The Munsell color system specifies colors based on hue, chroma (saturation) and brightness (value).
We used this system to classify color of images in this paper. The left image shows the combinations of brightness and chroma for the color Red. The right image illustrates how the Munsell system divides the hue space into 10 different colors.

The Munsell color system specifies colors based on hue, chroma (saturation) and brightness (value). We used this system to classify color of images in this paper. The left image shows the combinations of brightness and chroma for the color Red. The right image illustrates how the Munsell system divides the hue space into 10 different colors.

Source publication
Article
Full-text available
Many lab studies have shown that colors can evoke powerful emotions and impact human behavior. Might these phenomena drive how we act online? A key research challenge for image-sharing communities is uncovering the mechanisms by which content spreads through the community. In this paper, we investigate whether there is link between color and diffus...

Similar publications

Article
Full-text available
Little is known on how to encourage effortful (rather than effortless) conservation behaviors, and prior research investigated only single (rather than multiple) message appeals in terms of their persuasive power in promoting pro-environmental intentions. The current study uses a framework from evolutionary psychology to propose and test a blend of...

Citations

... Visual aesthetics, which influences human perceptions of visual attributes [40,46] and diffusion of visuals [1,50], have been extensively explored in prior research using various indicators, such as colors, composition, visual complexity, and quality [25,31,40,41,47]. In the assessment of AIGIs and other visuals, color and quality are two crucial factors in considering visual realism [5,8,33]. ...
... All variables demonstrated acceptable intercoder reliability. Following content analysis, we conducted a descriptive analysis using R. We separated images from two platforms for analysis and conducted a series of chi-square tests and independent t-tests to compare the results, as shown in Appendix A. 1 In addition to quantitative analysis, we conducted qualitative analysis on two high-level salient features emerging from both theory and our collected data-human-related surrealism and aesthetic professionalism. To better understand and categorize these features, we assessed a sample of images exhibiting each salient feature, using an open coding strategy. ...
Preprint
Full-text available
Advances in generative models have created Artificial Intelligence-Generated Images (AIGIs) nearly indistinguishable from real photographs. Leveraging a large corpus of 30,824 AIGIs collected from Instagram and Twitter, and combining quantitative content analysis with qualitative analysis, this study unpacks AI photorealism of AIGIs from four key dimensions, content, human, aesthetic, and production features. We find that photorealistic AIGIs often depict human figures, especially celebrities and politicians, with a high degree of surrealism and aesthetic professionalism, alongside a low degree of overt signals of AI production. This study is the first to empirically investigate photorealistic AIGIs across multiple platforms using a mixed-methods approach. Our findings provide important implications and insights for understanding visual misinformation and mitigating potential risks associated with photorealistic AIGIs. We also propose design recommendations to enhance the responsible use of AIGIs.
... Surface roughness caused by damage increases the reflection of light diffusion, intensifying the brightness of the damaged area (Jung and Weon, 2013;Sin et al., 2020). Therefore, the bright part (4/11 steps) of the value scale was used as the corrosion indicator according to the Munsell color system (Bakhshi and Gilbert, 2015), and the top 36 % of the greyscale was defined as the dissolution area (%). The grey value (0-225) was calculated using ImageJ. ...
... Using computer vision, images are automatically analyzed and classified according to their content (LeCun et al. 2015). Because of its automatic nature, computer vision quickly identifies many content features in a large and varied image set in a short time (Bakhshi and Gilbert 2015;Bianco et al. 2017). However, despite significant advances in this direction, in our study, we found a significant decrease in the accuracy of results when the respondent's position is unstable, which imposes certain limitations on the conduct of experiments and requires further improvements in computer vision models. ...
Article
Full-text available
Artificial intelligence technologies are improving the marketing toolkit, making it possible to process large amounts of data faster and more efficiently than ever before. Machine learning, a subset of AI, uses algorithms that can predict which ads will be most effective in specific situations, allowing for optimized ad targeting. This research explores the issues of coevolution and distribution of machine and human intelligence in various social practices, including marketing and advertising. The authors describe the key approaches to studying the visual component of advertising and suggest revising traditional methods of analyzing advertising messages. The tracking of biometric data combined with AI-based methods that capture human emotions while viewing video content is proposed as a promising direction for such analysis. This paper presents the results of a pilot study based on analytical face-tracking technology using AI, where the subject of the experiment was the analysis of video fragments that may have an impact on the emotional state of the viewer. The AI software platform used was Amazon Rekognition, and the results show that AI analytics provide the ability to track the level of audience engagement in perceiving video content, which helps to improve communication effectiveness. This allows the use of AI to make recommendations for the development of more directed and engaging advertising messages.
... Brand sincerity positively associates with blue and orange. Blue compositions denote peacefulness (Hsieh et al. 2018), relaxation (Bakhshi and Gilbert 2015), and security (Labrecque and Milne 2012), which may evoke pleasant feelings and a sense of calmness (Yu and Egger 2021) and result in greater perceived sincerity. Brightness, linked with the perception of calmness, cleanness, and warmth, also increases the perception of brand sincerity (Gorn et al. 1997). ...
... Despite colours' heterogeneous affective effects in varied situational or cultural contexts, universal colour-emotion associations were revealed across 30 nations worldwide (Jonauskaite et al., 2020). Research on the emotional role of colour has also spread across website media design, marketing and educational settings (Bakhshi & Gilbert, 2015;Khattak et al., 2018;Plass et al., 2020). Compared to the gray scaled poetic learning material, the colourful one could be more capable of eliciting emotions such as pleasantness or liking (Berlyne, 1970;Uzun & Yildirim, 2018) in low empathy learners, initiating their affective processing and fostering their aesthetic experience with the poem. ...
Article
Full-text available
In this study, we investigated whether a colourful design would foster the aesthetic appreciation of Chinese classic poetry, and the role of learners’ empathy trait in this process. Middle school students (N = 159) were assigned to one of four conditions in a 2 (Colour Design: Colourful vs. Grayscale) × 2 (Empathy Trait: High vs. Low) between-subject design. Results showed that learners with high empathy attained higher scores on comprehension test and aesthetic appreciation ratings than learners with low empathy. Moreover, a significant interaction of the two factors on aesthetic emotion and evaluations ratings was found, indicating that aesthetic ratings were higher in the high empathy group than in the low empathy group with grayscale learning materials, whereas aesthetic ratings were comparable between the two groups with the colourful design. The results indicated that a colourful design can foster low empathy learners’ aesthetic appreciation, and suggested that individual differences should be taken into consideration in instructional design in the multimedia learning of poetry.
... Color can influence learners' emotion (Bakhshi & Gilbert, 2015) and cognition (Pan, 2010). Combining color stimulation with visual experience can trigger learning imagination and associations (Elliot et al., 2007), bring different feelings and mental touch, as well as impress learners. ...
Article
Full-text available
The study aimed to examine how emotional design and digitalization/mobilization of textbooks influence students’ emotion, cognitive load, and cognitive performance. Research samples were 147 undergraduate students enrolling in the “Introduction to Internet” course at a university. They were randomly divided into three groups: 51 students in the control group using traditional paper textbooks; 49 students in the experimental group I using e-textbooks with emotional design; 47 students in the experimental group II consisted using e-textbooks without emotional design. The MANCOVA was adopted to analyze gathered data. The results revealed learners’ positive emotion was significantly higher when using e-textbooks with emotional design than when using paper textbooks. Three types of cognitive load was significantly better when using e-textbooks with emotional design than when using e-textbooks without emotional design and paper textbooks. Cognitive efficiency was significantly better when using two types of e-textbooks (with and without emotional design) than when using paper textbooks. In a word, “emotional design and digitalization/mobilization” of textbooks (i.e. e-textbooks with emotional design) would significantly improve positive emotion; emotional design would significantly decrease intrinsic and extraneous cognitive load when using e-textbooks, and significantly increase germane cognitive load; digitalization/mobilization of textbooks (i.e. e-textbooks) significantly improved cognitive efficiency. What the study learned was that learners might firstly adopt e-textbooks with emotional design, and those without emotional design secondly, and paper textbooks rank last; viz., digitalized/mobilized textbooks (namely e-textbooks) were better than paper textbooks. Practical and theoretical implications and suggestions were raised for practitioners and researchers to take as references.
... Visual aesthetics shape a variety of psychological and communication outcomes, such as aesthetic appeal, emotion, psychological distances, and persuasiveness (Peng, 2017;Peng & Jemmott, 2018;Sutton et al., 2019). In addition to identifying the content of images, computer vision tools can extract aesthetic features, such as brightness, contrast, composition, color, texture, blur, and complexity (Bakhshi & Gilbert, 2015;Liu et al., 2016;Peng & Jemmott, 2018). These features are often unrelated to the actual content of images, but they can still meaningfully impact our perceptions of images. ...
... Ovenden, 2017; see also Bideaux, 2021, p. 697-699). In addition, pink attracts more attention than other colours (Lindsey, et al., 2010), and a study found that pink is one of the colours -alongside red and violet -with the best impact on the diffusion of an image on Pinterest (Bakhshi and Gilbert, 2015), and presumably the same applies to Twitter, Facebook or Instagram. This could explain the quick coverage of this event in the national and international press, and even more so on social networks. ...
Article
Full-text available
On 13th January 2022, some Italian police stations received stocks of pink FFP2 masks to protect them from covid-19. Many officers however refused to wear them because the colour would not be appropriate for a police officer, alerting the Sindacato Autonomo di Polizia – SAP. This article aims to revisit this “Pink Mask Affair”, to identify and detail the reasons which could or might have led police officers to refuse to wear pink masks. It proposes two hypotheses: 1) the association of pink with feminine and femininity could have led male officers to refuse to wear pink masks for fear of being perceived feminine; 2) bright and pastel colours are not suitable for police officers because they are also associated with femininity.
... The popularity of online photos has been mainly investigated concerning engagement (Park et al., 2015), which has been usually measured by number of likes and/or comments from users. A group of studies has employed a random sample of photos, which was not confined to a particular domain or topic, to examine which visual attributes of photos induced more engagement, and it was found that having human faces (Bakhshi et al., 2014) and using red, purple, and pink (Bakhshi & Gilbert, 2015) had influences on the engagement of photos on Instagram and Pinterest. Another group of studies has focused on a particular domain or topic; SNS photos about politics (Muñoz & Towner, 2017), online social movements (Casas & Williams, 2019;, and public health were investigated with regard to the relationships between the visual attributes of the photos and engagement. ...
Article
Full-text available
The goal of the current study was to explore the visual representation of #ShoutYourAbortion hashtag movement on Instagram. The photos’ content and embedded texts in the photos were examined. And the photos were clustered using k-means clustering algorithm, and the resulting clustered were compared using the same criteria above. Photo features which shows the content- and pixel-level characteristics were extracted and used for comparison between clusters. The photo features were also used to examine their relationships with the public’s responses. It was found that text was the main type of content, and the texts presented in photos were mainly about stories told in first person point of view as a woman. The photos were grouped into two clusters, which differed in terms of content and photo features. And the public’s responses were found to be related to photo features. The results are expected to contribute to the understanding of hashtag movements via photos and making photos in hashtag movements more appealing to the public.
... showing that aesthetic attributes are associated with a broad range of communication and psychological outcomes such as viewers' emotional responses to visual stimuli, the aesthetic appeal of images, the popularity of social media content, and persuasiveness of visual messages (Bakhshi & Gilbert, 2015;Labrecque & Milne, 2012;Lazard & Mackert, 2014;Matz, Segalin, Stillwell, Müller, & Bos, 2019;Morgan, Palmgreen, Stephenson, Hoyle, & Lorch, 2003;Sutton, Yang, & Cappella, 2019;Valdez & Mehrabian, 1994). Meanwhile, the rapid development of computer vision has offered social scientists techniques to analyze large-scale visual content, for example, identifying protests in social media images or emotional displays of politicians (Joo, Bucy, & Seidel, 2019;Peng, 2018). ...
... One line of research has investigated how visual aesthetics influence various kinds of digital behavioral data related to visual media (Miniukovich & De Angeli, 2016;Totti et al., 2014). Extensive research has looked at the virality of visual content on social media platforms (Bakhshi & Gilbert, 2015;Deza & Parikh, 2015;Khosla, Das Sarma, & Hamid, 2014;Totti et al., 2014). For example, Bakhshi and Gilbert (2015) analyzed the popularity of images on Pinterest and demonstrated that the percentages of red, purple, and pink were positively related to virality whereas green, blue, black, and yellow showed negative associations. ...
... Extensive research has looked at the virality of visual content on social media platforms (Bakhshi & Gilbert, 2015;Deza & Parikh, 2015;Khosla, Das Sarma, & Hamid, 2014;Totti et al., 2014). For example, Bakhshi and Gilbert (2015) analyzed the popularity of images on Pinterest and demonstrated that the percentages of red, purple, and pink were positively related to virality whereas green, blue, black, and yellow showed negative associations. Other kinds of digital behaviors are also examined. ...
Preprint
Full-text available
Visual aesthetics are related to a broad range of communication and psychological outcomes, yet the tools of computational aesthetic analysis are not widely available in the community of social science scholars. This article addresses this gap and provides a tutorial for social scientists to measure a broad range of hand-crafted aesthetic attributes of visual media, such as colorfulness and visual complexity. It introduces Athec, a Python library developed for computational aesthetic analysis in social science research, which can be readily applied by future researchers. In addition, a case study applies Athec to compare the visual aesthetics of Instagram posts from the two candidates in the 2016 US presidential election, Hillary Clinton and Donald Trump, showing how amateurishness and authenticity are reflected in politicians' visual messages. With tools of computational aesthetic analysis, communication researchers can better understand the antecedents and outcomes of visual aesthetics beyond the content of visual media.