Keywords

1 Introduction

Through the circulation of the global economy, Kawaii, has drawn the attention of people both in and out of Japan. By outputting many ACG industry, Kawaii has fevered the whole world and become a sub-culture in the world. There were many popular Kawaii image such as Doraemon, the character in the comic of Fujiko • F • Fujio, Hello kitty which designed by Shimizu Yuko, as well as Sailormoon which swept the Europe in 1944, and the Pokemon just made a hit in 2016 (Allison 2003; Perez 2016). These widely known characters and their affiliated products flooded in our life and also brought great commercial value. According to the list of best-selling entertainment products unveiled by Forbes, Hello Kitty made $800 million in 2011 and tightly seized the world’s fifth (Goudreau 2012). The effects of Kawaii are not only in the animation, comics, and games but also in the entertainment industry, public facilities, transportation, industrial products, consumer goods and even food.

The description of Kawaii debuted during Heian era in 8th -12th AC. An archaic word, utsukushi, was denoted the meaning of Kawaii and the meaning changed to beautiful in modern time. At the end of Heian era, the word, kawayui, arose as predecessor of Kawaii which meant ashamed or blushing due to a twinge of conscience or pity and pitiable during medieval era in 12th–16th AC. Till 1945 in Taisho era, Kawaii was printed in Japanese dictionaries as kawayuishi and later changed to kawayui (Cheok and Fernando 2012; Nittono 2016). As the sweeping trend of Kawaii, the word Kawaii has been added in to the Collins English Dictionary in October 2014.

Recently, a new concept called Kawaii+ arose, where the plus mark means that Kawaii can evoke human being positive emotions, and has a lot of positive meanings based on experiments in previous studies. Kawaii refers to a special positive emotion whose features are related to smiling and approaching behavior (Nittono and Tanaka 2010). Kawaii can reduce people’s fear and make uninteresting information more acceptable such as a bitter pill with a layer of sugar (Cheok and Fernando 2012). Results of previous experiments showed that viewing Kawaii images can help improve performance on tasks that need carefulness, narrow the breadth of attentional focus, and reduce the global precedence effect. When viewing Kawaii images, people’s motor behaviors were more physically tender, and it furthermore increased perceptual carefulness (Nittono et al. 2012; Sherman et al. 2009). People had great interest and predilection for Kawaii objects. They were sensitive about, drawn to and excited by something Kawaii, which can bring them positive feeling of comfort. There was gender difference in Kawaii perception. Japanese females were more positive and sensitive to Kawaii objects than males, and female used the word Kawaii more often. (Tokosumi and Teng 2011; Nittono 2016).

Previous researchers have found stimulus factors of Kawaii from tactile, visual, motional and auditory aspects. The tactile textures which make people feel Kawaii such as yarn, cotton, and sheep fabric are bushy, flurry, soft, smooth, and elastic. As for visual stimuli, the visual textures which are tangible, flurry, and soft are commonly considered Kawaii (Ohkura and Komatsu 2013). As for colors, experiments showed that warm colors were generally chosen as more Kawaii than cold colors. The yellow and purple hues were chosen as Kawaii most frequently. The combinations of hue and brightness or saturation were important for evaluating Kawaii degrees of colors (Ohkura et al. 2014; Ohkura et al. 2008). The color in pink or basically like pink were chosen as Kawaii colors, while dark brown and dark green were chosen as non-Kawaii colors in previous experiments. Objects with small size were generally chosen as more Kawaii than larger objects. The curved shapes were chosen as more Kawaii than shapes with straight lines in two-dimensional space. While in the three-dimensional space, round objects were chosen as most Kawaii in general. The effect of shape on Kawaii perception is also dependent on color (Ohkura et al. 2011; Ohkura et al. 2014). In a study of Roomba, motions that were rotational, slow, and sudden changes in acceleration by collisions were evaluated Kawaii (Sugano et al. 2013). In a word, a Kawaii object is in bright, primary, pure, and warm colors (i.e. red, orange, and yellow), soft and with nature longer pile tactile texture, has animal like movements and small steps in motions, higher pitch in sounds, small size, baby like or bigger head in proportion, and roundness in shapes. These characters are all connected to the natural world. Newborn babies with these characters evoke Kawaii feelings and the feelings such as young, innocent, vigorous, pure, harmless and friendly (Cheok 2010). Kawaii features found in literature were summarized in Table 1.

Table 1. Variables contributing to Kawaii in previous literature
Full size table

The features mentioned above contribute to Kawaii from various aspects. Does a simple combination of these features lead to a perfect Kawaii design? If not, how do these features interact with each other in Kawaii design? Up to our knowledge there has been little research so far focusing on the integration of Kawaii features. In this study, we aim to explore Kawaii semantic space from an integration view. First, we conducted semantic questionnaires to find factors of Kawaii features. Then we conducted a full factorial experiment to explore the interactions between the factors.

2 Study I: Exploring the Semantic Space

2.1 Methodology

A semantic differential scale was used to construct the questionnaire of this survey. The semantic differential scale was proposed by Osgood and was used to measure the semantics or meaning of words, particularly adjectives and their referent concepts (Osgood 1952). The questionnaire contained two parts. The first part was constructed by 36 pairs of adjectives with opposite meanings. The adjectives were chosen according to the findings in pervious literatures. A five-point scale was used to construct the questionnaire (1-agree to the first adjective; 5-agree to the second adjective). The adjectives were 14 pairs in visual subpart containing color, size, shape, proportion, and textures; five pairs in acoustic subpart containing pitch, tone and volume; five pairs in tactile texture subpart; six pairs in motion subpart; and six pairs in motion subpart. The second part was to collect demographic information of the subjects and their willingness in purchasing something with Kawaii design. The demographic information included subjects’ gender, age, and their nationality.

An online questionnaire was used and each participant can have the opportunity to win a lottery draw as a reward. There were 125 questionnaires collected and 118 of them were valid. There were 51 male subjects, and 61 female subjects. Their ages range from 19 to 28 years old (M = 23.6, SD = 2.1). All subjects were Chinese. There were 90 subjects (76.3%) willing to buy something with Kawaii design.

2.2 Survey Results

For the survey results, if the average rating on one pair of subjects is smaller than 3.0, it means that the first adjective was considered more Kawaii than the second one; and vice versa. Following are the results of the survey.

Visual.

The visual aspect included color, size, shape, proportion, and visual texture. The mean ratings of both male and female subjects indicated that warm, pure, saturated, bright color, small size, curved line, orderliness, round shape, disproportion, babylike, soft and furry looked, tangible and natural features contribute to visually Kawaii. Gender difference existed. Although both genders rated on the same side on all the objectives, male subjects (M = 2.29, SD = 1.19) agreed more on pure color than female participants (M = 2.81, SD = 1.35, F = 4.61, p = .034) as Kawaii, while female participants agreed more on babylike (Female M = 2.10, SD = 0.90; Male M = 1.64, SD = 0.67; F = 10.00, p = .002) and furry look (Female M = 1.39, SD = 0.60; Male M = 1.71, SD = 0.80; F = 5.61, p = .020) than male participants as Kawaii.

Acoustic.

The acoustic aspect included pitch, volume and tone. The mean ratings of both male and female subjects indicated that high pitch, smooth, whispered, lightly and soft features contribute to acoustically Kawaii. Gender difference on acoustic aspect was not found.

Tactile.

For tactile aspect, the mean ratings of both male and female subjects indicated that bushy, soft, smooth, elastic and fluffy features contribute to tactilely Kawaii. Female subjects (M = 1.46, SD = 0.64) agreed more on fluffy than male subjects (M = 1.73, SD = 0.75, F = 4.23, p = .042) as a Kawaii feature.

Movement.

In movement aspect, the mean ratings of both male and female subjects indicated that rotational, obedient, liquid movement, obtuse and tardy features contribute to movement Kawaii. Female subjects agreed that brave is more Kawaii (M = 3.40, SD = 1.12) while male subjects agreed that shrinking is more Kawaii (M = 2.63, SD = 1.15), which was significantly different (F = 13.65, p < .001)

Emotional.

For emotional profile, the mean ratings of both male and female subjects indicated that innocent, harmless, cheerful, sympathy, young and vitality features contribute to emotionally Kawaii. Female subjects (M = 1.09, SD = 0.29) agreed more on cheerful than male subjects (M = 1.25, SD = 0.56, F = 4.34, p = .039) as Kawaii.

2.3 Factor Analysis

Exploratory factor analysis was applied to find the dimensional semantic space of Kawaii. The KMO measure of samplings adequacy was 0.78 > 0.70. The principle component analysis was conducted as extracting method and the factor whose Eigenvalue was more than 1.0 were extracted. Varimax rotation with Kaisar normalizing was operated as a rotation with the items, and the lower limit of the factor loading was 0.50 (Hair et al. 1995). The extracted factors were named by the common meanings of the comprising item. Six factors were extracted and 67.78% of overall variance were explained. The results were summarized in Table 2.

Table 2. Factor analysis of Kawaii features
Full size table

The Cronbach’s alpha of the six factors were 0.78 (baby schema), 0.75 (acoustics), 0.64 (sense of obedience), 0.61 (movement), 0.55 (sense of harmony), 0.37 (color), respectively. Four of them were larger than 0.60 (Hair et al. 2010). The fact that the last two alphas failed to reach 0.60 may due to the limited number (only 2) of items within those two factors.

3 Study II: Exploring Design Element Portfolio

3.1 Methodology

Based on the result of semantic questionnaire, we selected one typical element in each factor as experiment factors and applied them on three types of product design. In the experiment, 3D models of furniture (chair), household appliances (microwave oven), and robots were created by 3D blender 2.78 and showed to the participants. To control the number of factors, the item obedient was merged into sense of harmony and smooth was merged into acoustics. Thus, the five factors of Study II were baby schema (roundness), acoustics (soft sound), movement (obtuse movement), sense of harmony (pure color), and color (brightness). A 2 k-full-factorial experiment design was used. Table 3 showed the high and low levels defined for each factorial design object. The low and high levels for the factors were selected according to some preliminary experiments. There were 2 * 3 treatments in chair, 2 * 4 treatments in microwave oven, and 2 * 5 treatments in robot. For each type, a non-colored and unchanged prototype was used as the compared object. Participants watched a total of 56 trails and scored on whether the treatment is more Kawaii than compared object.

Table 3. Experiment design of Study II
Full size table

An 11-point Likert scale was conducted to measure whether the design is more Kawaii than compared object (1-relatively not Kawaii; 6-neutral; 11-relatively Kawaii). Figure 1 was part of designs used in the experiment. In the experiment, chair prototype refers to the general office chair design, microwave oven refers to the Media microwave oven prototype design, and the robot is a reference to the Japanese animation character Gundam as a prototype design.

Fig. 1.
figure 1

Examples of prototypes used in Study II

Full size image

In this stage, 30 Chinese students from Tsinghua university (15 males and 15 females) who aged 20–29 (M = 23.9, SD = 2.2) were invited to participate in the experiment. Each participant was given 80 RMB as a reward.

3.2 Results of the Experiment

Repeated measured ANOVA was used to determine the main and interaction effects of factors and Mauchly’s test was used to test the sphericity assumption. In cases in which the sphericity assumption was violated, the degrees of freedom were corrected using Greenhouse-Geisser estimates of sphericity. Effect size was measured with a generalized eta-squared (η2) which proposed by Olejnik and Algina (2003). Paired t-tests with Bonferroni adjustment were used for post hoc analysis. The main results are as following.

Furniture (Chair).

The means of ratings on all designs were higher than neutral point 6.0, except for design 1 (not rounded, not pure color, not bright) and design 3 (not rounded, pure color, not bright). The results of ANOVA indicated that the effect of roundness and brightness were significant on Kawaii ratings. Rounded design (M = 8.08, SD = 2.61) was rated significantly more Kawaii than non-rounded design (M = 6.78, SD = 2.89, p < .001, η2 = 0.12). Design with highly bright color (M = 8.39, SD = 1.88) was rated significantly more Kawaii than design without bright color (M = 6.47, SD = 2.62, p < .001, η2 = 0.25).

Household Appliances (Microwave Oven).

The means of all designs were higher than neutral point 6.0. The results of ANOVA indicated that the effect of roundness, the interaction effects of roundness * pure-color, roundness * brightness, pure-color * brightness, and roundness * sound were significant. Rounded design (M = 8.08, SD = 1.93) was rated significantly more Kawaii than non-rounded design (M = 6.70, SD = 2.22, p < .001, η2 = 0.15). The effect sizes of interaction effects of roundness * pure-color, roundness * sound were too small (less than 0.10) so they were not discussed further. Post-hoc tests showed that the effect of brightness was only significant when the design was not rounded or when the design was not pure color. When the design was not rounded, design with highly bright color (M = 7.05, SD = 2.33) was rated significantly more Kawaii than design without bright color (M = 6.34, SD = 1.86, p = .002, d = 0.34). When the design was not pure colored, design with highly bright color (M = 7.98, SD = 1.97) was rated significantly more Kawaii than design without bright color (M = 7.13, SD = 2.47, p < .001, d = 0.38).

Robot.

The means of all designs were higher than neutral point 6.0. The results of ANOVA indicated that the effect of roundness, movement, brightness, sound, the interaction effects of pure-color and brightness, roundness * pure-color * sound, movement * brightness * sound, roundness * movement * pure-color * brightness * sound were significant. Only the effect of roundness had an acceptable effect size (larger than 0.10). Rounded design (M = 8.31, SD = 1.45) was rated significantly more Kawaii than non-rounded design (M = 7.36, SD = 1.37, p < .001, η2 = 0.17).

4 General Discussion

To our knowledge, this research for the first time fuses multiple Kawaii features from visual, acoustic, tactile, movement and emotional aspects. Previous research on Kawaii perception are mainly conducted with Japanese participants, while the involved participants in this research are all Chinese, providing insights of Kawaii perception beyond Japanese culture. The first study examines Kawaii perception between Chinese male and female, as well as establishes a six-dimension Kawaii model. The second study in further investigates Kawaii design features of different product types with experimental approach.

The survey confirms that for Chinese participants, warm, pure, saturated and bright color, small size, curved line, orderliness, round shape, disproportion, babylike, soft, furry, tangible, natural looked; high-pitch, smooth, whispered, lightly and soft sound or voice; bushy, soft, smooth, elastic and fluffy touched; rotational, obedient, liquid, obtuse and tardy movement; innocent, harmless, cheerful, sympathy, young and vitality emotions contribute to a Kawaii design. Chinese understand Kawaii nearly the same way as Japanese do (Cheok 2010; Ohkura and Komatsu 2013; Ohkura et al. 2014, 2008). Gender difference exists in some aspects. In general, female Chinese agree more on furry (both visually and tactilely), baby schema and cheerful as Kawaii and male Chinese agree more on pure color as Kawaii, although the tendency for both genders are the same on these features. The divergence is in movement, where female think brave as Kawaii while male think shrinking as Kawaii. The survey establishes a full-scale semantic space for Kawaii, can provide detailed guidelines for Kawaii design, and help extends Kawaii definition from microcosmic perspective.

The six-dimensional Kawaii model is an integration of previous studies on Kawaii features and points out the six most important facets of Kawaii design. According to the model, a Kawaii design should consider baby schema, acoustic aspects, sense of obedience, movement, sense of harmony and color.

The experiment involves three types of products: furniture (chair), household appliances (microwave oven), and robots. Results of the experiment indicates that for all three products, roundness in shape is a key element to make a design Kawaii. Bright color is effective to make a furniture Kawaii. The results of effects on household appliance Kawaii is noteworthy. The analysis shows that only when roundness and pure-color level is low, the brightness in color is effective to make a household appliance Kawaii, while when the design is already rounded and pure-colored the effect of brightness fails to be significant. This finding is valuable to point out that Kawaii features do not simply accumulate to make a design Kawaii, but their influences interact with each other. Kawaii is not a result of aggregated Kawaii features, but an integrated concept. Change in one feature of a design can influence the Kawaii in other features. Kawaii is a Kansei concept (Ohkura et al. 2008). Designers should take full consideration of the relationship among every design element to provide a product with Kawaii value.

The research has some limitations. First, the participants are all Chinese. Although these years Chinese young people are highly influenced by Japanese culture and every participant in this research claims that they know Kawaii, there is still doubt that whether the model and the experiment results can be generalized and represent the “Kawaii” as its original meaning. Second, due to the difficulties to make real products with different designs, 3D models are used in the experiment. Previous research indicates that Kawaii perception in real and virtual presentation may differ (Ohkura et al. 2009). Third, the sample size of the survey is only 118 and this is not sufficient to confirm the factor analysis. Further investigation with sample from Japanese or other cultures, and experiments with real objects are needed in generalization of the findings of this research.

5 Conclusion

This research provides a dimensional semantic model of Kawaii design features. The experiment further tests the semantic model with controlled designs of typical artificial product types. The results of this study provides an integrated picture of what a Kawaii design should be like, and provide valuable guidelines for designers who want to take advantage of the power of Kawaii into their product design.