Abstract
This study experimentally investigated the impacts of major design variables of a liquid crystal display (LCD) home appliance control interface on interface text legibility. The home appliance of interest was a washing machine. The experimental setup simulated a real-world washing machine usage condition challenging in terms of visual information processing. Independent variables were surface glass transmittance (25%, 50%, and 75%), font brightness (60%, 75%, and 100% at 25, 44, and 90 cd/m²), and font height (3, 5, and 7 mm). Response time (RT) and proportion of errors (PE) were employed as the dependent measures. Regression analysis was conducted to analyze data. The results showed that: (a) an increase in the font height resulted in decreases in RT and PE, and (b) increasing glass transmittance and font brightness reduced RT and PE for the 3 and 5 mm font heights. The study findings would inform the design of home appliance control interfaces.
Due to their widespread applications, the legibility of text on liquid crystal displays (LCDs) has been extensively studied, particularly in everyday products such as computer monitors and tablets computers. Key research topics include luminance contrast (Fletcher et al., 2009; Ko, 2017), brightness combination (Li et al., 2022), font characteristics (Dobres et al., 2018; Hasegawa et al., 2008), all of which are crucial design elements for improving text legibility on these screens.
LCDs are also utilized as control interfaces in home appliances. Compared to computer displays, home appliance LCD interfaces have a few unique characteristics. First, despite the increasing complexity of the functions in modern home appliances, the screen area for displaying information remains relatively small. Second, The LCD interfaces of home appliances usually feature a surface glass panel on top of the LCD, unlike computer displays. This surface glass can affect text luminance by altering light diffusion and reflection (Cakir, 1980), thereby impacting text brightness and legibility. Third, the LCD interfaces of home appliances are typically positioned lower than the standing eye height of users and are often on the front side of the products. This setup poses challenges in text legibility due to the large eye-to-text distance and the downward viewing angle, which tilts and distorts characters (Cai & Green, 2009; Cai & Li, 2014).
Designing LCD interfaces for home appliances is an important research topic since many people use these products daily and are impacted by their design. Despite this importance, research on the text legibility of LCD interfaces in home appliances has been limited.
To address this research gap and inform the design of LCD interfaces for home appliances, this study experimentally examined the effects of surface glass transmittance, font brightness, and font height on text legibility under conditions representing real-world use. The target home appliance of this study was washing machines.
In this study, 48 participants (24 males and 24 females) with normal or corrected-to-normal vision, without visual impairments participated in a controlled laboratory experiment. The experimental environment was controlled to maintain consistent lighting conditions using artificial indirect lighting, which provided an ambient illumination of 300 lux.
The experimental apparatus included a tablet computer, a surface glass panel, a holder, and a height-adjustable stand. To represent the washing machine control interface, the LCD panel was positioned 984 mm above the ground. The surface glass panel, which is replaceable and serves as the LCD interface’s surface glass, was attached to the front interface using the holder. This surface glass panel varied in glass transmittance, with one of three different panels being selected based on the experimental condition.
Participants engaged in the experiment in an upright stance, maintaining a distance of 347 mm from the display to read the presented text. Participants were exposed to a total of 54 four-syllable Korean words, which were chosen for their complexity similar to that of commonly used terms on washing machine control interfaces. A san serif font was used for the presented words. To ensure the validity of the results, the conditions were randomly arranged.
Independent variables utilized in the experiment were glass transmittance (set at 25%, 50%, and 75%), font brightness (60%, 75%, and 100%, corresponding to luminance values of 25, 44², and 90 cd/m²), and font height (3, 5, and 7 mm). Glass transmittance is defined as the ratio of the luminance of the area where the text is presented on the LCD panel surface with the attached glass to the directly measured luminance of text on the screen. Font height was based on the most used values in the United States and Korean home appliances.
Response time (RT) and proportion of errors (PE) were utilized as dependent measures. RT refers to the duration from the onset of stimulus presentation to the completion of reading the word. PE is defined as the proportion of syllables mispronounced in a word consisting of four syllables. Responses exceeding 10 seconds were recorded as incorrect.
Polynomial regression was applied to analyze the collected data. In the case of PE (ranging from 0 to 1), the logit function of the Generalized Linear Model (GLM) was used. This was done to appropriately set the range of the dependent variable and to enhance model fit. The regression lines were calculated using the least squares method.
Figure 1 visually represents the effects of the three independent variables and their interactions on PE, along with the regression equation. Noteworthy findings from the regression plots are as follows: (a) PE generally decreases as font height increases; (b) for font heights of 3 mm or 5 mm, an increase in glass transmittance results in a decrease in PE; (c) when the font height is 3 or 5 mm, PE typically declines with higher font brightness; and (d) at a font height of 7 mm, PE remains consistently very low (mean value under 0.1) regardless of variations in glass transmittance and font brightness.
Regression plots and regression equation of PE variation with glass transmittance and font brightness. • – 60% font brightness, ■ – 75% font brightness, ◆ – 100% font brightness. (a) 3 mm font height, (b) 5 mm font height, (c) 7 mm font height.
The regression plots and the regression equation for RT is shown in Figure 2. The regression plots showed that: (a) as font height increased, RT generally decreased; (b) under the 7 mm font height condition, a consistent RT of approximately 2 seconds was observed across all font brightness and glass transmittance levels; and (c) when the font height was 3 or 5 mm, RT decreased as glass transmittance and font brightness increased
Regression plots and regression equation of RT variation with glass transmittance and font brightness. • – 60% font brightness, ■ – 75% font brightness, ◆ – 100% font brightness. (a) 3 mm font height, (b) 5 mm font height, (c) 7 mm font height.
The results indicate that font height has a high impact on users’ text legibility, affecting both PE and RT. This finding is consistent with results from other studies investigating text legibility (Boyarski et al., 1998; Huang et al., 2009). Notably, under the 7 mm font height condition, the large font height is thought to have facilitated easier character recognition regardless of other design variables.
Font brightness affects the luminance contrast between the background and the text, and changes in luminance contrast have been shown to significantly impact text legibility (Fletcher et al., 2009; Ko, 2017). A decrease in font brightness can lead to a reduction in luminance contrast between the background and the text, as observed in the 3 and 5 mm conditions, which demonstrated an increase in PE and RT. The luminance of the text displayed on the LCD screen affects legibility (Cai, 2008). Surface glass, when applied to the display surface, reduced this luminance, thereby impacting legibility. Based on the experimental results of this study, the use of surface glass with high glass transmittance and high font brightness is necessary to improve text legibility under conditions of small font height.
The existing text design guidelines provided by the United States Nuclear Regulatory Commission (U.S. NRC; O’Hara & Fleger, 2020) and the U.S. Military Standard MIL-STD-1472F (1999) recommend font heights ranging from 2.5 to 5 mm for viewing distance of 0.71 m. However, the results of this experiment indicate that most participants were unable to recognize text with a font height of 3 mm. This discrepancy appears to be influenced by the design variables of home appliances and the large viewing angle resulting from the height difference between the display and the user’s eyes. Similar to the findings of Cai and Li (2014), which suggest that change in viewing angles negatively impact reading performance.
In this study, experiments were conducted under a single illuminance condition. Given that variations in illuminance can affect legibility on LCDs (Lee et al., 2008), future research is recommended to determine the impact of these design variables on text legibility across a broader range of illuminance conditions.
This study differs from previous research by simulating real-life scenarios in which text information is read on consumer electronic devices. By incorporating factors that impair text legibility in real-life scenarios, the results of this study could provide a stringent criterion to be considered when designing control interfaces for household appliances.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
