the visual recognition of visually impaired children, which plays an important role in their growth [10,11]. Additionally, color design in indoor environments has an impact on their cognitive abilities, attention span, creativity, emotions, and personality development [12]. Special education schools catering to visually impaired children in China play a crucial role in providing fundamental education and serve as a cornerstone of special education [13]. Unfortunately, a solid foundation in design research within the architectural domain is conspicuously absent in these institutions [14,15,16]. Early studies focused on examining the methodologies of interior and furniture design for various functional spaces, with a keen consideration of ergonomics for individuals with disabilities [14,17,18,19]. Follow-up studies have underscored the significance of the environment for special groups, demonstrating that interventions in the built environment are instrumental in the developmental progress of children with disabilities [1,6,20,21,22]. Given the challenges faced by visually impaired individuals in spatial orientation [21] and the multifaceted role of color as a tool for identification, grouping, highlighting, warning, and interpretation within educational settings [23], the color design of school campuses plays a crucial role in the lives of visually impaired children [24,25].

Many scholars have a misunderstanding regarding the visual capabilities of the visually impaired population, often conflating individuals with visual impairments with those who are completely blind, thereby assuming they have entirely lost their vision. As a result, early research predominantly focused on Braille, sensory compensation, and obstacle avoidance devices, with very limited attention given to the study of color. Over time, research specifically targeting individuals with low vision has gradually emerged, including studies on ocular devices and VR games, where color was used as a tool within other research contexts. However, these studies were primarily aimed at improving vision through instruments, and their conclusions were not specifically centered on color. We have identified the critical role of special education schools for children with visual impairments and noted a significant lack of research on the use of color in indoor environments for the visually impaired. We believe that indoor color can assist in visual recognition for this population, which serves as the foundational motivation for our study.

There is a scarcity of research that connects the visually impaired population, indoor environments, and color. Our study represents foundational research on indoor color recognition for the visually impaired, with all our references drawn from indirect studies. However, we are confident that as more scholars focus on the role of color in indoor environments for the visually impaired, the standards for indoor color design will become increasingly refined. Our study is an exploratory attempt, and we hope that our research will provide new insights and research directions for future scholars in this field.

Gardner examined the impact of various combinations of figure–background contrasts on the visual function of visually impaired children. The findings indicated that neither contrast inversion nor chromatic changes effectively improved the visual ability of visually impaired children [26]. When addressing aspects like color matching, sign design, roads, and building accessories, Ying Li et al. argued that accessible colors should be both visually appealing and easily visible to visually impaired individuals [9]. Bin Cheng et al. introduced a visual system and instructional design tailored for those with low vision and blindness in indoor environments, emphasizing the importance of standardized text spacing, graphic size, and color brightness [27]. Huang, K.-C. delved into the effects of color and background brightness on LCD screens for individuals with low vision, with experimental results underscoring the importance of electronic text information and icon design for visually impaired individuals [28]. Baker et al. identified the shopping requirements of visually impaired individuals and proposed a color design approach [29]. Liu Jianying’s research, based on the deficiencies of visually impaired students, outlined design approaches for map color design and coordination [30]. Özkan et al. assessed the hotel experience for visually impaired individuals, discussing various aspects of signage, including braille, lighting, color, contrast, and layout [31]. Zou et al. proposed solutions such as greening and color design specifically for the visually impaired [32]. Recent interdisciplinary research has delved into the field of engineering techniques by utilizing grating orientation and color-aware interfaces for the visually impaired [33,34,35,36]. It is noteworthy that in recent years, color applications have expanded into industrial design, incorporating low-vision assistive devices and visual navigation systems that utilize color and machinery to augment visual capabilities [37,38]. Furthermore, there has been the emergence of low-vision VR games that employ color and spatial elements to enhance the visual experience for individuals with visual impairment [11,39].

Previous studies synthesized methods for interior color design in special education schools [24,40]. However, these studies did not explore the relationship between the indoor environment and color recognition, and the practical applications of color remain somewhat ambiguous. Research indicated that indoor lighting had a considerable impact on both visual comfort and color recognition [41,42,43,44,45,46,47]. Moreover, color and lighting are crucial for indoor wayfinding [48]. The visual perception of color in visually impaired children is also affected by the distance at which they observe objects [49]. Nevertheless, the specific impacts of these factors remain unclear. Previous experimental studies on color perception among the visually impaired have exhibited inconsistency in the choice of experimental settings. The research, whether it was color studies related to shopping preferences [29] or color investigations on hotel signage [31], was conducted in the public environment. However, it is important to note that special education schools are not considered the public environment, and the findings from previous studies may not necessarily be applicable to the specific requirements of special education schools.

Apart from the only two existing studies on color design in special education, other research has primarily focused on the strategic use of color to improve the living conditions of individuals with visual impairments. These studies have applied color in various contexts, including signage design, electronic texts, and VR games, with a particular emphasis on the interaction between visual impairment and color contrast. Building on this prior research, the present study delves into the impact of indoor environmental factors on the color recognition abilities of children with low vision, with a specific focus on variables such as lighting and distance. This study contributes to the growing body of literature on color research within the context of visual impairment.

2. Methods

2.1. Experimental Program for Color Brightness Perception

Participants were tasked with identifying the colors red, yellow, blue, green, and purple under varying lighting conditions (natural and artificial light) and distances (3 m and 5 m). Each color was classified into six levels of brightness. The participants aimed to choose 1 to 3 levels of brightness within the color palette based on recognition ability, with the data being recorded accordingly. A weighted scoring method was employed for data analysis, assigning 1 point to each participant’s choice with weights of 50%, 30%, and 20% for the three choices. Data organization, statistical analysis, and interpretation were conducted using SPSS software (v.29). Two independent sample t-tests were performed to compare the differences in weighted scores between the two groups, while an analysis of variance (ANOVA) was applied to assess variations across four groups. In cases of group differences, the Student–Newman–Keuls method was employed for pairwise comparisons. The significance level was set at α = 0.05, with p values less than 0.05 indicating statistically significant differences.

2.2. Participants

The participants, ranging in age from 6 to 13 years, exhibited variability in their perception of color hues. The experimental criterion for color recognition was not to ‘select the color that most closely resembles a specific hue’, but rather to ‘identify the three most attention-catching luminance levels’. Therefore, we designed the data analysis to identify the luminance levels that are most easily recognized by children aged 6 to 13, emphasizing their sensitivity to luminance rather than their understanding of color hues.

Table 1. List of experimental subjects.

Serial Number	Age	Gender	Class of Vision	Left Vision	Right Eye Vision
N1	6	Female	Extremely severe	0.1	0.15
N2	8	Female	Extremely severe	0.25	0.1
N3	12	Female	Extremely severe	0.1	0.1
N4	13	Female	Extremely severe	0.02	0.1
N5	13	Male	Extremely severe	0.02	0.12
N6	7	Male	Extremely severe	0.1	0.2
N7	10	Male	Extremely severe	0.2	0.15
N8	12	Male	Extremely severe	0.2	0.2
N9	11	Male	Extremely severe	0.1	0.15
N10	12	Male	Extremely severe	0.1	0.25

presents the visual acuity range for the 10 participants. Visual acuity values are expressed in “Decimal Visual Acuity,” a format commonly used in China and Japan. A visual acuity of 1.0 corresponds to 20/20 or 6/6 on the LogMAR chart, indicating normal vision. Visual acuity values between 0.8 and 0.5 represent subnormal vision, with descending values reflecting a degradation in visual acuity. In accordance with the World Health Organization (WHO) standards, individuals with a visual acuity below 0.3 in Decimal Visual Acuity are classified as having severe visual impairment.

In this experiment, we enrolled over 60 children with amblyopia from a special education school. In this group, 11 children were diagnosed with a visual impairment to be categorized as severe amblyopia, while the remaining children exhibited milder forms of the condition. The objective of our study is to identify luminance levels of colors that are more recognizable in an indoor setting. Consequently, if colors are recognized by children with severe amblyopia, they should also be more easily identifiable by those with less severe vision impairments. The inclusion of all children in the experiment could potentially compromise the validity of the results.

Among the 11 children with severe amblyopia, one was found to have a color vision deficiency, leading to an experimental sample of 10 participants. However, a sample size of 10 is relatively modest, which is acknowledged as one of the limitations of our study.

In the design of this experiment, we intentionally excluded a control group and did not engage children with normal vision. This decision was informed by Chinese educational policy that necessitates children with disabilities to undertake a 9- to 12-year residential education within specialized schools. These institutions are equipped with distinct facilities, each meticulously tailored to address the unique requirements of various disabled student populations, including dedicated buildings for the visually impaired, the hearing impaired, and others. Given this context, our experiment was conducted within the building specifically designated for visually impaired students, an environment where all facilities are thoughtfully designed to meet the specific needs of children with amblyopia. Therefore, the scope of our data collection was limited to this particular group, as they represent the target population for our research.

2.3. Ethical Approval

The ethical committee of Northeast Forestry University granted a waiver of ethics and informed consent for the study due to “not involving any drugs or medical-related equipment”. The research team obtained verbal consent from both the principal and minor participants along with their guardians. As the experiment did not involve the use of any drugs or specialized equipment, approval from an ethics committee was not necessary. In order to safeguard privacy, the authors were only provided with information regarding the participants’ age, vision, and gender, without access to their names, addresses, or family connections. The authors were not authorized to photograph the participants, obtain signed consent forms from them, or exercise any other rights. We guarantee that all procedures are conducted under the guidance of the Helsinki Declaration.

2.4. Color Brightness Test Sample

2.4.1. Selection of Experimental Colors

These five colors—red, yellow, green, blue, and purple—hold pervasive applications across various facets of daily life. Red, with its high contrast and attention-grabbing properties, is a universal choice for signaling emergencies, danger, or prohibition, evident in traffic lights and stop signs. Yellow, recognized for its visibility under low-light conditions, is a prevalent choice for warnings or alerts, as seen in cautionary signage. Green, often linked with safety or permission to proceed, is commonly utilized in green traffic lights and evacuation routes and is generally regarded as a calming and reassuring color, minimizing the likelihood of misinterpretation. Blue, while less frequently employed, can be found in informational prompts or directional guidance, such as in specific traffic signs. Purple, a composite color derived from red and blue, is less commonly utilized in daily life but is included in China’s color blindness tests alongside the aforementioned colors. The selection of lower-brightness shades of purple by visually impaired children may suggest a diminished sensitivity to composite colors, which could indicate a potential risk of latent color vision deficiency and potentially exclude them from the experiment. The significance of these five colors in the daily lives of visually impaired children is underscored by the fact that ambiguous hues, such as orange and pink, are less appropriate for them. Given that color purity significantly influences hue perception, this study focuses exclusively on brightness in its examination of color.

The authors categorized the 11 levels of brightness for the 5 colors into 6 color numbers based on the RGB color attributes, as illustrated in Figure 1. The colors labeled as No. 1 are standard red (255, 0, 0), medium yellow (255, 217, 0), standard blue (0, 0, 255), standard green (0, 255, 0), and standard purple (128, 0, 128). The choice of medium yellow over standard yellow (255, 255, 0) was deliberate, given the prevalence of medium yellow in China’s blind lanes and vehicle turn signals. These five colors were specifically selected for color blindness testing in China and hold practical significance in everyday life.

The authors categorized colors based on brightness into six groups. Group 1 and Group 4 are designated as the standard-color group, characterized by strong hue properties and distinct color tendencies. Groups 2 and 3 constitute the high-brightness group, where the standard colors have been adjusted with the white attribute. Groups 5 and 6 form the low-brightness group, with the standard colors modified by adjusting the black attributes. This grouping of brightness levels will facilitate the analysis of the specific effects of lighting and distance on color recognition in future experiments.

2.4.2. The Production of Color Palettes

Walls and ceilings in Chinese special education schools are predominantly white or gray. For the purpose of this experiment, six levels of brightness for five colors were randomly arranged on a gray background, as illustrated in Figure 2.

2.5. Setting Up the Experimental Environment

Rather than using a dedicated visual function training room for our experimental location, we decided to carry out the investigation in a regular classroom setting. This decision was informed by the consideration that visual function training rooms are equipped with an array of specialized equipment, including lighting, curtains, and furniture, all calibrated to facilitate visual training. Although such an environment is ideal for its intended function, it may not accurately represent the ambient conditions encountered in everyday settings. The colors appropriate for children with low vision should meet the needs of their daily living environments and not be strictly tailored for a specific indoor context.

The experimental site was located in a classroom in a special education school, which was equipped with nine lighting fixtures. These fixtures were LED (light-emitting diode) models with a warm white color, 40 W (watt) power, 350 Lux (luminance) illumination, 5000 K color temperature, and a color rendering index of 90. The staff in the image is demonstrating the recognition distance, as depicted in Figure 3.

2.5.1. Lighting Control

This study categorized indoor lighting into two distinct types: natural light and artificial light. These two sources of light have distinct color rendering indices, resulting in the display of objects in different colorations. The natural light will be simulated at 12:00 Beijing time, with indoor illuminance measured at a range of 370 to 460 Lux. On the other hand, the artificial light will be simulated at 19:00 Beijing time by turning on nine lights, after which the indoor illuminance is measured to be within the range of 300 to 450 Lux. The experiments were carried out under these two specified lighting situations.

2.5.2. Distance Control

In China, the benchmark distance for visual acuity testing is conventionally established at 5 m, a standard that also informs the calibration of the eye chart. This 5 m standard serves dual purposes: it is easily adopted in most indoor settings, including hospitals and schools, and it is a widely used distance not only in China but also in other countries like Japan and Germany, thereby enhancing the comparability of visual assessment outcomes.

In pursuit of a comprehensive understanding of the impact of various distances on color recognition, we introduced an additional testing distance of 3 m. This decision was motivated by the observation that the visual field of individuals with visual impairments is considerably reduced at distances exceeding 5 m.

3. Results

3.1. Recognition Results under Different Lighting and Distance Conditions

3.1.1. Legend Explanation

The experimental results are designed to demonstrate the variations in different brightness levels under various lighting and distance conditions. Line graphs are employed to effectively illustrate the trends in brightness levels. In Figure 4, Figure 5, Figure 6, Figure 7 and Figure 8 the respective conditions of lighting and distance conditions are represented, with solid lines representing the natural light and a 3 m distance and dashed lines indicating the artificial light and a 5 m distance. This graphical depiction affords a clear illustration of the changes in brightness levels. Additionally, the significance differences marked by letters further emphasize statistically significant results.

We conducted a comparative analysis across six levels of brightness, examining variations within specific lighting and distance conditions—namely, the natural light, artificial light, at 3 m, and at 5 m—as well as between these conditions, such as natural versus artificial light and the 3 m versus 5 m distances. For more details, please refer to Figure 4, Figure 5, Figure 6, Figure 7 and Figure 8.

3.1.2. Red Recognition

There was no statistically significant difference in the weighted score among the six levels of brightness within the natural light group (all with p > 0.05). In contrast, the differences in the weighted score within the artificial light group were statistically significant (p = 0.014). The pairwise comparison revealed that the weighted score for brightness level 5 was significantly greater than those for brightness levels 1, 2, 3, and 4. However, there was no statistically significant difference in the weighted score of the brightness level 6 and the other levels. Additionally, the differences between the lighting groups were not statistically significant. For more information, please refer to Figure 4A.

The weighted score of the six levels of brightness within the 3 m group showed a statistically significant difference (p = 0.006). The results of the pairwise comparisons indicated that the weighted scores for brightness levels 1, 4, and 5 were considerably greater than those for brightness levels 3 and 6. However, the weighted score for brightness level 2 did not exhibit statistically significant differences when compared to the other brightness levels. Within the 5 m group, the difference in the weighted score among the six brightness levels was also statistically significant (p < 0.001), with the pairwise comparison revealing that the weighted scores for brightness levels 5 and 6 were markedly higher than those for levels 1 to 4. The disparity among the distance group was statistically significant as well. Specifically, the weighted score at a distance of 3 m was higher than that at 5 m when the brightness level was set at 1 (p = 0.032). Conversely, when brightness was adjusted to level 6, the weighted score at a distance of 3 m was found to be lower than that at 5 m (p = 0.007). There were no notable disparities identified for the other levels of brightness. For more detailed information, please refer to Figure 4B.

Figure 4. Red—Comparative analysis of the weighted score differences between different lighting conditions and distance conditions. Note: The letters a and b represent statistically significant differences in the weighted scores between different brightness values.

Figure 4. Red—Comparative analysis of the weighted score differences between different lighting conditions and distance conditions. Note: The letters a and b represent statistically significant differences in the weighted scores between different brightness values.

3.1.3. Yellow Recognition

There was a statistically significant difference in the weighted score across the six levels of yellow brightness (p = 0.003) in the natural light group. The pairwise comparisons revealed that brightness level 5 yielded significantly higher scores than brightness levels 1, 2, 3, 4, and 6. Within the artificial light group, there was also a statistically significant difference in the weighted scores among the six levels of brightness (p < 0.001). The pairwise comparison demonstrated that the brightness level 5 scored significantly higher than levels 1, 2, and 6, but not when compared to levels 3 and 4. Differences between lighting groups did not reach statistical significance. Please refer to Figure 5A for more detailed information.

Within the 3 m group, there was a statistically significant difference in the weighted score among the six different brightness levels (p < 0.001). The pairwise comparison revealed that the weighted score for brightness level 5 was significantly higher than those for brightness levels 1, 2, 3, 4, and 6. Similarly, within the 5 m group, the difference in the weighted score among the six levels of brightness was also statistically significant (p < 0.001), with the weighted scores for brightness 5 and 6 being significantly higher than those for levels 1 to 4. The comparison between distance groups also showed statistical significance. Specifically, when brightness was set at level 3, the weighted score for 3 m was higher than for 5 m (p < 0.001). Conversely, when brightness was set at level 6, the weighted score for 3 m was lower than for 5 m (p < 0.001). No statistical significance was found in the differences between brightness levels 1, 2, 3, and 5 across distance groups (p > 0.05). Please refer to Figure 5B for more detailed information.

Figure 5. Yellow—Comparative analysis of the weighted score differences for different lighting and distance conditions. Note: The letters a and b represent statistically significant differences in the weighted scores between different brightness values.

Figure 5. Yellow—Comparative analysis of the weighted score differences for different lighting and distance conditions. Note: The letters a and b represent statistically significant differences in the weighted scores between different brightness values.

3.1.4. Blue Recognition

Within the natural light group, the difference in weighted scores among the six brightness levels was statistically significant (p < 0.001). The pairwise comparison indicated that brightness levels 5 and 6 were significantly higher than those for levels 2 and 3. Additionally, the brightness level 1 showed no statistical difference in the weighted score compared to levels 2, 3, 4, and 6. The brightness level 6 had a higher weighted score than the brightness level 1. In the artificial light group, the difference in the weighted score among the six brightness levels was also statistically significant (p < 0.001). Here, the weighted score for the brightness level 5 was significantly higher than those for levels 1, 2, and 3, while there was no statistically significant difference when compared to level 6. Similarly, the weighted score for the brightness level 1 was not statistically different from those for levels 2, 3, 4, and 6. Notably, differences between the two lighting groups did not reach statistical significance. Please refer to Figure 6A for further details.

Within the 3 m group, there was a statistically significant difference in the weighted score among the six levels of brightness (p < 0.001). The pairwise comparison revealed that the weighted score for brightness level 5 was significantly higher than those for levels 1, 2, 3, and 4. No statistical difference was found between the weighted score of brightness level 6 and those for brightness levels 1, 3, and 6. In the 5 m group, the difference in the weighted score among the six levels of brightness was also statistically significant (p < 0.001), with the pairwise comparisons indicating that the weighted scores for brightness levels 5 and 6 were significantly higher than those for levels 1 to 4. Specifically, at brightness level 2, the weighted score at 3 m was higher than that at 5 m (p = 0.025), whereas changes in the weighted score at other levels of brightness did not exhibit statistical significance across distance groups. It is important to note that there was a statistically significant difference in the distance group, particularly at the brightness level 2, where the weighted score at 3 m surpassed that at 5 m (p = 0.025). For a detailed illustration of these findings, please refer to Figure 6B.

Figure 6. Blue—Comparative analysis of the weighted score differences between different lighting and distance conditions. Note: The letters a and b represent statistically significant differences in the weighted scores between different brightness values.

Figure 6. Blue—Comparative analysis of the weighted score differences between different lighting and distance conditions. Note: The letters a and b represent statistically significant differences in the weighted scores between different brightness values.

3.1.5. Green Recognition

Within the natural light group, there was a statistically significant difference in the weighted score among the six levels of brightness (p = 0.005). The pairwise comparisons revealed that the weighted scores for brightness levels 5 and 6 were significantly higher than those for levels 2 and 3. In contrast, the weighted score for brightness level 4 did not exhibit statistically significant differences when compared to the other levels. In the artificial light group, a significant difference in the weighted score among six brightness levels was observed (p < 0.001). Here, the weighted scores for levels 5 and 6 were found to be significantly higher than those for levels 1, 2, and 3, whereas the weighted score for brightness level 4 showed no statistical differences when compared to other levels. Notably, differences between the two lighting groups were not statistically significant. Please refer to Figure 7A for a detailed illustration.

Within the 3 m group, there was a statistically significant difference in the weighted score among the six levels of brightness (p = 0.003). Specifically, the pairwise comparisons revealed that the weighted score for brightness level 6 was significantly higher than those for levels 1, 2, 3, and 4. However, the weighted score for brightness level 5 did not differ statistically from those of the other levels. In the 5 m group, the difference in the weighted score among the six levels of brightness was also statistically significant (p < 0.001), with the pairwise comparisons indicating that the weighted scores for brightness levels 5 and 6 were significantly higher than those for levels 1 to 4. No statistically significant differences were observed between the distance groups. Please refer to Figure 7B for more detailed information.

Figure 7. Green—Comparative analysis of the weighted score differences between different lighting conditions and distance conditions. Note: The letters a and b represent statistically significant differences in the weighted scores between different brightness values.

Figure 7. Green—Comparative analysis of the weighted score differences between different lighting conditions and distance conditions. Note: The letters a and b represent statistically significant differences in the weighted scores between different brightness values.

3.1.6. Purple Recognition

The difference in the weighted score across six levels of brightness was not statistically significant within both the natural light group and the artificial light group (all p > 0.05). Additionally, the differences between the lighting groups did not appear to be statistically significant. Please refer to Figure 8A for more details.

In the 3 m group, the difference in the weighted score among the six levels of brightness was not statistically significant (p = 0.195). In contrast, within the 5 m group, the differences in the weighted score across six levels of brightness were statistically significant (p = 0.004). The pairwise comparisons revealed that the weighted scores for brightness levels 2 and 5 were significantly higher than those for levels 1, 3, 4, and 6. Furthermore, a statistically significant difference was observed between the distance groups at brightness level 5, where the weighted score at 3 m was lower than that at 5 m (p = 0.012). For more detailed information, refer to Figure 8B.

Figure 8. Purple—Comparative analysis of weighted score differences between different lighting conditions and distance conditions. Note: The letters a and b represent statistically significant differences in the weighted scores between different brightness values.

Figure 8. Purple—Comparative analysis of weighted score differences between different lighting conditions and distance conditions. Note: The letters a and b represent statistically significant differences in the weighted scores between different brightness values.

3.1.7. Experimental Summary

The study’s findings within the natural light group revealed no significant differences among the six brightness levels of red and purple (p > 0.05), whereas notable variations were observed for yellow, blue, and green. In the artificial light group, all six brightness levels of red, yellow, blue, and green exhibited significant differences, mirroring the patterns seen with purple in the natural light group. In the natural light group, lower brightness levels of red, yellow, blue, and green were associated with higher scores, whereas higher brightness levels received lower scores. Purple displayed a unique pattern in the natural light group, with brightness level 5 attaining the maximum score and brightness level 6 the minimum score. No significant differences were found in the weighted scores when comparing the natural light and artificial light groups. Collectively, these findings suggest that indoor lighting conditions exert a negligible influence on the visually impaired children’s capacity to discern color brightness.

In general, there were significant differences in brightness levels among the red, yellow, blue, and green hues for both the 3 m and 5 m groups. As the scores for the low-brightness group increased, the scores for the high-brightness group decreased, with this trend being particularly enhanced in the 5 m group. The purple hue did not exhibit significant differences in the 3 m group, but it did in the 5 m group. This suggests that the observation distance has a considerable impact on brightness recognition in children with visual impairments.

3.2. Comparative Analysis of Differences between Lighting and Distance

3.2.1. Legend Explanation

The experimental results aim to present the weighted scores of six brightness levels under different conditions of lighting and distance, with the objective of identifying which brightness levels are more suitable for visually impaired children. To convey these results, a bar chart has been employed for visual representations. The chart is organized such that each brightness level is associated with four groups of data, which are represented from left to right as follows: natural light&3 m, natural light&5 m, artificial light&3 m, and artificial light&5 m. The letters on the chart denote significant differences, thereby highlighting the results that have achieved statistical significance. For more details, please refer to Figure 9, Figure 10, Figure 11, Figure 12 and Figure 13.

3.2.2. Red Recognition

There was no statistically significant difference in the weighted score between the natural light&3 m, the artificial light&3 m, the natural light&5 m, and the artificial light&5 m (all with p > 0.05) groups across all levels of brightness. Additionally, the weighted score of the natural light&3 m group, the artificial light&3 m group, the natural light&5 m group, and the artificial light&5 m group all showed no statistical difference among the six brightness values (all p > 0.05). Please refer to Figure 5 for details.

Figure 9. Red—Comparative analysis of weighted score differences between different groups.

Figure 9. Red—Comparative analysis of weighted score differences between different groups.

3.2.3. Yellow Recognition

When the brightness was set to level 3, there was a statistically significant difference in the weighted score among groups exposed to the natural light&3 m, the artificial light&3 m, the natural light&5 m, and the artificial light&5 m (p = 0.014). The pairwise comparison result indicated that the weighted scores for both the natural light&3 m group and the artificial light&3 m group were higher than those for the natural light&5 m group and the artificial light&5 m group. When the brightness was set to level 6, a statistically significant difference in the weighted score was also observed among these same groups (p = 0.015). However, the pairwise comparison result showed that the weighted score of the natural light&3 m group and the artificial light&3 m group were lower than those for the natural light&5 m group and the artificial light&5 m group. For more details, please refer to Figure 7.

In the natural light&3 m group, a statistical difference was observed in the weighted scores among the six different brightness levels (p = 0.018). The pairwise comparison revealed that the weighted score for brightness level 5 was significantly higher than those for brightness levels 1, 2, 3, 4, and 6. Please refer to Figure 7 for more detailed information.

Figure 10. Yellow—Comparative analysis of weighted score differences between different groups. Note: The letters a and b represent statistically significant differences in the weighted scores between different brightness values. † and ‡ represent statistical differences in weighted scores between different groups.

Figure 10. Yellow—Comparative analysis of weighted score differences between different groups. Note: The letters a and b represent statistically significant differences in the weighted scores between different brightness values. † and ‡ represent statistical differences in weighted scores between different groups.

3.2.4. Blue Recognition

In the natural light&5 m group, a statistical difference was observed in the weighted score among the six levels of brightness (p = 0.029). The pairwise comparison results indicated that the weighted scores for brightness levels 5 and 6 were significantly higher than those for levels 1 to 4. Similarly, in the artificial light&5 m group, the difference in the weighted score among the six levels of brightness was also significant (p = 0.042), with the weighted score for levels 5 and 6 being significantly higher than those for levels 1 to 4. Please refer to Figure 9 for more details.

Figure 11. Blue—Comparative analysis of weighted score differences between different groups. Note: The letters a and b represent statistically significant differences in the weighted scores between different brightness values.

Figure 11. Blue—Comparative analysis of weighted score differences between different groups. Note: The letters a and b represent statistically significant differences in the weighted scores between different brightness values.

3.2.5. Green Recognition

At brightness level 3, there was a statistically significant difference in the weighted score among the natural light&3 m group, the artificial light&3 m group, the natural light&5 m group, and the artificial light&5 m group (p = 0.014). The pairwise comparison results demonstrated that the weighted score for the natural light&3 m group was higher than those of the other three groups.

Within the artificial light&5 m group, the weighted score among the six levels of brightness exhibited statistically significant differences (p = 0.046). The pairwise comparison results indicated that the weighted scores for brightness levels 5 and 6 were significantly higher than those for brightness levels 1, 2, 3, and 4. Please refer to Figure 11 for more details.

Figure 12. Green—Comparative analysis of weighted score differences between different groups. Note: The letters a and b represent statistically significant differences in the weighted scores between different brightness values. † and ‡ represent statistical differences in weighted scores between different groups.

Figure 12. Green—Comparative analysis of weighted score differences between different groups. Note: The letters a and b represent statistically significant differences in the weighted scores between different brightness values. † and ‡ represent statistical differences in weighted scores between different groups.

3.2.6. Purple Recognition

Across all levels of brightness, there were no statistically significant differences in the weighted scores between the natural light&3 m, the artificial light&3 m, the natural light&5 m, and the artificial light&5 m groups (all with p > 0.05). Furthermore, the weighted score of the natural light&3 m group, the artificial light&3 m group, the natural light&5 m group, and the artificial light&5 m group showed no statistically significant differences in the weighted score among six levels of brightness (all with p > 0.05). Please refer to Figure 13 for more details.

Figure 13. Purple—Comparative analysis of weighted score differences between different groups.

Figure 13. Purple—Comparative analysis of weighted score differences between different groups.

3.2.7. Experimental Summary

Overall, the low-brightness group maintained a high level of recognizability in both lighting and distance conditions, with color brightness levels 5 and 6 achieving higher scores. A certain pattern was seen in the sequence of selection for the red, yellow, blue, and green color groups. However, the purple color group displayed a different distribution, with significant weight assigned to scores in the low-brightness, standard, and high-brightness groups.

4. Discussion

The impact of indoor lighting and distance on color recognition among severely visually impaired children constitutes a relatively underexplored area within the realm of visual perception in special education settings. The findings highlight that distance emerges as the primary factor influencing color recognition. Although previous studies have separately investigated color design in the context of both special education schools and the visually impaired, there is a dearth of studies that integrate these two domains. Prior research on color design in special education schools has been deficient in empirical validation of the range of color applications [24,40]. Given the pivotal role of color in the indoor environmental context of schools, which significantly impacts the development of visually impaired children [12,21,23,24], it is imperative to further explore the relationship between color and the indoor environment.

We posit that the higher scores for the low-brightness color under lighting conditions are less related to the gray–white background of the color palette and more associated with the illumination, color temperature, and color rendering index of the lighting. Research conducted by Garnder indicated that neither background color contrast nor differences in chromaticity were effective measures for improving the visual function of visually impaired children [26]. However, Mazza’s findings suggested that attention was typically allocated to the foreground elements. They believed that unless there is a deliberate focus on the background, the contrast in color may not be noticed [50]. Several studies have shown that indoor lighting significantly influences color perception, with color temperature and color rendering index impacting human visual comfort [41,44,47]. Excessively high values of color temperature and color rendering index can cause the low-brightness color to appear brighter, thereby enhancing their visibility. In experiment settings, natural light, compared to artificial light, tends to have a higher color temperature and a stronger color rendering index, which may be one of the reasons for the higher scores of low-brightness colors. However, Hidayetoglu proposed that the cool color tone and the high-brightness color are more effective in aiding spatial orientation in the low-illumination environment [48], which is in direct contrast to our findings. In fact, cool color tones tend to darken the perceived brightness of colors, indicating a similarity between cool color tones and low-brightness colors. Given that our experimental scenarios were computer-generated with complex lighting conditions and involved a diverse set of participants, we believe that the discrepancy in findings may be attributed to the methodology employed in the respective studies. Regarding the specific case of purple, there is a scarcity of literature discussing the phenomenon of the color purple. Therefore, we hypothesize that the outcome of purple hues may be intrinsically linked to the color itself. Purple, being a composite color formed by the combination of red and blue, falls outside the traditional primary colors of light (red, green, blue) and pigment (red, yellow, blue). Furthermore, purple is not commonly observed in large quantities in everyday life. The irregular patterns observed in the scoring for purple warrant additional research to elucidate these findings.

The experimental results concerning distance are consistent with previous studies, confirming that distance is a significant factor affecting color recognition [49]. Statistically significant variations in the weighted score suggested that as distance increases, the recognition rate for high-brightness colors decreases, while low-brightness colors consistently receive higher scores. It is noteworthy that distance did not produce a significant effect on green recognition [49]. Furthermore, other studies have also found that the minimum discriminable size for low-brightness backgrounds exceeds that for high-brightness backgrounds [28], suggesting better recognition for low-brightness colors, which is consistent with our findings. The research by Ying Li further indicated that as observation distance increases, the discriminability of colors decreases, while the perception of black and white by visually impaired individuals relatively improves [9], providing support for our findings.

The findings of this study exhibit both similarities and differences when compared to previous research. We attribute these discrepancies primarily to the differences in experimental settings and the selection of colors for identification. We conducted our experiment in a special education school, a long-term residence for children with visual impairments. We also identified standard hues commonly used in daily life, limiting the range of color samples to the attribute of lightness. In earlier studies, however, the experimental conditions did not replicate the typical living environments of visually impaired children, and the identified colors encompassed not only lightness but also chroma.

In China, the architectural design of special education schools is governed by national standards, as stipulated by the industry code JGJ76-2003. These regulations for various conditions, including classroom dimensions, facilities, lighting, and window arrangements, are all tailored to meet the needs of visually impaired children. Therefore, our conclusions are applicable to special education schools that operate within these standardized conditions. The experimental findings of this study indicate that in indoor settings, distance is the primary factor influencing color recognition for visually impaired children, surpassing the effects of lighting. Future research should focus on the impact of color area, contrast, and shape. Furthermore, it is imperative that the academic community expand its investigative scope to the application of colors in long-term residential environments for visually impaired children.

Our study has several additional limitations. Firstly, the study did not use a control group, which limits our ability to compare the results of the experimental group with those of individuals with lesser visual impairments or normal vision. While we used standardized values to demonstrate differences or similarities with accepted norms, the absence of a control group may affect the overall reliability and interpretability of the findings.

Additionally, the sample size is relatively small and includes a broad age range, which may introduce variability in motor skills and developmental stages. This raises the question of whether the observed color perception characteristics are specific to the studied group or if they can be generalized to a broader population with varying degrees of visual impairment. Special schools, including the one studied, typically serve a diverse student body with varying levels of visual impairment, and our focus on a more severely impaired group may not fully capture this diversity.

Finally, the practical significance of our findings is somewhat constrained by the scope of the study. Although special schools are designed to accommodate a wide range of disabilities, the results of this study may not be applicable to all students within such schools. Future research should consider a larger, more diverse sample and include a control group to enhance the generalizability and practical relevance of the findings.

5. Conclusions

This study examined the effects of lighting and distance on color recognition among children with profound visual impairments. The results indicate that lighting has a relatively minor impact on color recognition, and no significant difference was observed between natural and artificial light. The only notable difference appeared as a slight increase in scores within the low luminance group. Notably, the low luminance group exhibited superior performance under various lighting situations, with some colors showing overlapping characteristics between the high luminance and standard color groups. Conversely, distance had a more substantial effect on color recognition. Specifically, the scores of the low luminance group increased further, revealing a significant difference compared to the results of the high luminance group. Equally noteworthy is the absence of any significant differences in the identification of the color green over varying distances.

One constraint of our study is the limited sample size consisting of only 10 children with low vision, which may influence the generalizability of our findings. However, our findings highlight the importance of considering these factors when addressing the visual needs of children with visual impairments. Considering the crucial role of special education schools for these children, we advocate for the establishment of a specialized color application system within these institutions, separate from general areas. Our analysis indicates that color attributes such as hue, luminance, saturation, area, and shape should adhere to standardized guidelines based on the observed variations in distance. Furthermore, it is important to establish interior lighting requirements, such as illumination levels, color temperature, and color rendering index, in coordination with these color standards. This will guarantee that these requirements correctly correspond to the visual needs of children with visual impairments.