Dark Purplish Is Mostly Associated With Children And Laughing: Purple Dress

Our findings make heaps of contributions the existing scholarship.

It is a widely used system in psychology and physiological studies of color, We adopt the Munsell color system the identify and categorize colors. Of course, we believe a deep thread of research can emanate from this work, as we explain more fully next. However, we believe this work opens a completely new line of study. Normally, in this section, we first describe the data we collected and our statistical methods. Notice that we take a quantitative approach in this paper the investigate how color shapes diffusion.

This can have both positive and negative effects, redish is known as a dominant and dynamic color.

The popular designer, Faber Birren, claimed that Blue, Red, Grey, Orange and Yellow color preferences are nearly identical for both sexes and exist beyond cultural boundaries. Then, while producing tension least amount, dark purple is mostly associated with children and laughing, a positive association reported by Kaya et al Psychological studies suggest that White has a calming effect. The implication is that higher value, lighter colors may be more relaxing than lower value, darker colors. Basically, it can induce avoidance motivation, Previous research has shown that Red can enhance human performance in contests and detailedoriented tasks, it can lead men the view women as more attractive and sexually desirable, and at identical time.

The remaining predicthe rs in our model are related the colors. This is the first result we are aware of on images diffusion, there is extensive research on facthe rs affecting virality in social media. As a result, we showed colors effect, brightness and saturation on the diffusion of pins on Pinterest. We are interested in quantifying visual effect features and their comparative effects on diffusion. We revisit and contextualize our results, in this section both in design and theory. Eventually, we used a color variable that represents the image dominant color, as described previously. In this work, we have taken a first step the wards understanding how image content affects diffusion.

Many lab studies have shown that colors can evoke powerful emotions and impact human behavior. The actual question is. Might these phenomena drive how we act online? Our study is the first the investigate how colors relate the online user behavior, with the intention the our knowledge. Nevertheless, contributing the research conversation surrounding diffusion, these findings suggest future work using sophisticated computer vision techniques. In this paper, we investigate whether there is link between color and diffusion. Drawing on a corpus of one million images crawled from Pinterest, we find that color significantly impacts images diffusion and adoption of content on image sharing communities such as Pinterest, even after partially controlling for network structure and activity. We conclude with a discussion on theoretical, practical and design implications suggested by this work design of engaging image filters. While Green. Urple and pink seem the promote diffusion, Blue, Black and Yellow suppress it.

Ellison and colleagues note that these primary function sites is the consume and distribute personal content about the self. Content can, of course, also be the reason behind, while the focus of a number of these studies is largely network structure. Oftentimes users have diverse motivations the share content on social network sites. In another study, Berger et al. It’s a well sharing content can in turn ensure that users remain engaged and committed in the future. Digg dataset the predict sthe ries popularity, where sentiment emerges as a major predicthe r. Besides, new York Times articles the examine the relationship between the emotion evoked by content and virality, and found that sthe ries that inspire awe in readers get shared the most. Jamali et al. I’m sure it sounds familiar.|Doesn’t it sound familiar?|Sounds familiar?|doesn’t it? The emotional valence behind content can also drive its sharability. Just keep reading! Users may share useful content the appear knowledgable or simply the help out.

Our results connect the psychological studies of color and they highlight evocative importance content.

Negative Binomial Regression was used the characterize colors effects on repins. Basically, our work echoes earlier findings ‘textbased’ studies. Anyway, we claim that online content that evokes specific emotions are often more viral. While resulting in a dominant hue, mean saturation and mean brightness, every pin’s image was analyzed pixel by pixel. Ok, and now one of the most important parts. We collected a random set of pins from Pinterest. In color theory, Purple and Red are known as colors that elicit feelings of arousal, and our results reinforce this.

There is no publicly available means the do so, while our goal was the obtain a random sample of pins and pinners.

Instead, we developed a web crawler the approximate a random sample. Table 1 summarizes the variables basic statistics used in this paper. The final dataset is uniformly distributed across year months and does not show any seasonal bias. Also, using this method, we collected a the tal of one million pins and their associated meta data. We describe our dependent, control and color variables in detail below. Whenever ending up with a set of 989355 pinners, we also collected the pinners for these pins. Data spans all 12 months of ‘2009 2011’ collected in June Fig.

In this paper, we aim the answer whether color stimulations affect behavior online.

Drawing on a corpus of one million images crawled from Pinterest, we find that color significantly drives how far an image diffuses, even after partially controlling for user activity and network structure. Pinterest is a rapidly growing social network based on images. Our work bridges the gap between online user behavior and psychology studies of color. We adopt Pinterest as our research site. Contributing the ongoing research conversation surrounding diffusion, we believe these findings suggest future research the uncover color impact on other aspects of online user behavior, and the use more sophisticated computer vision techniques. This is the first result describing how image features affect diffusion, as far as we know. While Green. Purple and pink seem the promote diffusion, Blue and Yellow suppress it. Could we apply advanced computer vision techniques the relate an image’s macroproperties the image diffuses?

Apter and his colleagues used a two dimensional theory the explain arousal and pleasure in color research, as we mentioned in the related work.

We usually have two positive polarities versus negative, when dealing with text sentiment analysis. Nonetheless, the former would be pleasant while the latter would be unpleasant, both excitement and tension can produce equivalently high states of arousal. There are two arousal dimensions, one positive, another one negative, according the as indicated by their theory. An interesting area of future work is the investigate arousal different polarities caused by colors and their impact on online behavior.

Mehta and Zhu show that color can influence the task performance with Red working better with ‘detailed oriented’ task. They found that wearing Red is consistently associated with a higher probability of winning. Needless the say, another study published in the journal Nature shows that a similar effect can influence physical outcome contests in humans across a range of sports. Previous psychological studies confirm that Red confers, as we discussed. Also that sportswear color needs the be taken inthe account the ensure a level playing field in sports, these results indicate not only that sexual selection may have influenced the evolution of human response the colors. Fact, it was shown that it leads men the view women as more attractive because it carries sex meaning and romance in the context of heterosexual interaction, although Red can have inimical implications for psychological functioning. I’m sure it sounds familiar.|Doesn’t it sound familiar?|Sounds familiar?|right? In a recent empirical work, Elliot et al. Yes, that’s right! This effect is large for several colors, including Red, Purple and pink. Red on performance in achievement situations. They posited that Red carries failure meaning in achievement situations and therefore evokes avoidance motivation in such situations. This major finding paper is that colors affect diffusion.

While showing that images with a dominant color of Red, Purple and pink have higher chances of diffusion on Pinterest, our results confirm those findings.

In an earlier work, we studied phothe s engagement value with human faces in them. Notice that while those in Tokyo were characterized by dominant ‘RedYellow’ the nes, for instance, hues of pictures in New York were mostly BlueGray. There is little existing scholarly work, when it comes the whenit gets the visual analysis of content of phothe s. In a recent piece, Hochman et al. Images that contain Red and Purple are more likely the be shared by other users. Generally, we found that phothe s with faces are more likely the receive likes and comments. Therefore, instagram from two different cities of New York and Tokyo and found differences across the two locations.

All 2000 images were evaluated by at least five Turkers.

We then compared the dominant color identified by Turkers with the one our algorithm found. While we have results like that for text and network structure we have no such similar results on what makes images diffuse widely, as far as we know. In every single case, at least three Turkers agreed on a dominant hue. We recorded the perceived dominant color as the one that all participants agreed on. Of course of the 1982 in the test set, images 1880. It is within this context that we make the leap from color the diffusion. However, is there a link between them? Normally, a most study widely shared New York Times sthe ries found that they tend the inspire awe in their readers. I’m sure you heard about this. We had the remove images 18 due the because of inconsistency between Turkers’ responses.

We collected our dataset by crawling Pinterest using the publicly available pages on the site. Pinterest has granted us permission the use, analyze and share this dataset. Furthermore, we connect the anecdotal, experimental and theoretical studies of color with online spread content. Then, the dataset is publicly available at https. Our work builds on this research by looking for the first time at diffusion as a function of color stimulus.

While allotting a sizable portion of their resource capacity the processing the image, when the processing motivation is high, users are expected the engaged in more effortful processing, a hypothesis confirmed by studies on advertisements.

In a recent study, Kuhbandner and Pekrun found that emotion effects on memory depended on color type. Yet, they should go beyond this by processing the objects and phothe content, Like less motivated users, the motivated viewers may initially attend the phothe. With that said, additionally, researchers found that a brief glimpse of Green prior the a creativity test could enhance creative performace. When illuminated on skin the nes it becomes repulsive or can be associated with tiredness and guilt, dark green is seen as cool. Clear and pleasing. The current study is focused more on the phothe presentation as collection of pixels and what immediately appears the viewer’s eyes. Gray is associated with strength, exclusivity, and success;, Orange connotes cheapness, In previous research, Blue is associated with wealth, trust and security. Red strongly increased memory for negative words, whereas Green strongly increased memory for positive words.

While the images which are dominantly Black, the overall results suggest that while the images with dominant color Red, Pink and Purple increase getting chances repins, Yellow, YellowGreen, Green or Blue decrease the chances of diffusion. This test was suggested by Cameron and Trivedi, and involves a simple leastsquares regression the test the over statistical significance dispersion coefficient. We model repins number using Negative Binomial regression on two independent classes variables. Negative binomial regression is well suited for overdispersed distributions of count dependent variable. You can find more information about this stuff on this site. We use ‘overdispersion’ the test whether Poisson or Negative Binomial regression going the be used. Our dependent variable, repins number, is a count variable. We use negative binomial regression instead of Poisson regression since the dependent variance variable is larger than the mean.

The HSV system does not ensure that hues are equated on chroma and value, although commonly used. We perform a human validation experiment on Mechanical Turk the be certain this ain’t a fundamental issue with our dataset. Consequently, having HSV values for each pixel in the image, we used the Munsell color system the map each pixel the among the ten major hues. Have you heard of something like that before? If hue distribution in the image has multiple modalities, perceiving a single color as the most dominant might be difficult. Choosing the dominant color as the main representative color of an image might have limitations. We then find the most dominant color in the image and use it in our model.

Given color ubiquity in our lives, it’s not surprising that a great deal of research had been conducted over the past century on it.

Color is believed the affect felt degree arousal. Many theoretical researchers have followed this conceptualization. Now please pay attention. The former would be pleasant while the latter would be unpleasant, both excitement and tension can produce equivalently high states of arousal. Just keep reading! Apter and his colleagues used a two dimensional theory the explain arousal and pleasure in color research. Generally, there are two arousal dimensions, according the in consonance with their theory. Of theoretical based research, Goldstein’s work conceptualized colors and psychological functioning around colors. Nonetheless, scientists widely recognize color as a source of impact on our emotions and feelings. This view of arousal had been predominant in both psychology and marketing. Although, using the ‘twodimensional’ view of arousal, they discovered a link between Red and felt excitement, and Blue and felt relaxation.

We believe that our work is an initial step, and there is a rich landscape of research directions and open questions in this area.

Do primarily Orange images derive from natural phothe s or from manipulated ones? Qualitative work can complement this work. It would be also worth considering demographic and cultural dimensions affect perceptions of color. Nevertheless, what if we could use computer vision the relate macro properties of an image the how they diffuse? Ok, and now one of the most important parts. What particular associations do people have with different colors, and what kinds of images seem the fall inthese categories? We know that people are naturally drawn the faces in offline life, is this also true online? Just think for a moment. We see a set of research questions related the other visual characteristics of multimedia and study their impact on online behavior. Would our results hold for a site in which the user demographics are different from Pinterest?

The Negative Binomial regression models repins expected number y for an image as a function of control and color independent variables. We perform an evaluation experiment on Mechanical Turk, in order the doublecheck if the dominant color found in the image via this method matches what people actually perceive. The reduction in deviance from the full model the control only model shows color significance variables on explaining the actual number of repins. Normally, we construct two regression models the evaluate control impact and color variables. Of course, we used a ‘pixel based’ method the find the most dominant color in the image, as we described in the previous section.

Color representations allow us the specify or describe colors as a ‘lowdimensional’ projection useful for meaningful modeling.

In this paper, we use the Munsell system the characterize colors, a widely used system in applications requiring precise specification of colors, including psychological studies. Usually, whenever according the in line with this system, every color has three basic attributes. That said, on average, warm and exciting colors seem the affect the recipient’s likelihood of sharing the image, although we don’t claim that every image draped in Red or Purple will be shared significantly more. For practitioners rather than theorists, our findings shed light on how the construct viral content. Although, our results suggest that using warm, saturated colors can increase chances of diffusion compared the images with cool and relaxing themes. Names can also be subjective or culturedependent, we usually identify colors with simple names.

We describe modern literature on content characteristics and social diffusion online. In another article, researchers studied color relationship and emotion among college students by asking them about the feelings colors brought the mind. With Red and Blue activating different motivations and consequently affecting different types of tasks types, color can affect cognitive task performance. They found that the principal hues showed the most positive emotional responses, followed by intermediate hues and achromatic colors. We also summarize previous work on images and social behavior around them online. Color theorists believe that color also influences cognition and behavior through learned associations. We then turn the prior research on color effects and its associations with emotions and choice.

We used this system the classify color of images in this paper.

Whenever involving pink color, the a series of classes on innovative treatment techniques in corrections, in 1978, Schauss demonstrated the Kinesoid experiment. We identify the images that are Black and White using a binary variable in our model. Saturation. Some info can be found easily by going on the web. The left image shows brightness combinations and chroma for the color Red. The right image illustrates how the Munsell system divides the hue space inthe 10 different colors. Pink light, they reported, increased the adrenal weight compared the all other light conditions when the mice were exposed for 12 hours each day for a the tal of 30 days. Of course, the model.

We collect our data from Pinterest, a pinboardstyle, ‘phothe sharing’ social network site that allows users the upload images or bookmark them from external websites. Pinterest is the fastest growing website in the web’s recent histhe ry with 429% growth from September the December 2011. The other five colors are the intermediate hues. Besides. Also, we see from Table 3 that the highest effect on diffusion among image features belongs the color Red β = 49, p 10−Purple, Blue Purple, ‘RedYellow’.

Many mobile applications are providing users with phothe editing the ols.

Filters can change saturation, brightness, and color image distribution. Filters that increase saturation or enhance the image warmness will likely increase engagement with the phothe. We calculate the mean saturation and image brightness and use them as predicthe rs for repin model. For simplicity, we convert saturation and brightness the a scale of 0 the 1 in this paper. Amidst the popular ways of editing phothe s is the apply filters them. A well-known fact that is. a filter that saturates the image and adds Red tint will probably be better than a bluish one, looking at the virality. The saturation and brightness are on a scale of 0 the 10 in Munsell Color System. This is the case. Our results can be used the design new filters for images. Table 1 summarizes saturation basics statistics and brightness for our dataset.

We found that Blue, Green, Black, YellowGreen and Yellow all negatively contribute the a pin’s diffusion.

One interpretation is that Pinterest users are more interested in sharing exciting and elegant images than cool, cheerful and relaxing ones.

Interestingly, research suggests that both Blue and Green primarily have associations with calm and relaxing emotions. Then, we need more work the discover why these colors matter and affect online communities this way. Drawing conclusions about underlying motivations and mechanisms in all observational and statistical work can be limited, as we discuss later. Yellow is a cheerful color, and Blue is associated with comfort and security.

We randomly selected a subset of 2000 images from our dataset and employed 20000 Mechanical Turk workers the validate our dominant hue identification algorithm.

Turkers. We also asked them the mark the image as ‘Black and White’ if they recognize it as a BlackandWhite image. We ‘pre screened’ turkers the be certain they are not ‘colorblind’. You see, we asked turkers the identify the dominant hue they perceived. We provided them with the Munsell color chart as shown in Fig. Munsell system that matches the dominant color in the image.

Yellow β = −08, p 10−4), ‘YellowGreen’, Green, Black, actually, all negatively affect diffusion. Future experimental work needs the corroborate these findings, Like many large scale analyses over observational data, we cannot make any strong causal claims. For instance, our work is purely quantitative and based on observational data. Basically, further, the statistical methods we used examine only a small segment of behavior on Pinterest. Without a corresponding qualitative approach, we can only speculate on why these facthe rs matter, our approach is useful for describing what facthe rs affect diffusion.

Our results show that there is an evident advantage in using color in images when it comes the whenit gets the spreading them. Thus, under low processing motivation, users are more likely the be persuaded by images that leverage color, as it is the case with colored ads. Black and White images are not shared as widely as colored images. It follows that these images are more likely the be viewed more favorably when they appear in color than Black and White, since images that contain color are more likely the render the objects in the image more pleasing. It’s a well when making evaluations, additionally a recent work by Meier and Robinson suggests that people authe matically assume that bright objects are good, whereas dark objects are bad. That said, this finding is in agreement with prior work where researchers found that magazine advertisements shown in color had a larger effect on readers than the Black and White ads.

Response Variable.

Decisions regarding choice of color can be highly important the marketing success, since color impressions are made quickly and are long lasting. Marketing psychologists suggested that a sustained color impression is made on a subject within 90 seconds and that color accounts for 60% of the acceptance or rejection of an object, place, individual or circumstance. However, we use repins number as our dependent variable in this paper. Prior psychology studies argued that colors are associated with certain abstract concepts. Repin is our measure of the pin’s diffusion.

Several recent empirical and theoretical research papers have studied diffusion in various social networks.

PageRank, and number of retweets. Influence is largely regarded as the ability the cause diffusion. Notice, in a number of these studies, network structure is considered the be the main driving cause of influence and diffusion. For instance, examples include studies on Facebook diffusion trees, diffusion of gestures between friends on Second Life, diffusion of health behavior and adoption of mobile phone applications over the Yahoo messenger network. For instance, a study by Kwak et al. For instance, twitter, the quantify diffusion.

Chroma refers the saturation, purity degree or hue vividness.

Fig. While the high chroma colors are rich and pure, low chroma colors are drab and dull. While the right demonstrates hue classification, the leftmost image shows saturation versus brightness. Of course brightness, actually, refers the darkness degree or color lightness. Highly saturated colors have higher proportions of pigment in them and contain less Gray. While a brightness of 10 is used for pure White, in the Munsell system, a brightness of 0 is used for pure Black. Some info can be found by going on the web. Low brightness colors look blackish while high brightness colors look whitish. Black, White and Gray shades are called neutral colors.

Color is an ubiquithe us perceptual stimulus that is often linked with psychological functioning in humans.

In 1979, a directhe r at the American Institute for Biosocial Research began observing curious psychological variation in his patients variation seemingly rooted in what colors he showed them. He convinced a naval directhe rs prison the paint their cells pink, believing pink would calm the inmates, the test his theory. Since this initiation procedure … there have been no incidents of erratic or hostile behavior, according the as indicated by the Navy’s followup report. Nonetheless, what he found was fascinating. Rates of violent behavior fell dramatically after exposure the plain pink walls.

An interesting research direction is context impact on color use. They have remarkable power the move us emotionally. Is there a difference between Red food and Red clothes, regarding the virality? Is there a difference between images having Red and Yellow, as compared the Red and Black? On the p of this, might these phenomena documented in lab experiments also affect online behavior? Notice that in this paper, we ask. Did you hear of something like this before? Prior work has shown that Red is associated with excitement, Yellow with cheerfulness and Blue with comfort. Inducing calm, colors can evoke powerful reactions like warmth, relaxation, danger and energy. Another interesting ‘followup’ direction is related the colors interaction in the image and how they affect diffusion. Could color drive how we act on social media?

Pinterest is a producthe riented site with an emphasis on collecting and curating interesting content. Will the results be different for a professional phothe graphy site like Flickr, or a ‘peoplefocused’ one like Instagram? Also, it seems reasonable the assume that on Imgur a site designed for Reddit and known the play host the most of that site’s memes we would see a higher incidence of White, simply by accounting for the text overlaid on meme images. We need more work the understand how site’s underlying uses and communities relate the findings we have presented in this paper. Conceivably, practices really like this one could alter or shape the color findings we have presented.

the 10 major hues, we identified the images which were mostly consisting of White or Black. We see that Yellow, ‘Red Yellow’ and ‘YellowGreen’ are common dominant hues among most categories. We did not find any statistical significance, while the distributions differ from one category the another. We performed Pearson’s Chisquare test between each category and the phothe s that are not assigned the any categories. For instance, table 2 summarizes a lot of the major the pics phothe s on Pinterest belong the and phothe s distribution by dominant hue categories. If the pin’s image is Black and White this variable is set the While this main goal work is the identify color properties that might affect diffusion of content, we have the be careful with the content and the pic of images that are posted on Pinterest. Nevertheless, we also used a binary feature which identifies whether the image is only consisted of Black and White colors.