Sage Journals: Discover world-class research

Abstract

This study explored whether the color of letters could influence letter discrimination task performances and whether this effect of color could be modulated by processing level (global vs. local) and attention level of color (color-attended vs. color-unattended). We used the Navon letters in red, green, or white as stimuli at a relatively small (Experiment 1) or large visual angle (Experiments 2, 3, and 4). Each experiment included two tasks: color-attended task in which participants were told to respond only to target letters in a designated color; color-unattended task in which color was task-irrelevant. Experiment 1 found that the responses to red stimuli were significantly faster than those to the other color stimuli in the color-attended task. In Experiment 2, the same pattern occurred only at the local level in the color-attended task. Experiments 3 and 4 further controlled the brightness and chroma of stimuli and the results replicated the enhancement effect of red at the local level in the color-attended task and demonstrated an interference effect of red and green in the color-unattended task. These results suggested that red facilitated letter discrimination at the local processing level, reflecting the effect of avoidance motivation evoked by red on cognition and behavior which was consistent with color-in-context model. Moreover, this study found that the effect of color was modulated by attention level of color, and the interference effect of color in the color-unattended task confirmed that the color effect might mainly arise from controlled processes but not automatic processes.

Keywords

color Navon letters visual angle local/global processing attention level

Introduction

Color is a fundamental aspect of visual inputs and has some physiological and psychological functions (Elliot & Maier, 2012; Goldstein, 1942; Hanada, 2018). For example, longer wavelength colors such as red are deemed arousing, whereas shorter wavelength colors such as green are deemed calming (Kwallek et al., 1997). Blue background enhances performance on creative tasks (Mehta & Zhu, 2009). Viewing red may weaken intellectual performance (Elliot et al., 2007, 2009; Elliot & Maier, 2007; Zhang & Han, 2014).

The effect of color on detail-oriented cognitive processing is the focus of the present study. Several researches have provided the basis for further investigation. For instance, Kwallek and Lewis (1990) assessed the effects of a red (vs. a green and a white) office environment on workers’ proofreading productivity. They found that participants in the red office made the fewest errors. Mehta and Zhu (2009) explored the effect of red background screen (vs. blue and white) on cognitive task performances (memory task, proofreading task, the creative uses test, the Remote Associates Test) and discovered that viewing red can enhance performances on memory and proofreading tasks which can be categorized as detail-oriented cognitive tasks.

According to the color-in-context model (Elliot & Maier, 2012), color can carry specific meanings based on learned associations and biologically based proclivities. Perception of color can evoke automatic evaluative processes and further produce approach/avoidance motivation that may influence observers’ behavior and cognition. Moreover, the psychological meanings of color and its effects are contextual. Take red as an example. As the focus of most researches on psychological effect of the color, it is often associated with danger or failure and produces an avoidance-based motivation in achievement contexts; whereas in affiliation contexts, it is associated with sex and produces an approach-based motivation (Elliot et al., 2007, 2009, 2010, 2011; Elliot & Niesta, 2008; Mehta & Zhu, 2009). Based on this theory, some researchers argued that avoidance motivation activated by red made individuals become alert and risk-averse and hence can promote the processing of details (Mehta & Zhu, 2009). The idea that avoidance motivation facilitated detail processing was also supported by other researches (Derryberry & Reed, 1998; Finucane, 2011; Förster et al., 2006; Friedman & Förster, 2005, 2010).

This study would make further efforts to clarify the effect of the color red on detail processing. Unlike prior studies using the memory exercise or the proofreading as detail-oriented tasks (Kwallek & Lewis, 1990; Mehta & Zhu, 2009), this research conducted letter discrimination task. In addition, color manipulation in this study was different from previous work. Aforementioned research manipulated the color of office environment or computer background screen to evoke individual emotion and motivation, but we manipulated the color of target stimuli. For one thing, we hoped to replicate the previous result in a different situation in which the color changes across trials to gain insight into the fast effect of the color on cognition and behavior. For the other thing, we also attempted to investigate whether two factors, processing level (global vs. local) and attention level of color (color-attended vs. color-unattended), can modulate the effect of color.

Maier and colleagues (2008) proposed that red could evoke avoidance-based processes leading to a narrow attentional focus. In their Experiment 2, subjects were informed to look at color blocks (red vs. gray) and then complete a visual match task in which they were given a target figure and two comparison figures (one comparison figure was similar to global figure of the target, the other was similar to local elements of the target) in each trial and judged which comparison figure was more similar to the target figure. The authors found that subjects chose the local option in more trials in red condition, reflecting a narrow attentional focus. However, the result that red induces local processing tendency was not replicated by Zhang and Han (2014). Their study also employed the visual match task, but they failed to find the facilitation of red on local processing although the result showed that viewing color red worsened participants’ intellectual performance. The authors proposed that the visual match task may be influenced by response strategy. For example, if participants tend to make their judgments based on the same rule in all trials (e.g., all based on global figure), their test scores were likely to be zero or full mark and ended up with a null result. Thus, this study employed another common paradigm-Navon letter to explore the modulation of the global/local processing on the color effect (Navon, 1977). If red could induce local processing tendency, we would anticipate that the enhancement effect of red on letter discrimination performance would occur mainly at the local level but not at the global level.

In addition, this study also investigated whether another factor, that is, the attention level of color, modulated the effect of color. We designed two kinds of tasks: one was a color-attended task in which participants were asked to respond to target letters in a designated color and to ignore stimuli in the other color, the other was a color-unattended task in which participants were asked to respond to target letters whatever color they are. The main difference between the two tasks is the additional requirement of attending and selecting color (Daffner et al., 2012). We wonder whether the color effect is influenced by color-selective attention. If color has an automatic effect, then we would predict similar effect of color for both types of tasks. However, if the effect of color arises from controlled processes, then we would predict that the significant effect of color is mainly in the color-attended task. Thus, this study can further test whether color effects would be automatic or controlled if color effects did exist.

Unlike previous studies that mostly used blue as the contrast color of red (Elliot et al., 2011; Maier et al., 2008, 2012; Mehta & Zhu, 2009), we chose green as its contrast color, with white or gray served as a control. In daily life, green is a kind of common color which is often associated with positive meanings such as safety, life, and hope, providing us with a comfortable, calm feeling. For instance, green represents “go” in the traffic light, and green passage means an efficient or guaranteed way. In light of the color-in-context model, seeing green things might elicit approach motivation and positive emotion and then affect cognition and behavior. Lichtenfeld et al. (2012) tested the effect of color (green, red, blue, and gray) on creativity, and the result indicated that green facilitated creativity. It was approach motivation elicited by green that played an important role in creativity task, the authors held. Then, can green affect letter discrimination performance? This study was designed to address this question. If the green effect exists, given that approach motivation could broaden attention scope (Förster et al., 2006), we expect that the green effect will also be modulated by the global/local processing level.

In sum, we hypothesized that red could activate avoidance motivation, induce local processing tendency, make viewers more vigilant, and then promote the processing of details at the local level. In contrast, approach motivation would be elicited by green which can facilitate the global processing. Also, we expected that the attention level of color can modulate the effect of color, with the enhancement effect of color occurring mainly in the color-attended task.

Experiment 1

Method

Participants

Eighteen undergraduate students (M_age = 21.4 years, SD = 2.4; 15 women, three men) took part in the experiment. All participants had normal or corrected-to-normal vision and were not red–green or blue–yellow color deficient. Each participant signed an informed consent form prior to participating. This research was approved by the Institutional Review Board of Capital Normal University.

Stimuli and apparatus

Stimuli were large letters composed of small letters, including T composed of Hs, T composed of Fs, L composed of Hs, L composed of Fs, H composed of Ts, H composed of Ls, F composed of Ts, and F composed of Ls. These letters were chosen because they were made up of horizontal and vertical lines, and they were all consonants (Gable & Harmon-Jones, 2010). T and L were target stimuli, and H and F were nontarget stimuli. When T and L were presented as large letters, participants needed to respond to targets at the global level of compound stimuli. When T and L were presented as small letters, participants needed to respond to targets at the local level of compound stimuli (see Figure 1). Each stimulus appeared 72 times. It was presented in red (R:255, G:0, B:0) for a third of these times, in green (R:0, G:255, B:0) for another third, and in white (R:0, G:0, B:0) for the remaining times. Red and green are contrast colors; white is an achromatic color used as baseline. Stimuli were displayed against a black background on the computer screen of 17 inches whose resolution was 1024 × 768 pixels. The distance from eye to screen was about 60 cm. The visual angle of small letters was 0.53° × 0.72° (width × height), and that of large letters was 2.63° × 3.58° (width × height).

Figure 1.

Experimental design and stimuli used in Experiments 1 and 2: (A) Experimental design: In the color-attended condition, participants only responded to stimuli in particular color. In the red-attended blocks, the red “T” and “L” were targets; in the green-attended blocks, the green “T” and “L” were targets; and in the white-attended blocks, the white “T” and “L” were targets. In the color-unattended condition, regardless of the color, participants responded to all “T” and “L.” (B) Stimulus sets. The targets were always the letters “T” and “L.” The global “T” and “L” were composed of small “H” and “F,” whereas local “T” and “L” constituted large “H” and “F.”

Procedure

Before the experiment to create an achievement context, participants were informed that their performance would be compared with others. The experiment employed a within-subjects design; each participant completed nine blocks of 64 trials each. Among all the blocks, six blocks employed the color-attended task and others employed the color-unattended task, with the order of tasks being counterbalanced across subjects. In the color-attended task, participants distinguished T and L at the global or local level for red stimuli only (in two red-attended blocks, each block included 32 target trials and 32 nontarget trials), or green stimuli only (in two green-attended blocks), or white stimuli only (in two white-attended blocks; see Figure 1A); as for the nontarget stimuli, participants did nothing about it except waiting for it to disappear automatically after 1500 ms poststimulus; the target color order with those blocks was counterbalanced with a Latin square. In the color-unattended task, participants responded to T and L, no matter which color it was. In each trial, a fixation cross was presented at the center of the screen for 500 ms, and then a stimulus was present for a maximum of 1500 ms, but was terminated upon a button press. Participants were asked to make a response to the target correctly as fast as possible, and pressed the right key of mouse in response to the letter T and left key in response to letter L (the assignment of response keys was counterbalanced across participants). They could take a short break between blocks to avoid fatigue. Stimuli were presented in pseudo-random order in each block, and there were no more than two trials with the same stimuli in succession.

Results

Analysis of response times (RTs) included correct trials that were within 3 SDs of each participant’s mean RT. Table 1 presents the mean RTs to stimuli for each condition. A 2 (Task type: color-attended task vs. color-unattended task) × 3 (Color: red vs. green vs. white) × 2 (Level: local vs. global) repeated-measured analysis of variance (ANOVA) was conducted on RTs. The main effect of level was significant, F(1, 17) = 37.01, p < .001, $η_{p}^{2}$ = .69, with faster responses at the global level compared with those at the local level. Significant main effects were also found for task type, F(1, 17) = 29.07, p < .001, $η_{p}^{2}$ = .63, and color, F(2, 34) = 4.81, p = .014, $η_{p}^{2}$ = .22. Moreover, results revealed a significant interaction between task type and color, F(2, 34) = 3.84, p = .031, $η_{p}^{2}$ = .18. Simple effect analysis determined that participants responded faster to red stimuli than other colors in the color-attended task but not in the color-unattended task (see Figure 2).

Table 1.

Mean RTs (ms) (M ± SD) for Each Condition in Experiment 1.

	Red		Green		White
Task type	Local	Global	Local	Global	Local	Global
Color-attended	654 ± 73	612 ± 104	711 ± 76	656 ± 133	710 ± 93	654 ± 107
Color-unattended	642 ± 84	554 ± 91	647 ± 84	568 ± 95	634 ± 93	558 ± 103

Note. RT = response time.

Figure 2.

Behavioral data of Experiments 1 and 2: (A) Mean response times of Experiment 1. (B) Mean RTs of Experiment 2. (C) Mean accuracies of Experiment 1. (D) Mean accuracies of Experiment 2.

Mean accuracy on each condition is exhibited on Table 2. An ANOVA performed on accuracy showed a significant main effect of level, F(1, 17) = 15.03, p = .001, $η_{p}^{2}$ = .47, with better performance at the global level compared with the local level. Other effects were not detected (ps > .05).

Table 2.

Mean Accuracies (M ± SD) for Each Condition in Experiment 1.

	Red		Green		White
Task type	Local	Global	Local	Global	Local	Global
Color-attended	0.95 ± 0.04	0.98 ± 0.02	0.94 ± 0.04	0.97 ± 0.04	0.95 ± 0.05	0.97 ± 0.05
Color-unattended	0.94 ± 0.04	0.98 ± 0.02	0.96 ± 0.04	0.96 ± 0.04	0.97 ± 0.04	0.98 ± 0.03

In conclusion, Experiment 1 showed significant main effects of level on RT and accuracy. Subjects responded faster and more accurately to targets at the global level than at the local level, showing a global preference. This experiment also found a significant interaction between color and task type. Participants respond faster to red stimuli than other colors only in the color-attended task. Inconsistent with our expectation that participants’ responses to red stimuli would have an advantage compared with green and white stimuli only at the local level, the present results did not reveal an interaction between color and level. Because the difficulty of responding to letter discrimination was affected by the size of visual angle (Navon, 1981), we suspected that the absence of interaction between color and level might be due to relatively small visual angle of stimuli used in Experiment 1. Consequently, we planned to enlarge the visual angle in Experiment 2.

Experiment 2

In this experiment, the visual angle was extended from 0.53° × 0.72° to 1.05° × 1.43° at the local level, from 2.63° × 3.58° to 5.25° × 7.15° at the global level. We expected that the RTs for red letters were much shorter than those for white letters at the local level in the color-attended task, and the response to red letters would be slower than those to white letters in the color-unattended task.

Method

Participants

Fifteen undergraduate students (M_age = 21.3 years, SD = 1.9; 12 women, three men) took part in the experiment. All participants had normal or corrected-to-normal vision and were not red–green or blue–yellow color deficient. Each participant signed an informed consent form prior to participating.

Stimuli and apparatus

Stimuli and apparatus in Experiment 2 were the same as Experiment 1, except the visual angle. In Experiment 2, the visual angle was extended from 0.53° × 0.72° to 1.05° × 1.43° at the local level, from 2.63° × 3.58° to 5.25° × 7.15° at the global level.

Procedure

The design and procedure in Experiment 2 were the same as Experiment 1.

Results

Analysis of RTs included correct trials that were within 3 SDs of each participant’s mean RT. Table 3 presents the mean RTs to stimuli for each condition. A 2 (Task type: color-attended task vs. color-unattended task) × 3 (Color: red vs. green vs. white) × 2 (Level: local vs. global) repeated-measured ANOVA was conducted on RTs. The results revealed significant main effects of task type, F(1, 14) = 31.40, p < .001, $η_{p}^{2}$ = .69, and level, F(1, 14) = 41.84, p < .001, $η_{p}^{2}$ = .75, and an obvious interaction between task type and color, F(2, 28) = 4.65, p = .018, $η_{p}^{2}$ = .25. More importantly, the interaction among task type, color, and level was observed, F(2, 28) = 7.21, p = .003, $η_{p}^{2}$ = .34. Simple effect analysis showed that participants responded faster to the red letters than the other colors at the local level in the color-attended task, but no significant difference between green letters and white letters was found (see Figure 2). In the color-unattended task, there was no significant difference among these colors both at the local level and at the global level. Mean accuracy on each condition is exhibited on the Table 4. An ANOVA performed on accuracy showed no significant main effect or interaction.

Table 3.

Mean RTs (ms) (M ± SD) for Each Condition in Experiment 2.

	Red		Green		White
Task type	Local	Global	Local	Global	Local	Global
Color-attended	719 ± 113	688 ± 132	761 ± 135	682 ± 138	789 ± 129	702 ± 122
Color-unattended	683 ± 114	587 ± 99	671 ± 113	601 ± 119	656 ± 105	575 ± 96

Note. RT = response time.

Table 4.

Mean Accuracies (M ± SD) for Each Condition in Experiment 2.

	Red		Green		White
Task type	Local	Global	Local	Global	Local	Global
Color-attended	0.96 ± 0.05	0.97 ± 0.03	0.97 ± 0.05	0.99 ± 0.03	0.98 ± 0.03	0.98 ± 0.02
Color-unattended	0.98 ± 0.03	0.98 ± 0.03	0.97 ± 0.05	0.99 ± 0.03	0.98 ± 0.03	0.99 ± 0.02

To sum up, Experiment 2 revealed that the responses to red letters were faster than those to green and white letters only at the local level of the color-attended task. No significant difference was exhibited among three colors in color-unattended task.

There are some limitations in Experiments 1 and 2. First, we only used the RGB model to manipulate color, but a better method of LCh color model should be conducted to control the lightness and chroma of the color letters (Elliot & Maier, 2012). Second, the sample size was small and a large proportion of participants in our experiments were female. Therefore, in Experiment 3, we planned to use the LCh color model to control the brightness and increase the sample size to obtain more generalizable results.

Experiment 3

In this experiment, the visual angle was the same as Experiment 2 (1.05° × 1.43° at the local level, 5.25° × 7.15° at the global level). We further controlled the brightness of the stimulus according to the LCh color model which defines a color space in terms of three parameters: lightness (similar to brightness), chroma (similar to saturation), and hue (Fairchild, 2005). We expected that the enhancement effect of color red only occurred at the local level in the color-attended task.

Method

Participants

To detect the main effect of color with small-to-medium effect ( $η_{p}^{2}$ = .03), as power analysis with G-Power 3.1.9.2 statistical software (Faul et al., 2007) suggested 54 participants for achieving 80% power (alpha level of .05, repeated ANOVA), we therefore collected data from 54 participants (M_age = 23.9 years, SD = 3; 30 women, 24 men). All participants had normal or corrected-to-normal vision and were not red–green or blue–yellow color deficient. Each participant signed an informed consent form prior to participating.

Stimuli and apparatus

Stimuli and apparatus in Experiment 3 were the same as Experiment 2, except the brightness of stimuli. In Experiment 3, we measured the spectral properties of color stimuli via Tru-Image 2D Imaging Colorimeter from PHOTO RESEARCH in a room where the illumination was 80 lux. They come with Windows-based VideoWin 3 software to analyze the data (Fiske, 2015). The brightness of chromatic colors and achromatic color was equated (red: LCh [58.1, 85.1, 27.0]; green: LCh [59.6, 70.9, 148.1]; gray: LCh [58.1, 11.5, 259.1]). Equated here means functionally equivalent (i.e., within five units on each relevant parameter; Elliot et al., 2007).

Procedure

The design and procedure in Experiment 3 were the same as Experiments 1 and 2.

Results

Analysis of RTs included correct trials that were within 3 SDs of each participant’s mean RT. Table 5 presents the mean RTs to stimuli for each condition. A 2 (Task type: color-attended task vs. color-unattended task) × 3 (Color: red vs. green vs. gray) × 2 (Level: local vs. global) repeated-measured ANOVA was conducted on RTs. The results revealed significant main effects of task type, F(1, 53) = 36.23, p < .001, $η_{p}^{2}$ = .41, and level, F(1, 53) = 121.43, p < .001, $η_{p}^{2}$ = .70, and a significant interaction between task type and color, F(2, 106) = 7.03, p = .001, $η_{p}^{2}$ = .12. More importantly, the interaction among task type, color, and level was significant, F(2, 106) = 15.02, p < .001, $η_{p}^{2}$ = .22. Simple effect analysis showed that in the color-attended task, participants responded faster to the red letters than to the gray letters at the local level but not at the global level (see Figure 3). In the color-unattended task, participants responded slower to the red letters and green letters than to the gray letters at the global level, and responded slower to the red letters than to the green letters and gray letters at the local level.

Table 5.

Mean RTs (ms) (M ± SD) for Each Condition in Experiment 3.

	Red		Green		Gray
Task type	Local	Global	Local	Global	Local	Global
Color-attended	649 ± 114	608 ± 112	668 ± 129	601 ± 104	674 ± 129	606 ± 106
Color-unattended	606 ± 100	540 ± 97	588 ± 94	544 ± 97	580 ± 98	523 ± 91

Note. RT = response time.

Figure 3.

Behavioral data of Experiment 3: (A) Mean response times of Experiment 3. (B) Mean accuracies of Experiment 3.

Mean accuracy on each condition is exhibited on Table 6. An ANOVA performed on accuracy showed a significant main effect of level, F(1, 53) = 16.52, p < .001, $η_{p}^{2}$ = .24, with better performance at the global level compared with the local level. Other effects were insignificant (ps > .05).

Table 6.

Mean Accuracies (M ± SD) for Each Condition in Experiment 3.

	Red		Green		Gray
Task type	Local	Global	Local	Global	Local	Global
Color-attended	0.94 ± 0.05	0.96 ± 0.04	0.95 ± 0.06	0.97 ± 0.04	0.95 ± 0.06	0.97 ± 0.04
Color-unattended	0.95 ± 0.04	0.97 ± 0.03	0.94 ± 0.07	0.98 ± 0.04	0.96 ± 0.04	0.98 ± 0.03

In sum, Experiment 3 revealed that the responses to red letters were faster than those to gray letters at the local level of the color-attended task, showing the enhancement effect of red color. In addition, the experiment also found the interference effect of color red and green; participants responded slower to red and green letters than to the gray letters in the color-unattended task.

Experiment 4

Although Experiment 3 replicated the results of Experiment 2 that red would promote local processing in color-attended condition, it is difficult to determine whether the current results are due to categorical hue or instead due to chroma as we did not match the saturation of red and green. In Experiment 4, we further matched the brightness and saturation of red and green. We aimed to replicate the results of the Experiment 3 with same design and procedure to reduce the probability that the observed effects were due to chroma.

Method

Participants

We used the results of Experiment 3 (54 samples) to determine the appropriate sample size for Experiment 4. Considering both publication bias and sample bias (Anderson et al., 2017), we applied the effect size of the three-way interaction from Experiment 3 ( $η_{p}^{2}$ = .22) and found that the sample size to achieve a power of 0.8 (α planned = .05) was 20. Hence, we recruited 25 participants, and four participants were excluded from the analysis because they did not understand the procedure of the study. Finally, data of 21 participants were analyzed and reported in Experiment 4 (11 females and 10 males; M_age = 23.8 years, SD = 2.1). All participants had normal or corrected-to-normal vision and were not red–green or blue–yellow color deficient.

Stimuli

In Experiment 4, we matched the chromatic colors on brightness and saturation (red: LCh [54, 76.6, 29.1]; green: LCh [59.6, 70.9, 148.1]), and the achromatic color was equated on brightness (gray: LCh [58.1, 11.5, 259.1). The brightness and saturation of stimuli were measured via Tru-Image 2D Imaging Colorimeter.

Procedure

The design and procedure in Experiment 4 were the same as Experiment 3.

Results

Analysis of RTs included correct trials that were within 3 SDs of each participant’s mean RT. Table 7 presents the mean RTs to stimuli for each condition. A 2 (Task type: color-attended task vs. color-unattended task) × 3 (Color: red vs. green vs. gray) × 2 (Level: local vs. global) repeated-measured ANOVA was conducted on RTs. The results revealed significant main effects of task type, F(1, 20) = 6.04, p < .05, $η_{p}^{2}$ = .23, and level, F(1, 20) = 41.74, p < .001, $η_{p}^{2}$ = .68. Furthermore, the predicted interaction among task type, color, and level was observed, F(2, 40) = 4.99, p < .05, $η_{p}^{2}$ = .20. Simple effect analysis showed that in the color-attended task, participants responded faster to the red letters than to the gray letter at the local level; in the color-unattended task, participants responded slower to the red letters and green letters than to the gray letters at the global level and the local level (see Figure 4).

Table 7.

Mean RTs (ms) (M ± SD) for Each Condition in Experiment 4.

	Red		Green		Gray
Task type	Local	Global	Local	Global	Local	Global
Color-attended	596 ± 130	564 ± 138	618 ± 158	555 ± 131	634 ± 145	554 ± 111
Color-unattended	584 ± 115	518 ± 118	582 ± 127	533 ± 127	556 ± 112	501 ± 120

Note. RT = response time.

Figure 4.

Behavioral data of Experiment 4: (A) Mean response times of Experiment 4. (B) Mean accuracies of Experiment 4.

Mean accuracy on each condition is exhibited on Table 8. An ANOVA performed on accuracy showed a significant main effect of level, F(1, 20) = 25.19, p < .001, $η_{p}^{2}$ = .56, with better performance at the global level compared with the local level. Other effects were not significant (ps > .05).

Table 8.

Mean Accuracies (M ± SD) for Each Condition in Experiment 4.

	Red		Green		Gray
Task type	Local	Global	Local	Global	Local	Global
Color-attended	0.94 ± 0.04	0.95 ± 0.05	0.93 ± 0.06	0.96 ± 0.04	0.93 ± 0.05	0.98 ± 0.03
Color-unattended	0.93 ± 0.06	0.97 ± 0.03	0.94 ± 0.05	0.97 ± 0.03	0.95 ± 0.04	0.99 ± 0.03

Experiment 4 replicated the results of Experiment 3, which showed that the enhancement effect of color red only occurred at the local level of the color-attended task, and also found the interference effect of color red and green in the color-unattended task.

Discussion

In this study, we used Navon letters to test whether the color of letters influences performance on letter discrimination task and whether the effect of color is modulated by global/local processing and attentional level of color. The results of four experiments indicated an advantage of global processing, which means that subjects responded much faster to global letters than to local letters, consistent with previous studies (Navon, 1977, 1981; Navon & Norman, 1983). Navon and Norman (1983) proposed that the global advantage could be held in 2° to 17.5° of visual angle. Therefore, we observed global precedence when we enlarged the visual angle of letters in Experiments 2 to 4. In addition, RTs were longer on the color-attended task than the color-unattended task in four experiments, reflecting additional requirement of attending and selecting color in the color-attended task.

This study showed that color red could facilitate or impair letter discrimination compared with white or gray, depending on the attention level of color. Specifically, four experiments all showed that the enhancement effect of red occurred only in the color-attended task, and Experiments 3 and 4 also revealed the interference effect of red only in the color-unattended task, reflecting that the effects of red were modulated by attention level. As stated in the introduction, the enhancement effect of red on cognitive task performance was reported by several previous studies. For example, Mehta and Zhu (2009) used different cognitive tasks to explore the effect of color red (vs. blue and white) and discovered that viewing red background screen enhanced performances on memory and proofreading tasks which were categorized as detail-oriented cognitive tasks. A similar result was also reported by Kwallek and Lewis (1990) using the proofreading task and manipulating the color of office. Researchers (Elliot et al., 2007; Kwallek & Lewis, 1990; Mehta & Zhu, 2009) argued that red as a kind of cue indicates implicit danger which evokes avoidance motivation, making viewers more vigilant and risk-averse. In that case, individuals mobilize limited cognitive resources to attend to the current situation so that they have a better chance to avoid harm and failure. Therefore, red might narrow the scope of attention and facilitate detail or local processing, which is crucial to individuals in the aspect of evolution. Given that the letter discrimination task, similar to proofreading task, was a detail-oriented cognitive task, the result of Experiments 2, 3, and 4 that the enhancement effect of red occurred only in the local level supported the above view. However, one might argue that participants gave faster responses to red targets than to white or gray targets just because they could identify red letters more quickly to white or gray letters. Although we agreed on this view, it cannot explain our finding that the enhancement effect of red emerged only in the local processing level.

With respect to the role of global/local processing level, the result of Experiment 1 was different from that of Experiments 2, 3, and 4. Experiment 1 showed the enhancement effect of red regardless of processing level, whereas Experiments 2 to 4 showed the enhancement effect of red only at the local processing level. Because stimuli and procedure of Experiment 2 were the same as those of Experiment 1 except the visual angle of stimuli, and Experiments 3 and 4 further controlled the brightness and chroma of stimuli. Consequently, we thought that the inconsistency of results could be due to the difference of visual angle. For compound large, global letters made up of smaller, local letters, the global processing involves local element grouping and identification of global targets but the local processing involves selection of individual’s local elements and identification of local targets (Han et al., 1999). We assume that a local processing tendency and the more vigilant state induced by red can speed up not only the selection of individual’s local elements but also the identification of targets. Moreover, the enhancement effect of red on target identification might decrease with the increase of the visual angle. Therefore, when the visual angle was relatively small (Experiment 1), red facilitated the local processing by accelerating the selection of individual’s local elements and identification of local targets, and also facilitated the global processing by speeding up the identification of global targets. However, when the visual angle of targets increased to a certain extent (Experiments 2–4), the enhancement effect of red on global target identification disappeared, and thus no effect of red was found in the global level but the effect still existed in the local level.

It should be noticed that, unlike previous research manipulating color as a between-subject factor (Kwallek & Lewis, 1990; Mehta & Zhu, 2009), the color of target letters changed across trials in this study. The enhancement effect of red still found in the present design might suggest that this effect could rapidly produce and disappear. Nevertheless, as in the color-attended task each block required participants to respond only to stimuli with one color, participants would suppress the responses to the stimuli with other colors as far as possible. Thus, in this design, although color changed across trials, color manipulation was more like between-block one which was conducive to observing the effect of color.

Unexpectedly, in the color-unattended task, Experiments 3 and 4 demonstrated interference effect of red. We thought that this interference effect in the color-unattended task might stem from unconscious attention bias to color stimuli. Some researchers argued that red carries the message that a present stimulus was important and worthy of attention (Buechner et al., 2014; Elliot & Maier, 2012). Participants usually pay more attention to red stimuli. On the contrary, under the color-unattended task, color was a task-irrelevant dimension. More attention to red stimuli induced slower letter discrimination response. Furthermore, red letters cannot automatically activate an avoidance motivation or it can only activate a weak avoidance motivation because of unconscious attention on color dimension. Therefore, in the color-unattended task, the influence of red on letter discrimination was mainly manifested on its interference effect. In addition, we suspected that the white stimuli that were brighter than other stimuli in Experiments 1 and 2 were the reason for the lack of interference effect of red compared with white in these two experiments.

Apart from red, we also explored the effect of green on letter discrimination performance. The result did not show an enhancement effect for green stimuli in four experiments, both at the global level and at the local level. Inconsistent with our expectation, this result might suggest that, when using compound letter stimuli, green cannot facilitate global processing although green is often associated with safety, life, and hope, providing us with comfortable and calm feeling. According to previous theory, watching green might elicit approach motivation (Elliot et al., 2007; Lichtenfeld et al., 2012), which could broaden attention scope (Förster et al., 2006). However, this study showed that both larger attention scope and comfortable feeling seemed unable to promote global processing. One possibility is that the presentation time of green stimuli or the area occupied by green stimuli might affect the effect of green. In the present research, color served as stimuli’s feature. Compared with color block operation in prior researches, the color of letters was presented for a much shorter time, and the area occupied by green letters was much smaller in the present study. It is possible that colors with obvious emotional meanings, like red, a small area of presentation, or a short presentation time suffices to produce corresponding color effect. But for colors with less obvious emotional meanings, like green, maybe it takes a large area of exposure or a long presentation time to produce the color effect on cognitive task. However, this hypothesis cannot explain the result of Kwallek and Lewis (1990) that no enhancement effect of green on proofreading performance was found when green was the color of office, which had a large area of presentation and a long presentation time. In addition, in the color-unattended condition of our Experiments 3 and 4, participants responded slower to green letters than gray letters. We thought that the mechanism of the interference effect of green was similar to those of red.

Conclusion

This study used Navon letters to test whether color (red vs. green vs. white) of letters influences letter discrimination task performance and whether the effect of color is modulated by global/local processing and attentional level of color. Experiment 1 revealed that participants responded faster and more accurately to targets at the global level than at the local level and respond faster to red stimuli than to other colors only in the color-attended task. Employing same stimuli, procedure, and larger visual angle of stimuli, Experiment 2 showed that the responses to red letters were faster than those to green and white letters only at the local level of the color-attended task. Experiments 3 and 4 further controlled the brightness and chroma of stimuli, and the results replicated the enhancement effect of red at the local level in the color-attended task and demonstrated an interference effect of red and green in the color-unattended task. These results suggested that the effect of color was modulated by attention level of color. Red facilitated letter discrimination at the local processing level, reflecting the effect of avoidance motivation evoked by red on cognition and behavior which was consistent with color-in-context model. Moreover, this study found the interference effect of color in the color-unattended task and confirmed that the effect of color might mainly arise from controlled processes but not automatic processes.

Footnotes

Authors’ Note

This research was approved by the Institutional Review Board of the University.

Author Contributions

Meng Sun and Xiaorong Zhang contributed equally.

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Natural Science Foundation of China (31470980 and 31571143).

ORCID iDs

Qin Zhang

Lixia Cui

References

Anderson

S. F.

Kelley

Maxwell

S. E.

(2017). Sample-size planning for more accurate statistical power: A method adjusting sample effect sizes for publication bias and uncertainty. Psychological Science, 28(11), 1547–1562.

Buechner

V. L.

Maier

M. A.

Lichtenfeld

Schwarz

(2014). Red-take a closer look. PLOS ONE, 9(9), Article e108111.

Daffner

K. R.

Zhuravleva

T. Y.

Sun

Tarbi

E. C.

Haring

A. E.

Rentz

D. M.

Holcomb

P. J.

(2012). Does modulation of selective attention to features reflect enhancement or suppression of neural activity? Biological Psychology, 89(2), 398–407.

Derryberry

Reed

M. A.

(1998). Anxiety and attentional focusing: Trait, state and hemispheric influences. Personality and Individual Differences, 25(4), 745–761.

Elliot

A. J.

Maier

M. A.

(2007). Color and psychological functioning. Current Directions in Psychological Science, 16(5), 250–254.

Elliot

A. J.

Maier

M. A.

(2012). Color-in-context theory. Advances in Experimental Social Psychology, 45, 61–125.

Elliot

A. J.

Maier

M. A.

Binser

M. J.

Friedman

Pekrun

(2009). The effect of red on avoidance behavior in achievement contexts. Personality and Social Psychology Bulletin, 35(3), 365–375.

Elliot

A. J.

Maier

M. A.

Moller

A. C.

Friedman

Meinhardt

(2007). Color and psychological functioning: The effect of red on performance attainment. Journal of Experimental Psychology: General, 136(1), 154–168.

Elliot

A. J.

Niesta

(2008). Romantic red: Red enhances men’s attraction to women. Journal of Personality and Social Psychology, 95(5), 1150–1164.

10.

Elliot

A. J.

Niesta Kayser

Greitemeyer

Lichtenfeld

Gramzow

R. H.

Maier

M. A.

Liu

(2010). Red, rank, and romance in women viewing men. Journal of Experimental Psychology: General, 139(3), 399–417.

11.

Elliot

A. J.

Payen

Brisswalter

Cury

Thayer

J. F.

(2011). A subtle threat cue, heart rate variability, and cognitive performance. Psychophysiology, 48(10), 1340–1345.

12.

Fairchild

M. D.

(2005). Color appearance models (2nd ed.). New York: John Wiley.

13.

Faul

Erdfelder

Lang

A.-G.

Buchner

(2007). G* Power 3: A flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39(2), 175–191.

14.

Finucane

A. M.

(2011). The effect of fear and anger on selective attention. Emotion, 11(4), 970–974.

15.

Fiske

(2015). Image quality and metrology. Information Display, 31(5), 12–15.

16.

Förster

Friedman

R. S.

Özelsel

Denzler

(2006). Enactment of approach and avoidance behavior influences the scope of perceptual and conceptual attention. Journal of Experimental Social Psychology, 42(2), 133–146.

17.

Friedman

R. S.

Förster

(2005). Effects of motivational cues on perceptual asymmetry: Implications for creativity and analytical problem solving. Journal of Personality and Social Psychology, 88(2), 263–275.

18.

Friedman

R. S.

Förster

(2010). Implicit affective cues and attentional tuning: An integrative review. Psychological Bulletin, 136(5), 875–893.

19.

Gable

P. A.

Harmon-Jones

(2010). Late positive potential to appetitive stimuli and local attentional bias. Emotion, 10(3), 441–446.

20.

Goldstein

(1942). Some experimental observations concerning the influence of colors on the function of the organism. Occupational Therapy, 21, 147–151.

21.

Han

Humphreys

G. W.

Chen

(1999). Parallel and competitive processes in hierarchical analysis: Perceptual grouping and encoding of closure. Journal of Experimental Psychology: Human Perception and Performance, 25(5), 1411–1432.

22.

Hanada

(2018). Correspondence analysis of color–emotion associations. Color Research & Application, 43(2), 224–237.

23.

Kwallek

Lewis

C. M.

(1990). Effects of environmental colour on males and females: A red or white or green office. Applied Ergonomics, 21(4), 275–278.

24.

Kwallek

Woodson

Lewis

C. M.

Sales

(1997). Impact of three interior color schemes on worker mood and performance relative to individual environmental sensitivity. Color Research & Application, 22(2), 121–132.

25.

Lichtenfeld

Elliot

A. J.

Maier

M. A.

Pekrun

(2012). Fertile green: Green facilitates creative performance. Personality and Social Psychology Bulletin, 38(6), 784–797.

26.

Maier

M. A.

Elliot

A. J.

Lichtenfeld

(2008). Mediation of the negative effect of red on intellectual performance. Personality and Social Psychology Bulletin, 34(11), 1530–1540.

27.

Mehta

Zhu

R. J.

(2009). Blue or red? Exploring the effect of color on cognitive task performances. Science, 323(5918), 1226–1229.

28.

Meier

B. P.

D’agostino

P. R.

Elliot

A. J.

Maier

M. A.

Wilkowski

B. M.

(2012). Color in context: Psychological context moderates the influence of red on approach-and avoidance-motivated behavior. PLOS ONE, 7(7), Article e40333.

29.

Navon

(1977). Forest before trees: The precedence of global features in visual perception. Cognitive Psychology, 9(3), 353–383.

30.

Navon

(1981). The forest revisited: More on global precedence. Psychological Research, 43(1), 1–32.

31.

Navon

Norman

(1983). Does global precedence reality depend on visual angle? Journal of Experimental Psychology: Human Perception and Performance, 9(6), 955–965.

32.

Zhang

Han

(2014). Experience reverses the red effect among Chinese stockbrokers. PLOS ONE, 9(2), Article e89193.

The Effect of Color on Letter Discrimination Can Be Modulated by the Global/Local Processing and Attention Level

Abstract

Keywords

Introduction

Experiment 1

Method

Participants

Stimuli and apparatus

Procedure

Results

Experiment 2

Method

Participants

Stimuli and apparatus

Procedure

Results

Experiment 3

Method

Participants

Stimuli and apparatus

Procedure

Results

Experiment 4

Method

Participants

Stimuli

Procedure

Results

Discussion

Conclusion

Footnotes

Authors’ Note

Author Contributions

Declaration of Conflicting Interests

Funding

ORCID iDs

References