Journal of information and
communication convergence engineering
(J. Inf. Commun. Converg. Eng.)

Through play, children naturally learn various things, such as rules and skills, and continue to grow and develop themselves [1].

In the digital environment according to the development of IT technology, there have been various approaches to play and cultural enjoyment, and the existing analog types of play and culture have evolved in combination with IT technology.

Recently, the expansion of the experience area using immersive media and the feedback of immediacy and simultaneity, which are characteristics of media, have increased immersion. Children form their own sense of subjectivity, resulting in high satisfaction with experiential learning [2].

Therefore, there is increasing interest and demand for the effect of media conversion and experience development of play using immersive media on children; however, systematically applied research in the field of child education has not kept up with the development of play that incorporates the latest IT technology [3].

Although digital content has sufficient value in terms of play and educational effects for children, it is necessary to explore the medium in terms of content to make digital media more advanced and useful for children.

In this study, we attempted to determine whether threedimensional puzzles [4], i.e., the analog types of block building and assembly play that children prefer most, have a significant effect on children’s experience as they are converted into immersive media.

First, the researchers considered the play theory of Roger Caillois and the concept of puzzle play through theoretical consideration and examined the various interactive characteristics of puzzles as trends in preceding research and development along with the changes in children's puzzle play. After the consideration stage of preceding research was summarized, a questionnaire was prepared based on a survey filled out in preceding research on the play properties of Roger Caillois. The questionnaire was given to children who experienced puzzles at the Children’s Cultural Center of the Asia Culture Center, where both analog and immersive media puzzles can be experienced, in an experimental group and control group, and significant differences in play properties were tested. Finally, we checked the significant change effect of play properties by converting them to immersive media through the analysis results and suggest future research directions.

A. Play Theory of Roger Caillois

Play is an instinctive human behavior that has evolved from the beginning of human civilization. The main purpose of play is to experience. The joy of play revives excitement and adventurous spirit, arouses optimistic thoughts, and enriches human life as a driving force for creativity and innovation [5]. Such play enriches human life and has important value. That is why the academic theory of play has also been defined and studied by philosophers, thinkers, and scholars. Roger Caillois, a representative sociologist of play theory, critically inherited and developed the theory of Johan Huizinga, who defined play academically and suggested four concepts of play: Agon (competition), Mimicry (imitation), Area (luck), and Ilinx (vertigo), in addition to the essence of play [6]. Fig. 1 rearranges the rules of the four properties of play, giving the direction for strengthening human will.

Figure 1. Roger Caillois’ four concepts of play, rules of play, and direction of strengthening will.

Table 1 describes each property of play. Baseball and basketball, where there is an opponent and players compete with their will to win the target, are examples of Agon. Area includes random and accidental elements in the rules, such as drawing or diceplay. Mimicry is something that imitates other characters without any rules, such as imitation, house play, and mask play, while Ilinx causes extreme dizziness while relying on something without rules, such as a Viking or roller coaster.

Table 1 . Properties of play.

Division	Explanation	Origin
Agon	Execute at your will in the rules	Greek ‘game’
Mimicry	Enjoy only by your will without any rules	English ‘mimicry, imitation, denial’
Area	Enjoy enforced by rules, enjoy random results in rules	Latin ‘incident’
Ilinx	Rely on something without rules, naturally enjoying	Greek 'swirl'

The four properties of play can be classified according to the player’s willingness/unwillingness and the regulation/ deregulation of the play itself, which is not necessarily represented by one property; rather, each property is combined to explain the characteristics of the play [7]. When the properties of play are evenly distributed, it is important to increase the level of immersion by doubling the fun of play [8].

Recognizing play as a cultural phenomenon, Roger Caillois argued that the element of play was found at the beginning of all forms of culture and that human communal life itself is a form of play [9]. In other words, Roger Caillois discussed the nature of the formal part of play as a general phenomenon of a community rather than its content.

He defined essential play as separate, free, undefined, rule-based, unproductive, and fictitious activities [10].

Play is a voluntary activity by human free will within restrictions on time and space that are distinct from daily life, and it is a source of amusement that gives joy and fun, which are distinct from labor.

B. Hands-on Exhibitions for Children and Media Conversion

Hands-on exhibitions refer to exhibits that enable physical interaction based on clear educational goals and activities that can produce various results according to the audiencés personal interpretation [11]. A hands-on exhibition is an exhibition method that allows direct experience, meaning hands-on and interactive or interaction type exhibitions. It is also used as a term interchangeable with one concept [12]. Hands-on exhibitions evolve with the development of media, as shown in Fig. 2. Therefore, analog hands-on exhibitions are often reprocessed and provided by converting media in line with the period of technological development. Medium conversion literally refers to the conversion of one piece of content to another medium and is widely used in hands-on exhibitions for children to expand their user experience.

Figure 2. Immersive media of children's experience center by media development.

The Children’s Experience Center is a place that stimulates children’s curiosity and imagination and enables them to have real-life experiences. Hands-on exhibitions at the Children’s Experience Center, which are the most widely used in Korea, are mainly divided into two types: direct and indirect participation. Direct experiences are divided into exhibition operative, interactive, participatory, demonstrative, experimental, playing, and on-site experiential, whereas indirect experiences are divided into video, diorama, and special directing exhibitions [13]. However, the classification of hands-on exhibitions highlights the need for new systematization in line with the development of contemporary science and technology and the evolution of exhibition techniques. Isak Kim applied Tim Caulton’s hands-on classification and reflected Roger Caillois’ play studies on hands-on exhibition types, as shown in Table 2 [14].

Table 2 . Experiential exhibition type classification according to experiential maturity.

Hands-on	Reactive
Exploratory	Operative
Role playing
Expressive
Riding
Participatory		Reproductive
Game type		Appreciative
Random		Demonstrative

Hands-on exhibitions are subdivided into seven types, as shown in Table 2, and are produced in a variety of formats according to exhibition techniques, such as physical contact or virtual reality using digital media. Each type of hands-on exhibition uses elements of play in a complex manner, which can be considered a playful feature.

Fig. 3 summarizes the types of hands-on exhibitions according to the distribution and combination of play elements.

Figure 3. Distribution of play elements and experiential exhibition.

C. Immersive Media and Three-dimensional Puzzles

Immersive media satisfy users by conveying information through the five senses to maximize immersion and realism [15]. Immersive media include virtual reality, augmented reality, and mixed reality, which provide users with an immersive experience and have recently been referred to as extended reality [16]. With the commercialization of immersive media technology, it has been widely used throughout the cultural content industry. In line with this reality, leading museums worldwide have aimed to present various hands-on exhibitions using immersive media over the last 5-6 years. Experience is a constructive activity that accumulates through interactions that occur when users actively participate in the communication process with the content [17]. Overall, using immersive media provides an interface that enables various interactions and supports the expansion of children’s experiences to achieve holistic development.

As the children’s market has grown since the 1990s, interfaces for children’s interaction have gained recognition for potential future customers, and methods for identifying children’s needs and objective evaluation guidelines for interfaces have been developed [18]. Interfaces for children are developing and diversifying into physical spaces for analog experiences, mixed use of physical and electronic space, electronic space, and omnipresent space for immersive media experiences [19].

Table 3 shows the development status of puzzles, a representative component of play for children. The puzzle involves moving the shape of a figure left and right for matching, and the experiencer recognizes that a certain shape can be formed as the pieces are combined [20]. Among the types of play that help children’s development, puzzle play, in which three-dimensional structures can be constructed, is the most enjoyable activity for children [21]. By leading various compositional tasks through puzzle play, children experience social learning opportunities as well as good opportunities in math, science, and reading [22].

Table 3 . Puzzle experience at Children’s Experience Center.

In physical space, which is an analog experience, the puzzle itself is used as an interface to be assembled by hand.

Immersive media experiences can be divided into mixed use of physical and electronic spaces, electronic spaces, and omnipresent spaces. A puzzle that combines physical and virtual space shapes an arbitrary piece and completes it in the physical space; when linked to the virtual space, the shape of the puzzle changes into various shapes that symbolize and move. Puzzle matching in a virtual space is controlled by finger touch, and the game is completed step-bystep. In the omnipresent space, play controls and reacts according to the movements of children or the assembly of puzzles using sensor operation.

A. Study Subjects and Experimental Design

The target institution for the survey in this study was the Children’s Cultural Center of the Asia Culture Center, where both analog and immersive media experiences are available. The survey was conducted on children between the ages of six and nine. The content of the questionnaire was explained through basic interviews with the docents operating the hands-on exhibition for children, organizing 49 children who could fill out the questionnaire. The research groups comprised the same children, who were organized into 49 persons in the control group experiencing analog puzzles and 49 persons in the experimental group experiencing immersive media puzzles.

In this study, the earthenware matching puzzle was selected as one with an almost identical puzzle shape and rules of play in omnipresent space as an immersive media experience and physical space as an analog experience of the Children’s Experience Center. However, there is a difference in interface operation owing to the change in the interaction caused by media conversion.

Table 4 . Group scheme.

Comparison group	Experimental group	Place
Build a physical space puzzle (A)	Build an omnipresent space puzzle (B)
49 persons	49 persons	National Asian Cultural Center Children’s Cultural Center
49 persons

Table 5 . Analog and immersive media puzzle experiences at the Children’s Experience Center.

B. Research Method

The research questions of the survey for empirical analysis of the core hypotheses of this study were as follows:

Research Question 1: Is there a significant difference in Agon among the properties of Roger Caillois’s plays?

Research Question 2: Is there a significant difference in Area among Roger Caillois’ play properties?

Research Question 3: Is there a significant difference in Mimicry among Roger Caillois’ play properties?

Research Question 4: Is there a significant difference in Ilinx among Roger Caillois’ play properties?

The categories of the survey were evaluated based on the 5-point Likert scale in Table 5, and eight questions were formulated based on the four play properties of Roger Caillois in Table 6.

Table 6 . Characteristics of the sample.

Not at all	⇠	Average	?	Very true
1	2	3	4	5

The survey contents in Table 7 were based on Cho Okhee's questionnaire [23] and Kim Seo-In's questionnaire [24], which produced a measuring tool using the above-mentioned usability.

Table 7 . Questionnaire for usability assessment.

Division		Question
Research question 4	Agon	Use List 1- When I did puzzles, did I want to do more puzzles than my friends? Use List 2- Did I try to finish the puzzle in 10 minutes?
	Area	Use List 3- Did I line the puzzle pieces correctly during my experience? Use List 4- Did I fit the puzzle size in my experience?
	Mimicry	Use List 5- Did I feel like someone else (creator, architect, etc.) was in my experience? Use List 6- Did it feel like putting together a puzzle based on my experience?
	Ilinx	Use List 7- Did I get excited about the experience? Use List 8- Was it amazing to me that puzzle play felt real at the moment?

D. Research Procedure and Data Processing

Children were asked to experience A (physical space), followed by B (omnipresent space). The researcher conducted the experiment with the help of a docent, and both the experimental and control groups had the same experience. After the experience, the children were asked to complete questionnaires for A and B based on their experiences. When the questionnaire content was difficult, the children listened to the docent’s explanation and checked the questionnaire items. For data processing in this study, two experiences were conducted for the same children in a control group and experimental group using the statistical program SPSS18.0 before the survey, and Cronbach’s alpha coefficient was used for the questions.

In this study, all measurement variables showed a Cronbach’s alpha coefficient of 0.5 or higher in reliability verification, indicating that the measurement items could be trusted. Cronbach’s alpha coefficient is generally considered reliable when it is 0.5 or higher at the group level and 0.9 or higher at the individual level [25].

A. Results and Analysis of Play Properties

The analysis of the research results is shown in the following tables:

Table 8 . Reliability verification.

Variable		Number of first questions	Number of items removed	Number of final questions	Cronbach’s alpha
A	Agon	2	0	2	.865
	Area	2	0	2	.773
	Mimicry	2	0	2	.865
	Ilinx	2	0	2	.651
B	Agon	2	0	2	.840
	Area	2	0	2	.648
	Mimicry	2	0	2	.840
	Ilinx	2	0	2	.858

Table 9 . Sample group statistics.

variable		N	Average	Standard Deviation	Standard error of the mean
Agon	A	49	3.4694	1.06755	.15251
	B	49	3.8061	1.01968	.14567
Area	A	49	3.2041	1.07490	.15356
	B	49	3.8061	1.01968	.14567
Mimicry	A	49	3.2041	1.07490	.15356
	B	49	3.8061	1.01968	.14567
Ilinx	A	49	3.2551	1.02125	.14589
	B	49	3.8061	1.01968	.14567

Table 10 . Sample correlation coefficient.

N		N	Correlation	Significance level
Agon	A+B	49	.616	.000
Area	A+B	49	.526	.000
Mimicry	A+B	49	.526	.000
Ilinx	A+B	49	-.257	.075

Table 11 . Sample population statistics.

Group		Difference in Response		T	Degrees of freedom	Significance level (two-tailed)
		95% confidence interval of the difference,
		Lower bound	Upper bound
Agon	A+B	-.59958	-.07389	-2.576	48	.013
Area	A+B	.30891	.89517	4.130	48	.000
Mimicry	A+B	-.89517	-.30891	-4.130	48	.000
Ilinx	A+B	.08635	1.01569	2.384	48	.021

In the analysis of the results combining Tables 9, 10, and 11, the correlation coefficient of the corresponding paired samples of variables A and B for the Agon play attribute with Roger Caillois was .616 indicating a strong positive correlation. The mean difference and confidence interval between the two groups did not include zero, suggesting that the Agon play attribute in Group B was significantly improved compared to Group A. Furthermore, the result of the two-sample t-test shows a p-value (two-tailed) of .013< 0.05, which is less than the significance level of the 0.05 confidence interval (95%), indicating that the mean difference between the two groups was statistically significant.

Area shows a moderate correlation of .526 between the two variables A and B in the paired samples. Because the difference in the means and confidence intervals of the two groups does not include 0, it can be concluded that B has improved compared to A for Area. Furthermore, the mean difference between the two groups was statistically significant at a significance level of 0.05, as the two-sided p-value from the t-test was .000 < 0.05. Mimicry showed a moderate correlation coefficient of .526 between variables A and B in the paired sample. As the difference in means and confidence intervals between the two groups does not include 0, it can be concluded that Mimicry has improved more in B than A. Moreover, the t-test indicates a significant difference between the two groups at the 0.05 (99%) significance level, with a significance probability of .000 < 0.05. Ilinx showed a weak negative correlation, with a correlation coefficient of −.257 between variables A and B. The difference in means between the two groups had a 95% confidence interval of .08635 to 1.01569, which does not include 0, indicating that Ilinx has improved more in B than A. Furthermore, with a two-tailed t-test, the p-value was .021 < 0.05, so the mean difference between the two groups was significant at a significance level of 0.05 (95%). In the statistical analysis, greater emphasis was placed on the significance of the overall formula than on the detailed results. In this context, because there is no zero in the 95% confidence interval in the paired sample test, the results regarding the payability of the three-dimensional puzzle are significant. An empirical study found that the three-dimensional puzzle in electronic space through media transformation showed improvements in Roger Caillois' theory of all four play attributes (Agon, Mimicry, Area, and Ilinx). These results suggest that playing in virtual reality can enhance fun and enjoyment for children.

The purpose of this study was to determine whether there is a significant difference in the effectiveness of the play properties of changed interaction after converting media by combining three-dimensional puzzle play preferred by children with immersive media through correlation and paired ttest analysis.

The results showed that the properties of play were better and more evenly distributed in the puzzle experience in the omnipresent space than in the physical space, suggesting that play properties evenly distributed in the puzzle experience in the omnipresent space can have a positive effect on experiential learning and holistic development by increasing fun and concentration in children's experiences. The pleasure of this content appears to be of great interest to child experiencers. Therefore, digital media that can stimulate children's interest and curiosity from an instrumental perspective will require active research in various fields with purposefulness, as well as the potential for the holistic development of children.

In future, we will suggest a specialized space for the holistic growth of children by studying a digital experience exhibition model in which play elements are composed in a balanced manner.