Cuteness in Japanese Design: Investigating Perceptions of Kawaii Among American College Students

Abstract

Japanese products and pop culture, such as Hello Kitty, Pokemon, J-pop, and Anime, have gained global popularity, including in the United States. As a result, Japanese Kawaii (cute) design has also spread. It is not clear, however, how American perceptions of Kawaii compare to Japanese perceptions. In previous work by the third and fourth authors, Japanese college students rated 225 images with respect to Kawaii and other characteristics. This work compared perceptions between male and female Japanese college students. In the current paper, we report on a cross-cultural study of perceptions of Kawaii between Japanese students and American students. The study uses data we collected from American students who rated most of the images from the prior study that involved Japanese students.

1 Introduction

Japanese cuteness, described as Kawaii, is ingrained in Japanese contemporary society in many forms. As Yomota reports in the book “Kawaii Ron”, the word Kawaii stemmed from the word kawayushi that appeared in Makura no Sōshi (The Pillow Book) in classical Japanese literature [1]. At this time, the word Kawaii meant pitiful, shameful, or too sad to see. During the course of Japanese history, the meaning of the word started to describe the small, weak, and someone or something that invokes the feeling of “wanting to protect” [1].

The meaning of Kawaii has been extended to the concept of “Japanese cuteness” in contemporary society. Indeed, in their book “Cuteness Engineering: Designing Adorable Products and Services” Marcus, Kurosu, Ma and Hashizume confirm that Kawaii is the closest Japanese word to the English word cute [2]. However, the authors also acknowledge that the words are not perfect translations for each other and they note that Kawaii can also take on aspects of the English words “pretty” and “interesting” [2]. In the context of cuteness, the notion of Kawaii is pervasive in Japan and ranges from Hello Kitty products to road signs to posters created by the Japanese government, just to name a few examples. Japanese products are consciously tailored to accommodate widely preferred “cuteness.” Therefore, when designing a product, it is important to understand how specific groups of target users perceive Kawaii.

Japanese products and pop culture, such as Hello Kitty, Pokemon, J-pop, and Anime, have gained popularity around the globe, including in the United States. As a result, Japanese Kawaii design has also spread to some extent. For example, a recent survey of 25 college students at an American University revealed that more than 70% of the respondents were familiar with the concept of Kawaii. It is not clear, however, how closely American students’ perceptions of Kawaii match their Japanese counterparts [3].

In previous work, the third and fourth authors studied the extent to which perceptions of Kawaii in 225 specific photographs differ between male and female Japanese college students [4]. The photographs were divided into subgroups including products, objects, foods, geometric shapes, animals, characters and people. Gender differences were identified, depending on the subgroup of object studied.

2 Details of the Current Study

2.1 Participants

Participants were recruited from the student body at DePauw University, which is an undergraduate, residential, liberal arts college in the Midwestern United States. A general invitation to participate was posted on an electronic bulletin board and email invitations were sent to a variety of student groups. Due to University Institutional Review Board requirements, in order to be eligible for the study, participants had to be at least 18 years of age. In order to limit the study to students who would view the images primarily through an American cultural lens, we excluded participants who had not been raised primarily in the United States and we further excluded participants who had taken classes in Japanese language or culture.

In total, 47 students participated in the study at DePauw University. All 47 of these participants were raised primarily in the United States and 37 of the participants reported that they had never lived outside of the United States. One participant had lived in Ghana for 7 years, one had lived in England for 2 years, and 9 had lived outside of the United States for at most one year, likely during college study abroad programs that typically last between 2 and 4 months. In the remainder of this paper, for convenience, we will refer to these 47 participants as “American” participants since they were raised primarily in the United States, even though we did not ask them about citizenship.

The American participants ranged in age from 18 to 22 with a mean age of 20.29. Of the 47 American participants, 23 identified as female and 22 identified as male. The remaining two American participants self-reported a non-conforming gender identity. Data from these two participants were excluded from consideration when comparing results of male and female participants.

Data from the American participants are compared to the data that were previously collected from 89 university students in Japan, who we will refer to as the Japanese participants. These participants ranged in age from 18 to 24 with a mean age of 20.07. The Japanese participants included 54 males and 34 females [4].

2.2 Materials and Setup

In previous work by the second and third authors, Japanese participants had been asked to rate each image in a series of 225 photographs that depicted objects, people, scenes, small animals, sweets, geometric shapes, female celebrities, etc. [4]. In order to receive expedited Institutional Review Board Approval, we excluded eight images with potentially sensitive material (for example naked people) from consideration in the current study. Thus, the American participants viewed 217 images, which were a subset of the 225 images viewed by the Japanese participants.

The current study took place in a computer laboratory in an academic building at DePauw University. Data were collected in three sessions over a four-day period and each session had between 11 and 19 student participants.

2.3 Procedure

The procedure for the current study consisted of six parts.

• Participants were welcomed to the computer laboratory, shown to a seat and given an informed consent form and a participant number.

• After all the participants arrived, one of the experimenters read the informed consent form while the participants followed along. Each participant then signed an informed consent form.

• Students navigated to a Google Form where they entered their subject number and answered some basic demographic questions.

• The experimenters then presented a series of 217 images one by one on a screen at the front of the computer laboratory. Each image was shown for approximately fifteen seconds. While each image was presented, students used a Google Form to respond to several questions about the image.

• After the final image was presented, participants answered an additional question about their perceptions of cuteness and beauty as well as an additional question about their overall experience during the study.

• Finally, the students were debriefed.

Figure 1 shows the survey questions that were presented after each image.

Fig. 1.

Questions presented after participants viewed each image.

3 Results

A list of the descriptive titles of each image used in this study can be found in [4]. Where we provide image titles in this paper, the titles are often taken verbatim from [4]. The actual images are not displayed in this paper due to copyright limitations. However, Fig. 2 provides an example of an image, purchased from iStock with rights for re-use, similar to one of the images (Image 22) used in the study.

Fig. 2.

Japanese sweets

Five pair-wise comparisons were made for each of the images: (a) Japanese males were compared to Japanese females, (b) American males were compared to American females, (c) Japanese males were compared to American males, (d) Japanese females were compared to American females, and (e) all Japanese participants were compared to all American participants.

Tables 1, 2, 3, 4 and 5 report mean results for one of the images of Japanese Sweets (Image 22) used in the study. See Fig. 1 to review the full questions summarized in the tables. When participants responded that they were familiar with the image (Know) the response was recorded as 1, otherwise, the response was recorded as 0. For the remaining questions (Cute, Beauty, Like) answers were recorded on a five-point Likert scale with 1 meaning strongly disagree through 5 meaning strongly agree. A t-test was computed to test for differences between populations. Significant differences at the p < .01 level are marked with ** and significant differences at the p <  .05 level are marked with *.

Table 1. Japanese males versus Japanese females for Japanese sweets (Image 22.)

Table 2. American males versus American females for Japanese sweets (Image 22.)

Table 3. Japanese males versus American males for Japanese sweets (Image 22.)

Table 4. Japanese females versus American females for Japanese sweets (Image 22.)

Table 5. All Japanese versus all American for Japanese sweets (Image 22.)

As shown in the Table 1, Japanese females were more familiar (Know) the image than Japanese males (p < .01) but there were no differences in ratings for Cute/Kawaii, Beauty, or Like between these groups. Table 2 indicates that American females rated the image more highly than American males with respect Cute/Kawaii (p < .05) but there were no other differences between these groups.

As shown in Table 3, Japanese males were more familiar with the image than American males (p < .05) but American males rated the image more highly with respect to Cute/Kawaii (p < .05). Similarly, as shown in Table 4, Japanese females were more familiar with the image than American females (p < .01). While there were no significant differences in Cute/Kawaii ratings between these groups, Japanese females rated the image more highly than American females with regard to Beauty (p < .05).

Finally, Table 5 compares responses between all Japanese participants (combining males and females) and all American participants. As reported in the table, Japanese participants were more familiar with the image than American participants (p < .01) but American participants rated the image more highly than the Japanese participants with regard to Cute/Kawaii (p < 0.1), while the Japanese participants rated the image more highly with regard to Beauty than the American participants (p < .05).

A similar set of 5 tables were produced for each of the 217 images used in the study, which resulted in 1085 tables in total. Space limitations prevent printing all of the tables.

We used the tables and Excel conditional formulas to identify images that had a Cute/Kawaii rating of 4.0 or greater for at least one population (Japanese males, Japanese females, American males, and American females). Using this set of images as a base, we then identified images for which the Cute/Kawaii ratings differed significantly between a pair of groups. In particular:

• There were significant differences between Japanese males and Japanese females in the Cute/Kawaii rating for 16 of these images.

• There were significant differences between American males and American females in the Cute/Kawaii rating for 6 of these images.

• There were significant differences between American females and Japanese females in the Cute/Kawaii ratings for 20 of these images.

• There were significant differences between American males and Japanese males in the Cute/Kawaii ratings for 18 of these images.

• There were significant differences between all Americans and all Japanese in the Cute/Kawaii ratings for 27 of these images.

Excel conditional formulas were used to help us identify additional patterns in the data. These patterns are discussed in the next section, particularly with respect to cross-cultural differences.

4 Discussion

We analyzed the data with a focus on understanding differences in perceptions of Cute/Kawaii between American participants and Japanese participants. Key findings are presented in the remainder of this section.

Table 6 shows the number of images for which each group’s mean Cute/Kawaii rating was 4.0 or higher. These images represent the notion of Cute/Kawaii for the respective group. The table also shows the mean rating for those images that were rated 4.0 or higher. For example, as indicated in the first row of the table, Japanese males rated 11 different images with a mean Cute/Kawaii score of 4.0 or higher. The mean rating provided by Japanese males for these 11 images was 4.16.

Table 6. Number of Cute/Kawaii images and mean ratings for these images.

As shown in Table 6, Japanese males rated fewer images to be Cute/Kawaii than any other group. In particular, they rated half as many images to be Cute/Kawaii as American males, and they rated about one-third as many images to be Cute/Kawaii as Japanese females. Japanese males also had lower mean ratings for those images they did judge to be Cute/Kawaii. The lower Cute/Kawaii ratings provided by Japanese males largely accounts for the difference in ratings between the All Japanese group and the All American group.

Table 7 shows the number of images that had a mean Beauty rating of 4.0 or higher for each group. These images represent beauty for the respective group. Japanese males rate fewer images as Beautiful than any other group. However, for those images they do find beautiful, the mean Beauty rating for Japanese males is similar to other groups. The difference in the number of images all Japanese participants find beautiful, compared to all American participants, is primarily due to the differences between Japanese males and other groups.

Table 7. Number of beautiful images and mean ratings for these images.

Table 8 shows the number of images that had a mean rating of 4.0 or higher with regard to “Like” for each group. Once again, Japanese males rate fewer images as likeable than any other group. However, for those images they do like, the mean rating for Japanese males is similar to that of other groups. The difference in the number of images all Japanese participants liked, compared to all American participants, is primarily due to the differences between Japanese males and other groups.

Table 8. Number of liked images and mean ratings for these images.

In summary, when compared to all other groups, Japanese males rate fewer images at 4.0 or above in terms of Cute/Kawaii, Beauty and Likeability. For those images they do rate at a 4.0 or above, Japanese males seem to have lower ratings than other groups with regard to Cute/Kawaii but not with regard to Beauty or Likeability.

There were only four images that had mean Cute/Kawaii ratings of 4.0 or higher for each group (Japanese males, Japanese females, American males, American females). These images are described in Table 9. Three of these images depict animals and one image (Pikachu) is an animation character that represents a small animal.

Table 9. Images with a Cute/Kawaii Rating >= 4.0 for each Group.

In addition to the four images listed in Table 9, there were seven images that Japanese males assigned a mean Cute/Kawaii rating of at least 4.0. These images are listed in Table 10. Note that each of the 11 images listed in Tables 9 and 10 depict animals or adult humans with the exception of the Pikachu image, which is an animated animal character. Japanese males did not assign any other types of images a mean Cute/Kawaii rating of 4.0 or more.

Table 10. Additional images Japanese males assigned a Cute/Kawaii Rating >= 4.0.

The 19 images shown in Table 11 received a mean Cute/Kawaii rating of 4.0 or greater from at least one female group (Japanese females or American females) and also had ratings that differed significantly between these groups.

Table 11. Comparison of Cute/Kawaii between Japanese and American females.

The 9 images shown in Table 12 received a Cute/Kawaii rating of 4.0 or greater for at least one male group (Japanese males or American males) and also had ratings that differed significantly between these groups.

Table 12. Comparison of Cute/Kawaii between Japanese and American males.

In Tables 11 and 12, bold rows indicate images that appear in both tables. In each table a * indicates significance at the p < .05 level while a ** indicates significance at the p < .01 level.

For 13 of the 19 images in Table 11, Japanese females rated the image significantly higher than American females with regard to Cute/Kawaii. For the remaining 6 images, American females rated the image significantly higher than Japanese females with regard to Cute/Kawaii.

For 8 of the 9 images in Table 12, American males rated the image significantly more highly then Japanese males with regard to Cute/Kawaii. For the remaining image (a female Japanese model), Japanese males rated the image more highly with regard to Cute/Kawaii than American males.

For the images presented in this study, the differences between Japanese females and American females seem more varied with regard to perceptions of Cute/Kawaii as compared to the differences between Japanese males and American males. For images that at least one male group finds cute (mean rating of at least 4.0) the groups only differ for images representing animals or people. Additionally, as shown in Table 12, when the male groups differ, American males are more likely to give higher Cute/Kawaii ratings (with an apparent exception when they are rating adult females).

As shown in Table 11, ratings of Cute/Kawaii differ more frequently for female groups. In addition to images of animals and people, image of characters, sweets and toys were rated significantly differently with respect to Cute/Kawaii by Japanese females versus American female. Additionally, American males and females seem less likely than Japanese males and females to use “cute” to describe adults.

Evaluating an image in the context of Kawaii for Japanese females may not be just a matter of observing the characteristics of the two-dimensional photo. Rather, it is possible that Japanese females look at familiar “Kawaii” photos in richer contexts. For example, the group of Japanese female participants highly rated a pop group, Perfume, with respect to Kawaii, while Japanese males did not. Perfume is a well-known celebrity group active in music and TV programs. Japanese females might have associated the photo of the group with other attributes, which brought a wider sense of cuteness.

5 Future Work

Given the increasing global popularity, including popularity in the United States, of Japanese products that exhibit Kawaii, it would be interesting to perform a cross-cultural comparison between American participants and Japanese participants with regard to ratings of Cute/Kawaii in physical representations of commercial products such as toys, school gear, sweets, clothing, home decor and electronic goods.

It would also be interesting to determine the specific attributes of products (color, size, shape, decoration, etc.) that lead to feelings of Cute/Kawaii. Previous work by Laohakangvalvit, Achalakul and Ohkura [5] has developed a preliminary model for Kawaii feelings based on ratings of pictures of various proposed spoon designs. Additional studies reported by Ohkura [6] have reported on affective values of Kawaii in physical products. All of this work would inform the proposed cross-cultural study of commercial products. Results of such a study could be useful to manufacturers who are interested in designing products that appear to consumers outside of Japan.