Color as a Trigger: the Effect of Chromatic Characteristics of Touch Buttons on the Motivation

Griber Yuliya Alexandrovna Doctor of Cultural Studies Professor, Director of Color Laboratory at Smolensk State University 214000, Russia, Smolensk region, Smolensk, Przhevalskogo str., 4 y.griber@gmail.com Other publications by this author     Tsygankova Karina Yurevna Researcher at the Color Laboratory of Smolensk State University 214000, Russia, Smolensk region, Smolensk, Przhevalskogo str., 4 kayasmolensk@gmail.com

Ustimenko Yuliya Aleksandrovna PhD in Pedagogy Head of the Department of Design, Smolensk State University 214000, Russia, Smolensk region, Smolensk, Przhevalskogo str., 4 ustimenkoya@mail.ru Other publications by this author

IntroductionA familiar element of human interaction in modern society are smartphones – (from the English smartphone – smart phone) mobile phones supplemented with the functions of a personal computer.

According to statistics, the number of smartphone users in the world already exceeds 6 billion and increases by several hundred million annually ^[1]. People use smartphones to chat with friends, listen to music, browse social networks and read the news. Half of the owners describe their smartphone as something they "cannot live without" because, in addition to the obvious functional advantages, they receive emotional benefits from it – in particular, a sense of psychological comfort, a subjective sense of privacy, stress relief and tactile satisfaction ^[2].

As a rule, smartphones are manufactured with touch screens, which over the past decades have become an integral part of visual communication. The use of touch screens provides a quick response to user actions and makes it easier to perform many tasks. For example, they help to easily view the contents of the application using the scroll function, quickly switch between screens, easily select the necessary options using touch buttons.

For the convenience of the user, various colors are used when developing buttons, which help to make the interface intuitive, provide quick navigation through the main sections of the application and significantly simplify interaction with the program. However, recently, due to the growing interest in multisensory experience (see, for example: ^[3-6]) and design strategies based on evidence and scientific evidence (Eng. evidence-based design – evidence-based, scientific design), the emphasis in color selection has begun to shift noticeably from its exclusively aesthetic and functional characteristics [see sub-section: [7]). Of particular importance in modern industrial design is the understanding of the influence of color on cognitive processes, primarily on interpretation and motivation (see appendix: ^[8]).

According to the available data, among the five human senses, touch and sight dominate in our knowledge of the things of the surrounding world. Through touch, we can get information about texture and hardness, temperature and weight (see sub-section: ^[9]). However, in most cases we limit ourselves to visual observation only. To use tactile sensations and touch an object, motivation is needed, which, on the one hand, depends on the personal qualities of the subject of perception (see, for example: ^{[10; 11]}), and on the other – on the specific properties of the object (its shape, size, texture (see, for example: ^[12]) and the conditions in which all this happens ^{[13; 14]}.

The experiments conducted convince us that some visual qualities stimulate the motivation to touch an object that possesses these qualities to a much greater extent than others.  In modern studies of motivation, such qualities are designated by the term "trigger" (from the English trigger – trigger mechanism), which is used in a broad sense, denoting any cause of action; a stimulus that "triggers" a certain behavioral reaction (see, for example: ^[15]).

For example, one of the strong visual triggers of touch is a special wool-like structure of the material; on the contrary, the texture of cotton practically does not cause a desire to touch. Natural materials (wood or ceramics) motivate more than artificial ones (plastic or glass). Rounded shapes (circle, ball) – stronger than angular (triangle, cube, square). Small objects are more likely to attract than large ones (see subsp.: ^[16]). The motivation to touch is closely related to the emotions evoked by the object: people more often touch those things that they like, arouse interest and sympathy ^[13-16].

At the same time, an analysis of studies devoted to motivation shows that most of them are aimed at studying certain classes of real–world objects - food products on store shelves, clothing, fabrics, museum exhibits made of various materials. In this case, the study, as a rule, involves direct observation of the object (see, for example.: ^{[13; 14; 16]}) or (less often) a demonstration of its printed image or display on a computer screen (see, for example, ^[15]), and color is considered only as an attribute of an object and extremely rarely it is investigated separately (see e.g.: ^{[17; 18]}).

Continuing the tradition of studying the influence of color on motivation, the aim of our study was to experimentally test the hypothesis that various chromatic characteristics of buttons that a person sees on a touch screen (their brightness, tone, saturation) can act as independent stimuli of behavioral reactions (visual triggers) and have a noticeable effect on the desire to touch them.

The object of the study is modern smartphone users, the subject is the influence of chromatic characteristics of touch buttons on the motivation of touch. 

MethodParticipants of the experiment

The experiment involved 48 people aged 19 to 21 years.

All of them were born and raised in Russia and had no problems with color perception and color discrimination. The minimum sample size was determined according to the methodology of K.A. Otdelnova and, since the study was of a pilot nature and assumed an approximate acquaintance with the phenomenon under study, at a significance level of p = 0.05 it was 44 people (see appendix: ^[19]). To exclude the influence of age-related changes on the results, the criterion for inclusion in the experimental group was a relatively homogeneous age of participants (20 years ± 1).  To establish possible gender differences, subsamples of men and women were formed equal in volume (24 men and 24 women).

Experimental incentivesThe color stimuli of the experiment were developed on the basis of the PCCS (Practical Color Coordinate System) system proposed by the Japanese Color Institute in 1964.

The standard color circle of the PCCS system is divided into 24 tones (Fig. 1 on the left). The shades are grouped into 17 groups (Fig. 1 on the right), which differ in lightness (vertical axis) and saturation (horizontal axis).

  Figure 1. Color circle (left) and shade groups (right) of the PCCS system

52 shades were included in the experiment palette: 12 tones each from four groups – groups of saturated shades (v), light (lt), dark (dk) and gray (gy) shades.

The tone and group of each shade were encoded using traditional PCCS system designations (Table 1). This structure of the toolkit allowed analyzing the data obtained in several directions at once, grouping shades by tone (location on the color wheel), lightness (vertical axis) and saturation (horizontal axis) (Fig. 1 on the right).

Table 1Symbols of shades

Designations of groups of shades

To create pairs of colors, all shades were divided into 7 clusters (see subsp.: ^{[14; 15]}). Four clusters contained shades of only one specific chromatic group and achromatic colors; three more clusters included shades of all groups together:

Cluster 1 (N=16): 12 saturated shades (v) and 4 achromatic colors (gy).

Cluster 2 (N=16): 12 light shades (lt) and 4 achromatic colors (gy).

Cluster 3 (N=16): 12 bright shades (b) and 4 achromatic colors (gy).

Cluster 4 (N=16): 12 dark shades (dk) and 4 achromatic colors (gy).

Cluster 5 (N=16): 4 tones (2, 8, 12 and 18) from groups of light (lt), saturated (v), dark (dk) and 4 achromatic (gy) shades.

Cluster 6 (N=16): 4 tones (2, 8, 12 and 18) each from groups of light (lt), bright (b), saturated (v) and 4 achromatic (gy) shades.

Cluster 7 (N=8): a group of the most frequently selected colors based on the results of the previous experiment (lt8, b6, b8, v2, v4, v6, v8, gy 9.5) (Sato).

46 pairs of color stimuli were made up of the shades of all groups, each of which was presented to the participant once during the experiment.

Experimental designThe experimental tools were developed and tested in 2015-2016 in Japan (N=48) ^[17].

At the first stage, the participants' color vision was checked using the Ishihara test. At the second stage, the participants had to choose 46 times on the smartphone screen one of the two colors they wanted to touch and click on it.

Thus, there were 2208 data in the database (46 decisions of 48 participants).

Data analysisDuring the analysis , the following were evaluated:

(1) frequency of occurrence of individual shades;

(2) groups of shades;

(3) shades of a certain tone;

(4) the probability of choosing a certain shade out of two.

The frequency of occurrence was assessed using procedures and methods of visual statistics.

To assess the probability of choosing a certain shade from two, the method of searching for associative rules was used – a data mining procedure that allows you to find patterns between related events in databases (see appendix: ^[20]). To implement the search for associative rules in the Python programming language using the apyori library, a computer program was written that searches for data using the Apriori algorithm. This algorithm is based on the concept of a frequent set. At each stage, the algorithm scans the database, creates a set of candidates and calculates their support – estimates how often a candidate is found in the database. Next, the algorithm cuts off those candidates whose support is less than the minimum. The remaining sets are called frequently occurring.

ResultsShade rating

The analysis of color choices for all groups of color stimuli together made it possible to compile a popularity rating of individual shades.

The most popular were bright green (b12), saturated yellow (v8), bright green-blue (b16) and achromatic black (gy1.5) and white (gy9.5). The rating of the 10 most frequently chosen shades also includes two blue colors (lt 18 and v18), dark red (dk2), moderate gray (gy7.5) and saturated yellow-orange (v6) (fig. 2).

Figure 2. Rating (from left to right) of the most popular shadesThe lists of frequently chosen shades had pronounced gender differences (Fig. 3).

In men and women, they were only half the same. Women were much more likely to choose red shades (dark red dk2 and pink lt2), lilac (lt22), achromatic dark gray (gy2.5) and black (gy1.5). On the contrary, men preferred achromatic white (gy9.5) and light gray (gy7.5), which were not included in the rating of the 10 most popular shades among women at all; they more often chose saturated yellow (v8), yellow-orange (v6) and blue (v18), bright red-orange (b4) and green-blue (b16).

Figure 3. 10 most popular shades for women (top) and men (bottom); matching shades are connected by lines; codes of shades that were included in one list, but did not get into the second, are highlighted in redGroups of shades

A comparative analysis of the frequency of shades from different groups showed that participants were slightly more likely to choose saturated (v) shades than bright (b), dark (dk) and light (lt).

Achromatic greys (gy) were the most unpopular. We did not find any pronounced gender differences (r = 0.90 p = 0.0347) (Fig. 4).

Figure 4. Frequency of selection of shades from different groups in women (solid fill) and men (patterned fill); shades are arranged in descending order of frequency of occurrence in the whole sampleTone shades

The most popular were unmixed with other colors blue (18:B) and green (12:G) shades, the most unpopular – mixed yellow-green (10:YG) and purple (blue-purple) (20:V).

Women were much more likely than men to choose red (2:R) and red-purple shades (24:RP). On the contrary, among the shades chosen by men, there were more yellow (8:Y), yellow-orange (6:yO) and red-orange (4:rO). In addition to yellow shades, men more often chose green (12:G), blue-green (14:BG) and green-blue (16:gB) tones (Fig. 5).

Figure 5. Frequency of selection of shades of different tones in women (solid fill) and men (patterned fill); shades are arranged in descending order of frequency of occurrence in the whole sampleThe probability of choosing one of two shades in a pair

At the next stage, we applied the method of searching for associative rules to each of the seven clusters of stimuli.

This method allowed us to estimate the probability with which a person will choose one of the two shades in a pair. In this case, we used tone as criteria (in the code of experimental stimuli, tone was indicated by a number), brightness and saturation of shades (in the study, these characteristics were set by letters – b, v, dk, lt, gy).

The application of the Apriori algorithm showed that, choosing one shade out of two, men and women in many cases use the same strategies. So, from a combination of achromatic (gy) and chromatic shade from group b (bright shades), both men and women are almost equally likely to choose a bright shade (b) (probability 18.84% and 14.84%, respectively) (rows 3 and 7 of Table 2). On the contrary, from a combination of achromatic (gy) and chromatic shade from the dk group (dark shades), they are more likely to choose achromatic, and women are much more likely than men (probability 10.16% and 4.62%, respectively) (rows 6 and 13 of Table 2). In a combination of achromatic (gy) and chromatic shade from group v (saturated shades), the probability of choosing one or the other shade is almost the same for both men and women (21.88% for women, 7.88% and 7.34% for men) (lines 1-2 and 11-12 of Table 2).

Table 2The results of the algorithm A priori: the probability of choosing a shade from a certain group of brightness and saturation

Women

In terms of tone, the choice strategies have a pronounced gender specificity. So, according to the results of the Apriori algorithm, unlike men, in a combination of achromatic (0) and red (2), red-purple (24) or green-blue shade (16), women are much more likely to choose the chromatic option – red (probability 17.97% and 7.03%, respectively), red-purple (probability 8.07%) or green-blue (probability 7.81%). On the contrary, in combination of an achromatic shade (0) with yellow (8), they are more likely (4.69%) to prefer an achromatic shade (Table 3).

Table 3The results of the algorithm A priori: the probability of choosing a shade of a certain tone; all achromatic shades are indicated by the digit 0

Women

ConclusionsThe analysis shows that the various chromatic characteristics of the buttons that a person sees on the touch screen (their tone, lightness and saturation) have a noticeable effect on the motivation to touch them.

Firstly, the most powerful triggers of touch are shades with high saturation of the four pure tones to which the optic nerve reacts – bright blue and yellow, green and red colors.

Secondly, both the rating of the most popular shades and the strategies for choosing one shade from a pair have pronounced gender differences. Women are more motivated by red and red-purple shades, men – yellow, orange, green and green-blue.

Thirdly, the strategy of choosing one shade out of two in a pair is more predictable for women than for men. In a combination of achromatic and red, red-purple or green-blue shades, women are much more likely to choose the chromatic option. On the contrary, in a combination of an achromatic shade with yellow, they will prefer an achromatic one.

The results of the experiment confirm the conclusions obtained earlier that colors associated with tactile sensations do not coincide with the boundaries of lexical categories of color; these associations are not lexical in nature and are not derived from lexical meanings fixed in the language ^{[21; 22]}. This, in particular, is indicated by the "blurred" silhouette of the dominant color triggers on the color wheel.

We do not exclude that, as in experiments with real objects ^[12-15], the motivation to touch the touch button on the smartphone screen may be closely related to the emotions caused by color, and the strongest triggers are those colors that arouse interest and sympathy (cf.: ^{[23; 24]}). However, this assumption needs additional empirical verification.

The data obtained have a wide application potential. They can be used in the design of websites and web applications for users with different socio-demographic characteristics, in the design of interactive educational materials, in the development of the color environment of training and training programs.

In order to form final conclusions, the study should be continued on a larger sample, primarily in relation to possible age characteristics and cross–cultural specifics.