Does artistic training affect color perception? A study of ERPs and EROs in experiencing colors of different brightness

Color is a visual cue that can convey emotions and attract attention


Introduction
As is commonly known, among all visual elements, especially in design and art, color perception plays a crucial role in our perception and understanding of the world around us (Poulin, 2018).The choice of color is an important consideration in the artistic creation and design process as it can influence aesthetic appeal, accessibility, brand identity, communication, and emotion.
Color can capture attention and elicit emotional responses, leading to an aesthetic evaluation of the product and influencing consumer decisions (Fiore et al, 2000).Previous studies (Albers, 2013;Bourges, 1997)have indicated that color is the most influential product feature for customers.Color stimuli with pleasant associations elicit an approach reaction, whereas negative connotations elicit an avoidance response (Berger and Fitzsimons, 2008).Artists are frequently required to control color explicitly in their work, and it is likely that artists have a stronger capacity to notice fundamental color features or are just more accustomed to attending to these lower-level elements as a result of their creative training.Yet much of the current research on artists has focused on painting appreciation and the process of painting rather than basic visual perception (Kottlow et al, 2011;Kozbelt and Seeley, 2007;Long et al, 2011;Ostrofsky et al, 2012Ostrofsky et al, , 2013;;Pang et al, 2013).
There is a growing body of evidence suggesting that art experts may have perceptual advantages at a higher level of visual cognition and a more accurate appreciation of artworks and various painting styles (Biederman and Kim, 2008;Devue and Barsics, 2016).Devue and Barsics discovered that portrait artists were more accurate than non-artists at facial discrimination and recognition, as well as object recognition and mental rotation (Devue and Barsics, 2016).Pang discovered that artistic expertise was related to lower event-related potential (ERP) responses to general visual stimuli, which can be attributed to enhanced brain efficiency as a result of intensive visual arts contemplative practice (Pang et al, 2013).
There is, however, little study on the variations in color perception in artists and non-artists using ERP.According to early research, both artists and non-artists utilized color to differentiate between the negative spatial borders of colorful patterns and the contours of black-and-white pictures (Washburn, 2000).Another study discovered that during a color-naming task, painting majors had more activation in V4, a color-selective region, than controls, and that this activation linked with that in the left ventral lateral prefrontal cortex (Long et al, 2011).
In addition, there is no research to show whether artistic training affects the perception of a single color.Color perception is a highly complex process, and we should not ignore the most basic low-level visual perceptions such as color brightness, which may be influenced by artistic experience.
In recent years, there has been a growing interest in combining electroencephalography (EEG) with the research of color perception, as this method can detect differences in the neural correlates supporting performance.However, most of these studies were done from the point of view of the wavelength of the color, or hue.
A number of empirical studies on the relationship of color to various psychological qualities found that colors with long wavelengths increased the arousal level of the component under examination (Bakker et al, 2013;Buechner et al, 2014;Costa et al, 2018;Elliot, 2015;Olsen, 2010;Stephen and McKeegan, 2010).Additionally, there are also many studies comparing several different colors from a semantic and functional perspective, with red and blue, the two longest and shortest wavelengths, being the two most studied colors (AL-Ayash et al, 2016;Ali, 1972;Baek and Min, 2015;Cabeza and Nyberg, 2000;Caldwell and Jones, 1985;Elliot, 2019;Erwin et al, 1961;Mikellides, 1990;SHEN et al, 1999;Yoto et al, 2007;Bakker et al, 2013).
Brightness is a significant factor which determines color perception; the link between it and emotionality has been little systematically investigated, and a few studies were only focused on the realms of product preference and marketing.For instance, Chen discovered that color attractiveness depends on brightness; high-brightness colors are more appealing and have a greater attraction (Chen et al, 2017).Additionally, Park and his colleagues reported that subjects felt more pleasant in bright conditions (Park et al, 2013).
To precisely control the stimulus variables, we used the Munsell Color System, which has been widely used in various fields, including art, design, and science, as a method of color classification and calibration to differentiate between the varied brightness of the stimuli.Furthermore, to explore the effect of art training on the emotional and attentional perception of color brightness, two groups of subjects were recruited for this study: one with at least three years of professional art training and the other with no art training at all.Both groups of subjects viewed two different brightness intensities of the color stimuli to examine whether and how artistic training experience and color brightness might interact to affect emotion and attention.
This study mainly aims to objectively demonstrate how brightness affects the emotion and attention of color perception and if there are differences in the perception of color brightness between people with and without artistic expertise from a physiological perspective using ERP events and event-related oscillations (EROs) from EEG data.In most of the EEG-based studies (Herrmann et al, 2014;Price, 1997;Zhang et al, 2018b), desired features are derived from the frequency bands of the signal (e.g., delta, theta, alpha, and gamma).On the other hand, P2, N2, and P3 are commonly investigated ERP components in EEG investigations examining the association between vision, emotion, and cognition.
The P2 component is a positive ERP component that may be observed in the anterior and central regions (Luck and Hillyard, 1994).Strong P2 amplitude is associated with more positive emotions (Herbert et al, 2008;Kanske and Kotz, 2007;Schacht and Sommer, 2009;Kissler et al, 2006).Moreover, great interest and attention can cause an increased P2 amplitude (Shedden and Nordgaard, 2001;Eason, 1981;Mangun et al, 1986).Additionally, P2 amplitude decreases as cognitive stress increases (Allison and Polich, 2008;Deeny et al, 2014;Ghani et al, 2020;Horat et al, 2016).N2 appears to be tightly correlated with the cognitive processes of perception and selective attention and is related to instinctive stimulus recognition (Patel and Azzam, 2005).Preferred stimuli, which were highly linked to heightened attention, interest, and pleasant sentiments, were linked to an elevated anterior N2 (Guo et al, 2016;Vogel and Machizawa, 2004).
The P3 is always considered as an indicator of how cognitive attention and workload are distributed.Enhanced P3 is closely connected with increased attention (Miller et al, 2011) and associated with decreased cognitive workload (Allison and Polich, 2008;Goodin et al, 1983).In addition, the emotional aspects of the stimuli are also connected to P300; stimuli with more emotional features can elicit a larger P3 amplitude than neutral stimuli (Munoz and Martin-Loeches, 2015;Radilova, 1982;Radilova et al, 1983;Radilova, 1989).
Moreover, time-frequency analysis of EEG activity within specific frequency bands allows the extrapolation of information that is not available using ERP analysis (Cohen, 2014;Morales and Bowers, 2022;Munneke et al, 2015).Specifically, delta (1-4 Hz) and theta (4-8 Hz) waves are strongly associated with P3 ERP components and attention (Demiralp et al, 1999).Additionally, enhancement of theta waves has also been reported to be involved in positive mood changes (Aftanas and Golocheikine, 2001;Sammler et al, 2007).Alpha waves (8-12 Hz) are related to neural mechanisms of sophisticated cognitive processes and are highest in the occipital and frontal cortices, where they are assumed to signify both relaxed awareness and attentiveness (Başar et al, 2001).For the gamma band (beyond 30 Hz), many studies have found that it is associated with emotion and that strong emotional fluctuations cause an increase in gamma band energy (Balconi and Lucchiari, 2008;Keil et al, 2001).
Based on previous studies on color perception (Chen et al, 2017;Park et al, 2013), we have the following hypotheses.
Hypothesis 2: For the ERO analysis, we expected that higher-brightness color stimuli could elicit power for the delta, theta, and alpha bands.The delta and theta waves have been observed to be pronounced components correlated with the P3 wave (Demiralp et al, 1999), and alpha waves have been thought to indicate both relaxed awareness and attention (Başar et al, 2001).
Hypothesis 3: We also hypothesized that people with art training experience would have a sharper perception of color brightness, as art experts may be more accustomed to focusing on these lower-level features on the basis of their long-term art training compared with non-training people.
Moreover, we also computed the correlation coefficients between the behavioral data and the EEG data to examine if there is an intrinsic correlation between behavioral data (emotion values and reaction time), years of artistic training, ERP components, and each frequency band of interest.

Participants
Paid volunteers were recruited from undergraduate and master's students at Dalian University of Technology, and those with moderate to severe depressive or anxiety tendencies were excluded by the Self-Rating Scale for Depression (SDS) and the Self-Rating Scale for Anxiety (SAS).Finally, fortyfour subjects with right-handedness were recruited, of whom 22 belong to the artistic training group All participants had a normal or corrected-to-normal vision, and they were free from color blindness and color weakness.For the artistic training group, all subjects were art or design majors and had professional artistic training of at least three years (artistic training years: M = 6.45,SD = 2.84, range = 3-12).Those in the non-artistic group were also strictly controlled, all of whom were from non-art-related majors and had never received hobby classes or professional training related to art and painting.The experiment was conducted in accordance with the Declaration of Helsinki and was approved by the Medical Engineering Ethics Review Group of Dalian University of Technology, Dalian, China.

Stimuli
The experiment was programmed using E-Prime 3.0.All color stimuli in the experiment were selected from the Munsell color stereo and displayed at a size of 1024 × 798 pixels.The Munsell system is a color-coding system based on the visual perception characteristics of color from a psychological perspective.This system has clear standards for the three elements of color: brightness, saturation, and hue.There are nine levels of brightness in this color system, from 1 to 9, indicating decreasing brightness in descending order.Two sets of color stimuli, with high and low brightness, were selected from the 10 main side profiles of the color spectrum based on a rigorous brightness scale.These two categories are one-to-one, corresponding to the same color saturation and hue but with a 6-step difference in brightness levels.Fig. 1 a shows how the two groups of stimuli were selected in the yellow and red intermediate hue (5YR hue) and Fig. 1 b shows all the stimuli colors used in this experiment.

Design and Procedure
During the experiment, the participants were asked to observe two types of color stimuli and assess the emotional value of the stimuli by keystroke.Each participant completed 132 trials (66 trials for high-brightness color and 66 for low-brightness color separately).As shown in Fig. 2, each trial began with a fixation cross (1200 ms), then participants were shown a low-brightness stimulus image (1200 ms), followed by a gray screen for keystroke feedback.The numbers 1-3 represent a gradual increase in the emotional value of the color.Numbers 1, 2, and 3 represent negative emotions, neutral emotions, and positive emotions, respectively.After the participants had performed the keystroke feedback, they were presented with a new fixation cross (1200 ms) and the high-brightness stimulus picture (1200 ms) and continued with another round of keystroke feedback.The high-brightness stimuli were presented alternately with the low-brightness stimuli.There is a brief practice session where the color stimuli are chosen differently from those used before the formal experiment.The formal experiment would not be started until the participant had been fully briefed on the procedure.

EEG recordings and analysis
Participants sat around 100 cm from a 50 Hz CRT monitor (59 × 33cm).Electroencephalogram (EEG) data were recorded from 65 scalp electrodes using a 10-20 system in an electrically shielded and darkened room.An ANT Neuro EEG amplifier was used to record EEG signals with a 1000 Hz sampling rate.The electrode impedance for each electrode for each participant was below 5κ℧, and the EEG was online referenced to the CPz.

Preprocessing pipeline
The following preprocessing steps were included for the offline analysis that was conducted using EEGLAB 2020 (Delorme and Makeig, 2004).EEG data were first resampled to 500 Hz, and the data were re-referenced to the average reference, excluding M1, M2, and EOG channels.Afterward, the  data were filtered with a notch filter that was to remove 50 Hz line noise; a high-pass FIR filter with 0.1 Hz and a low-pass FIR filter with 30 Hz with the default parameters in EEGLAB were also used to filter the EEG data.Thirdly, independent component analysis (ICA) based on ICASSO software (Himberg and Hyvarinen, 2003) was applied to correct the EEG by removing artifactual independent components (ICs) that were associated with eye blinks and eye movement activities.After that, the artifact-corrected EEG data were segmented from 200 ms before stimulus onset to 1000 ms after stimulus onset, and segments were baseline-corrected by subtracting the mean amplitude of the time window of -200 ms to 0 ms from all time points.Bad epochs containing eye blinks and other artifacts were automatically discarded.To remove excessive magnitudes and to ensure that each subject had a similar number of trials, a maximum magnitude (50-100µV ) was set for each subject's data acquisition quality, resulting in 80.95% ± 12.06% of trials remaining for each subject.Additionally, we performed a wavelet filter transform on the ERP data to improve the signal-to-noise ratio as applied in the previous studies (Cong et al, 2012;Samar et al, 1992;Qi, 1993).Finally, the remaining epochs were averaged for each experimental condition and each subject.

ERP quantification
ERPs of interest were further analyzed using the ERP ERO toolbox (Zhang, 2021;Zhang et al, 2020a) in the MATLAB 2019b environment.The amplitude and latency scores of ERP components of interest were quantified across different electrodes within a specific time window (P2: FCz and Cz, from 155 ms to 220 ms; N2: FCz, from 220 ms to 310 ms; and P3: CPz, from 310 ms to 530 ms).
We computed the time-frequency representations (TFRs) of ERP data based on the ERP ERO toolbox using the continuous wavelet transform (Tallon-Baudry and Bertrand, 1999;Zhang et al, 2018aZhang et al, , 2020a,b),b).The complex Morlet wavelet was adopted for the mother wave, and both the bandwidth and center frequency were 1.Additionally, TFR is baseline corrected using the subtraction method.That is, we subtracted the mean power of the baseline period (from -200 ms to 0 ms) from all time points and from all frequency bins.To quantify the oscillatory dynamics associated with color perception at different brightnesses, we focused on four frequency bands.Specifically, we measured the delta band (1-4 Hz) from 50-800 ms, the theta band (4-8 Hz) from 50-300 ms, the alpha band (8-12 Hz) from 50-300 ms, and the low gamma band (30-48 Hz) from 50 ms to 300 ms.

Statistical analysis
To examine the possible interactions between artistic training experience and brightness, repeated measurement variance of analysis (rm-ANOVA) with artistic training experience as a between-subject factor (art vs. control) and with the brightness (low vs. high) as a within-subject factor was employed.
Partial eta-squared (η 2 p ) and Cohen's d were used to describe effect sizes (Gignac and Szodorai, 2016;Sawilowsky, 2009;Sullivan and Feinn, 2012).When Mauchly's sphericity test indicated a violation of sphericity, the degrees of freedom were corrected using the Greenhouse-Geisser method.Finally, Pearson correlation coefficients between behavioral performance (RTs and emotional values), artistic training years, ERP components (the amplitude and latency of P2, N2, and P3), and EROs (delta, theta, alpha, and lower gamma frequency bands) were calculated to investigate their relationships.
The two-tailed method was used to test the significance of the correlation.
We found no significant interactions or main effects in the RT analysis, but we did observe a marginally significant main effect for brightness (F (1,43) = 3.400, p = 0.072, η 2 p = 0.075, See Table .1 and Fig. 3 b).

P2 component
Mean amplitude: P2 was quantified at FCz and Cz ( Fig. 4 a and b) from 155 ms to 220 ms based on the previous studies (Luck, 2014;Luck and Hillyard, 1994) and inspection of the grand topography wave and topography (Fig. 4 d).Statistical analysis results (Fig. 4 g) showed that there was no interaction effect between brightness and group (F (1,43) = 0.274, p = 0.603, η 2 p = 0.006), and there was also no difference between the art group and control group (F (1,43) = 0.046, p = 0.832, η 2 p = 0.096).

Discussion
The present study aimed to investigate the relationship between artistic training experience and the electro-cortical correlates of processing different brightness colors based on EEG data.We used ERP and ERO to examine the effects of different brightnesses of color on art-trained and non-art-trained individuals and recorded both the subjects' scores and reaction times for assessing the emotional value of color.All color stimuli in this experiment were selected according to the Munsell color system scale of brightness, saturation, and hue.
As Hypothesis 1, high-brightness color stimuli yielded more positive emotion ratings and stronger P2, which is consistent with previous research on P2 components.Enhanced P2 amplitude is also strongly associated with more positive emotions (Herbert et al, 2008;Kanske and Kotz, 2007;Kissler et al, 2006;Schacht and Sommer, 2009), more attention (Eason, 1981;Mangun et al, 1986;Shedden and Nordgaard, 2001), and less cognitive stress (Allison and Polich, 2008;Deeny et al, 2014; Ghani effects.However, no such significant findings were detected for the control group.
301 Also, we found a significant between-group effect for P2 latency, with long latency for the art 302 group, which may be related to art training making single colors less visually appealing to the artists.
303 Meanwhile, there was a significant between-group effect for N2 latency, with the art group having a short latency.The peak latency of the N2 effect is the time when a person has enough visual information to respond (Augustin et al, 2011;Zhang and Damian, 2009;Schmitt et al, 2000), which may indicate that artistic training improves executive and attentional control (Chamberlain and Wagemans, 2015;Chamberlain, 2017;Chamberlain et al, 2019;Perdreau and Cavanagh, 2014) and that artists are more sensitive to the distinction between colors of different brightness and can make judgments more quickly.
Compared to the P2, an early visual component, the N2, and the P3 are associated with high cognitive function.Perhaps the rationale provided above explains why interaction effects were found only in the analysis of N2 and P3 amplitude and the between-group differences that emerged in P2 and N2 latencies, which supports previous research findings that, while artists may have perceptual advantages in certain high-level aspects of visual cognition, these advantages are not observed in low-level aspects of visual perception (Kozbelt and Ostrofsky, 2018;Kozbelt and Seeley, 2007).There are two aspects of art creation and art appreciation to support the art instruction experience that impact color brightness perception.On the one hand, the subjects in the art group have years of painting experience; they need to fine-tune the brightness of colors when painting and creating and therefore have a more sensitive and accurate perception of color brightness.On the other hand, long-standing aesthetic practices make them more inclined to pay more attention to bright colors.
A good artistic creation requires a companion and a protagonist, and with the protagonist often in a high-brightness color, the artist tends to be able to distinguish the main focus and focus more quickly on the appreciation of the artwork.
The analysis of the behavioral data also yielded the same results, i.e., high positive affective scores for bright colors, which reinforces the analysis of the EEG data and is consistent with previous studies (Pang et al, 2013;Chen et al, 2017).Additionally, the response time for the high-brightness stimuli was short, which is consistent with earlier research where feedback on the preferred stimuli took less time to respond (Bamford et al, 2015;Chen and Bargh, 1999;Guo et al, 2016).The automatic predisposition in cognition will drive people to approach pleasure and avoid unpleasant stimuli; stimuli that can elicit pleasure feelings will drive a faster response (Bamford et al, 2015;Chen and Bargh, 1999).
In contrast to ERP analysis, no significant group differences were found in the analysis of behavioral data.This is partly due to the limitations of the small group sample size and the fact that neuroimaging methods are more accurate and may discover variations in brain correlates that support performance even when no obvious behavioral differences exist (Gauthier et al, 2000).
The results of the analysis of the energy in the frequency bands were equally consistent with Hypothesis 2. High-brightness stimuli triggered much more energy in the delta and theta bands than in low-brightness colors.Significant differences were obtained in frontal and central regions in the delta band and in frontal, central, and occipital regions in the theta band.Given that both bands are associated with attention (Demiralp et al, 1999) and that high energy in the theta band tends to be involved in positive emotions (Sammler et al, 2007;Aftanas and Golocheikine, 2001), this result could imply that high-brightness colors have a more positive emotional value and trigger more attention.
Consistent with previous research, high-brightness stimuli in the alpha band derived significantly more energy, which is similar to the findings that subjects felt more pleasant and had increased alpha power in high-brightness conditions (Park et al, 2013).It is also in line with the previous study (Guo et al, 2016), viewing stimuli with high visual aesthetics raised alpha power considerably.This finding may suggest that bright colors have a high aesthetic value and are more likely to make the audience feel relaxed and cheerful.
In addition, the energy of high-brightness stimuli is also significantly increased in the low gamma band, which, according to previous studies, is strongly correlated with emotion (Balconi and Lucchiari, 2008;Keil et al, 2001).These results may represent the fuller emotional value of high-brightness colors.
Finally, in the correlation analysis, we found a negative correlation between reaction time and P2 latency as well as a negative correlation between P2 and N2 latency.Meanwhile, the number of years of art training was positively correlated with P2 latency and negatively correlated with N2 latency.
We can reason that art training experience can diminish the early perception of all color stimuli but at the same time sharpen the artist's differentiation and evaluation of colors and ultimately shorten the reaction time.Moreover, we observed a positive correlation between theta and low-band gamma energy and behavioral data emotion levels, which may be explained by the strong association between these two bands and emotion (Liu et al, 2022) and the formation of a subjective preference (Lindsen et al, 2010).
Our research suggests that artists may show a perceptual advantage in simple color perception, in addition to art appreciation and creativity.This means that art training is beneficial for color perception and can be popularized in other areas, such as in a study showing that formal art observation training enhances the visual diagnostic skills of medical students (Naghshineh et al, 2008).In addition, this may help artists or designers understand the gap between themselves and their audience and choose colors that better match the emotional needs of their audience in the process of creating art.The limitation of this study, which is difficult to avoid, is that it cannot distinguish whether the differences between the two groups are due to artistic training or individual artistic talent.Because people with superior motor coordination, visual perception, memory, or concentration may self-select into the arts (Chamberlain, 2018), the group differences could be related to some other unknown and uncontrolled background variable other than the difference in the level of artistic training or talent.
The relatively small group size may not enable control of such variables by randomized sampling.
Overall, the EEG may serve as a useful biomarker for further research into the effects of experience with art training on color cognitive function.In our future research, we will investigate the impact of artistic training experience on the visual perception of other elements of color, saturation, and hue, as well as aesthetic preferences and emotional perceptions of different color combinations.

Conclusions
The

Fig
Fig.1(a) The process of selecting two stimuli from the 5YR phases of Munsell color stereo.The selection is done strictly on the scale of hue, saturation, and brightness of this color system.High-brightness stimuli and low-brightness stimuli correspond one by one, with a difference of 6 degrees in the brightness scale while keeping the hue and saturation the same.(b) Presentation of the two groups of stimuli in the 10 most dominant hues of the color stereoscope.

Fig. 2
Fig.2Example of the experiment for two adjacent trials.The low-brightness stimuli alternated with high-brightness stimuli, and the brightness difference was uniform for each set of stimuli.Each trial begins with a 1200 ms fixation, followed by a 1200 ms stimulus presentation and free-time keystroke feedback.
Fig.3(a) Emotional ratings for high and low brightness color stimuli for the art and non-art groups.Numbers -1, 0, and 1 represent emotional values: negative, neutral, and positive, respectively (corresponding to numbers 1-3 in the experiment).(b) Reaction time for high-brightness and low-brightness color stimuli for the art and non-art groups, respectively.

Fig. 4
Fig. 4 a-f illustrates the grand average wave and topography corresponding to P2, N2, and P3 separately, and Fig. 4 g-i shows the latency and amplitude of the three components under different conditions.

Fig. 5 a
Fig.5 a and Fig.5 b show the grand average TFRs over the Fz, Cz, Pz, and Oz electrodes, as well as the topographic distribution corresponding to the delta, theta, alpha, and low gamma bands.The selected time windows are 50-800 ms for the delta band, 50-300 ms for the theta band, 50-300 ms for the alpha band, and 50-200 ms for the low gamma band, respectively.

Fig. 6
Fig.6 displays the Pearson correlation coefficients among behavioral data (emotion values and reaction times), ERP data (amplitude and latency of the three components, P2, N2, and P3), EROs (the Fig. 5 (a) Time-frequency representations for four experimental conditions at Fz, Cz, Pz, and Oz electrodes; (b) Topographies of the delta, theta, alpha, and low gamma bands; and the selected time windows are 50-800 ms, 50-300 ms, 50-300 ms, and 50-200 ms, respectively.

Fig. 6
Fig. 6 Pearson correlation coefficients among behavioral data (emotion values and reaction times), artistic training years, ERP components (amplitude and latency of the three components, P2, N2, and P3), and EROs (the delta, theta, alpha, and lower gamma frequency bands at Fz, Cz, Pz, and Oz).
current study investigated the neural responses of people with artistic training and the general population in relation to emotion and attention to colors of different brightness using ERPs and EROs.The results indicate that brighter colors trigger more positive emotions and stronger attention in both groups.Additionally, artistic training diminishes the early perception of color, while having a positive effect on top-down visual perception, making the artist more sensitive to the distinction between colors of different brightness.Acknowledgments.Scholarship from the China Scholarship Council (No. 201906060241), Fundamental Research Funds for the Central Universities [DUT20LAB303&DUT20LAB308] at Dalian University of Technology in China, and the Science and Technology Planning Project of Liaoning Province (no.2021JH1/10400049) all contributed to the funding of this study.The authors would like to extend their gratitude to Dr. Guanghui Zhang and Dr. Xiaoshuang Wang for help with data processing; Dr. Guanghui Zhang and Dr. Johanna Silvennoinen for their help with manuscript revision; and Prof. Ma and Prof. Cong for their help with their suggestions of experiment design.Declaration of competing interest.None of the authors has a potential conflict of interest to declare.

Table 1
Emotional ratings and reflection times for both high-brightness and low-brightness stimuli for art and control groups

Table 3
Statistical analysis of the delta,theta, alpha, and low gamma band energies at four electrodes(Fz,Cz, Pz, and Oz)