For the present study, three distinct sample recruitment processes were conducted. First, a focus group was assembled to identify the main domains of the virtual reality program. Second, a sample was recruited for a content validity study, and finally, a sample was gathered for the face validity study.

This study involved two interconnected studies: the first focuses on content validity, and the second on face validity. The overall objective was to assess the validity of the BSR-RV 2^® virtual reality program. A summary of all phases is shown in Figure 1 (Appendix).

To determine the content validity (Step 4) of the BSR-RV 2^® game scenes, a 9-item questionnaire, the Virtual Reality Content Validity Questionnaire, was created. This questionnaire measures both objective and subjective aspects of the game, such as gameplay, understanding of instructions, and usability of controls. The Cronbach’s alpha for this scale was α = 0.727, indicating acceptable internal consistency. This questionnaire evaluates the domains explained in Step 1.

For both the first and second studies, the Simulator Sickness Questionnaire (SSQ) (Kennedy et al., 1993) was administered using Spanish validation (Campo-Prieto et al., 2021) to assess the potential side effects of the game. The SSQ consists of three subscales: oculomotor, disorientation, and nausea. In the content validity study, the overall alpha was α = 0.886, with subscale alphas ranging from α = 0.593 (nausea) to α = 0.864 (disorientation). In the face validity study, the overall alpha was α = 0.818, with subscale alphas between α = 0.641 (nausea) and α = 0.755 (oculomotor). This questionnaire was applied in Steps 4 and 5 of this study.

To measure the degree of acceptance, understanding, and familiarity of users with the virtual experience, the GAMEX Index (Gameful Experience in Gamification), developed and validated in English by Eppmann et al. (2018) and further validated in Spanish by Parra-González and Segura-Robles (2019), was used. The scale consists of 27 items, with response options ranging from “strongly disagree” (1) to “strongly agree” (5) on a Likert-type scale. The questionnaire included five subscales: enjoyment (items 1–6), absorption (items 7–12), intrinsic motivation (items 13–16 and 24–27), activation (items 17–20), and absence of negative affect (items 20–23). The answers for items in the “absence of negative affect” subscale were recoded, meaning that higher scores on this factor indicated a lower incidence of negative affect. The original alpha value for the scale was above α = 0.90, and in the Spanish version, the alpha was above α = 0.80. In the face validity study, the alpha was α = 0.875. This questionnaire was applied in Step 5 of this study.

Descriptive statistics, including mean and standard deviation, were calculated for variables such as participant age, gender, and level of expertise in both the content and face validity studies. The Kolmogorov-Smirnov test was used to assess the normality of the distributions. Once normality was established, tests for mean differences and ranges were conducted using the t-test and Mann-Whitney U test.

For the content validity study (Step 3), the Content Validity Ratio (CVR) and the Item-Level Content Validity Index (I-CVI) were computed using data from the Virtual Reality Content Validity Questionnaire and the Simulator Sickness Questionnaire (SSQ). Non-parametric tests, specifically the Mann-Whitney U test, were applied to analyze differences by level of expertise.

In the face validity study, both the impact score (Zamanzadeh et al., 2015) and the Item-Level Face Validity Index (I-FVI) (Yusoff, 2019b) were calculated using data from the GAMEX and SSQ. Parametric t-tests, including independent sample t-tests and paired sample t-tests, were used to examine differences across gender, age, Age × Gender interaction and number of trials. Chi-square tests were conducted to analyze the influence of age groups on face validity outcomes. The effect sizes were calculated using Cohen’s d and Cramer’s V was calculated to determine the effect size for chi-square tests.

All inferential analyses were conducted with the significance level set at p < 0.05. The statistical software SPSS version 27 was used for all analyses.

Following a comprehensive literature review, we have selected the following domain-specific terms for further exploration: “attitudes towards disability and wheelchairs” (195 articles), “skill acquisition through virtual reality” (1169 articles), “content validity and face validity in virtual reality” (414 articles), “virtual simulation skill acquisition” (43 articles), and “advanced methodology in virtual reality and disability” (34 articles).

We have also reviewed 11 articles related to the constructs of “attitudes towards disability of people and wheelchairs” (Carter et al., 2014; Evans et al., 2015; Grenier et al., 2014; Lundberg et al., 2008; Matera et al., 2021; Medland and Ellis-Hill, 2008; Moss et al., 2020; Sapey et al., 2005; Sofokleous and Stylianou, 2023; Tamm and Prellwitz, 2001; Vilchinsky et al., 2010).

In addition, 11 studies on “skill acquisition through virtual reality” were reviewed (Alsalamah et al., 2017; Alvarez-Lopez et al., 2020; Babar et al., 2023; Bright et al., 2012; Chuan et al., 2023; Dorozhkin et al., 2017; Feeley et al., 2021; Harris et al., 2021; Runswick, 2023; Shen et al., 2022; Willaert et al., 2012).

We have also examined 15 articles on “content validity and face validity in virtual reality” (Carter et al., 2005; Dulan et al., 2012; Feeley et al., 2021; Harris et al., 2020; Hung et al., 2011; Kelly et al., 2012; Kenney et al., 2009; Marlenga et al., 2017; McDougall et al., 2006; Morgado et al., 2023; Rosenberg et al., 2013; Schwebel et al., 2008; Seixas-Mikelus et al., 2010; Verdaasdonk et al., 2006; Xiao et al., 2014). These articles discussed the perceptual limitations, sensory appearance, and degree of realism associated with virtual reality.

In conclusion, the literature on content and face validity in virtual reality in the educational and psychological fields is limited, although there are more references in the medical field.

A table of critical CVR values computed by Lawshe (1975) was adopted, in which it was assumed that the level of agreement exceeded that of chance, with values ranging from 0 to 1. A critical CVR of 0.42 is assumed for a value of 20 experts (Lawshe, 1975; Romero Jeldres et al., 2023). A critical CVR value of 0.42 is assumed for 20 experts (Lawshe, 1975; Romero Jeldres et al., 2023).

In the face validity study (Step 5), 395 children (lay experts) were recruited to evaluate the items. The formula proposed by Yusoff (2019b) was applied, with a criterion of 0.79 as a minimum value to maintain the item. Subsequently, the results were compared with those of other randomly selected small samples (N = 30, N = 60, N = 90, N = 300 and N = 360) to determine if sample size affected I-FVI regarding the Age × Gender interaction across the GAMEX questionnaire according the criteria of the original authors (Eppmann et al., 2018).

Before establishing the Content Validity Ratio (CVR), the impact of the level of expertise (lay experts versus content experts), gender, and age was evaluated. The responses from the content and lay experts groups were compared, and no differences were observed in the “Virtual Reality Content Validity Questionnaire.”

The responses between the content and lay experts groups were analyzed comparatively and no differences were found in the nine variables except for the item “the objects in the game seemed real to me,” which had a better rate of the content experts versus the lay experts (U = 19,5, Z = −2.193, p = 0.028).

In relation to side effects, no relationship was observed between the level of expertise and side effects, except for item 7 “sweating” (U = 28.00, Z = −2.21, p = 0.027) which affects the content expert group. Age correlated with item 10 “headache” (Rho = 0.46, p = 0.041). Finally, female gender correlates with item 7 “sweating” (U = 16.00, Z = −2.902, p = 0.004).

Items 1, 2, 5, 6, 7, and 8 had CVR values greater than 0.42 (Lawshe, 1975; Romero Jeldres et al., 2023) and I-CVI values higher than 0.79 (Yusoff, 2019a). The primary outcomes of this study are presented in Tables 2, 3, respectively.

The decision was made to evaluate the items that impacted the “Virtual Reality Content Validity Questionnaire” scale, which had a validity of less than 0.42 in the CVR values (Lawshe, 1975; Romero Jeldres et al., 2023) and less than 0.79 in the I-CVI values (Yusoff, 2019a). Consequently, modifications were implemented to the sound and music of the various scenes (volume of the soundtrack, type of music presented), as well as the instructions (visual position and information displayed to the user, type of instructions given to the user), adaptations related to the operation of the controllers to improve the users’ intuitiveness of the simulation (ease of moving the wheelchair, the way the user could take and pass the ball, ease of throwing the ball into the basket, and natural movement of the controls to enhance the user’s understanding of the simulation’s functioning) (Figures 3, 4).

In step five, a face validity study was conducted to evaluate the Validity Index using 395 lay experts. The sample size was reduced by 17 individuals in the GAMEX study due to unforeseen circumstances beyond the control of the researchers. The results of the study indicate that the GAMEX scale exhibits strong face validity in terms of enjoyment and absence of negative affect, but lower validity in factors such as absorption, intrinsic motivation, and negative activation (Table 4). The measurement of activation presents a methodological issue as it contains contradictory meanings in its items. The two positive activation items (“While playing I felt activated” and “While playing I felt excited”) and the two negative activation items (“While playing I felt jittery,” and “While playing I felt frenzied”) are problematic. If the I-FVI criterion proposed by Yusoff (2019b) is applied, the absorption, intrinsic motivation, and negative activation factors would not be considered adequate. However, if the Impact Score criterion is used, only the negative activation factor is considered inadequate. The low levels of agreement in negative activation observed are a positive finding, as they suggest low levels of jittery and frenzy.

The relationship between age groups and face validity assessed through the GAMEX was examined, revealing a correlation between age and absorption, intrinsic motivation (Items 13 and 24), and negative activation (Item 18) (Table 4).

A subsequent within-group analysis examining the Age × Gender interaction found variations within the 7–9 and 10–12 age groups. However, due to the small sample size of 12 years (N = 7), the 12-year-old group was excluded from the analysis (Figure 7). The chi-square analysis of Age × Gender interaction revealed that the highest degree of agreement in absorption occurred in the 8-year-old male group within the 7–9 years age group, and in the 11-year-old female group within the 10–12 years age group for item 7 (χ2 (36) = 53,641; V = 0.186, p = 0.030), item 8 (χ2 (36) = 55,768; V = 0.194, p = 0.019), Item 9 (χ2 (36) = 54,229; V = 0.192, p = 0.026), item 10 (χ2 (36) = 66,673; V = 0.212, p = 0.001), item 11 (χ2 (36) = 57,102; V = 0.197, p = 0.014), item 12 (χ2 (36) = 57,301; V = 0.197, p = 0.013), and item 24 (χ2 (36) = 56,417; V = 0.195, p = 0.016).

According to the criteria established by GAMEX Eppmann et al. (2018), we assessed the face validity based on the sample size. In our sample of 376 cases, we selected 30, 60, 90, 300 and 360 cases to determine if sample size affected I-FVI regarding the Age × Gender interaction across the GAMEX questionnaire (Table 5), with the 8 years old male age group showing the most notable differences in absorption.

For the purpose of analyzing the SSQ questionnaire, the impact score and the I-FVI were assessed using the criteria of Zamanzadeh et al. (2015) and Yusoff (2019b), respectively. In each instance, the scores were close to the minimum range, as indicated in Table 6.

The extent of agreement regarding side effects was assessed during the two trials. A t-test for paired samples indicated variations between the first and second trial in item 3 “headache” (t (393) = −2.73; p = 0.007, d = 0.812; 0.43 versus 0.38), item 5 “difficulty focusing” (t (388) = 3.509; p < 0.001, d = 0.881; 0.53 versus 0.38), item 9 “difficulty concentrating” (t (392) = 4.938; p < 0.001, d = 0.919; 0.55 versus 0.32), and item 11 “blurred vision” (t (392) = 2.743; p = 0.006, d = 0.846; 0.52 versus 0.40).

Table 7 displays the differences between trials and age groups (7–9 years old versus 10–12 years old) in trials 1 and 2. Although the differences were statistically significant, the level of symptomatology in both the trials was relatively low.

Analysis of Age × Gender interaction revealed significant differences in fatigue (χ2 (27) = 41,763; V = 0.189, p = 0.035) in the first trial, as well as in eyestrain (χ2 (27) = 45,458; V = 0.199, p = 0.015), difficulty in focusing (χ2 (27) = 40,283; V = 0.186, p = 0.048), increased salivation (χ2 (27) = 44,385; V = 0.195, p = 0.019), and dizziness with eyes open (χ2 (27) = 46,785; V = 0.200, p = 0.010) in the second trial. The differences were particularly pronounced among individuals aged 7 years, as shown in Figure 8. However, it should be noted that the values for all other Age × Gender interactions were very close to zero, indicating a negligible effect.

In our examination of the domains involved in constructing and validating virtual environments, we did not find any literature analyzing the content and face validity of virtual tools designed to change attitudes towards individuals with disabilities among typically developing children. However, several studies have focused on changing attitudes towards people with disabilities through virtual reality simulations (Ceberio, 2022; Chowdhury et al., 2017; Chowdhury and Quarles, 2022; Matera et al., 2021; Pivik et al., 2002), but they were not designed, created and validated through the children perspective.

Regarding the simulation of sports practice, a significant challenge is the scarcity of studies assessing the accuracy of children’s motor and movement skills. Most studies relate to pictorial images and do not focus on virtual reality (Lopes et al., 2016; Morgado et al., 2023; Valentini et al., 2018). The primary objective of the present study was to identify the different types of mastery. Those related to physical fidelity involve reproducing movements and motor skills related to chair movement, passing the ball between seated players, or changing the visual perspective when throwing the ball to the basket. The realism domain involves reproducing a game with two or more players. The goal is for a typically developing child to discover the limitations or barriers experienced by a disabled child when playing a wheelchair basketball game. In doing so, the child may change his or her perspective and promote inclusion in the future.

While the difficulty level of the motor and coordination activities in the different scenes of the study was inspired by general recommendations from medical research, it was based on several factors, including the sports hall environment (Kelly et al., 2022; Runswick, 2023), handling of tractors by children (Marlenga et al., 2017), and objectives and development of the present study, which appear to be novel. The wide range of elements considered in the study, such as familiarization with the movement of the wheelchair; the performance of two or more simultaneous tasks; and aesthetic aspects, sounds, images, or degree of realism adapted to the age of the participants, make it a unique contribution to the field. The target group consisted of children aged 8–12 years, and while there were few studies that analyzed domains adapted to developmental development, most notably, the works of Marlenga et al. (2017) and Schwebel et al. (2008), the age groups did not follow the developmental criteria proposed in the present study.

The development of virtual items in the present study involved a focus group consisting of adult players with over 12 years of experience in adapted basketball. Four participants agreed that the main thematic areas should focus on two domains: technical considerations and degree of difficulty and congruence with physical laws.

To the best of our knowledge, this is a pioneering effort to promote inclusion in children aged 8–12 years through an adapted basketball virtual reality game. The game features three main scenes, each comprising two subscenes with increasing difficulty levels. In the “Final Game,” users put into practice all they have learned previously. In addition, each user was provided with various options for wheelchair avatars at the beginning of the simulation to identify someone who used a wheelchair.

The primary focal point of the initial scene is the manipulation of the wheelchair, in which the user must traverse various obstacles while navigating the wheelchair towards a designated location. The subsequent main scene entails a passing drill, where the user acquires proficiency in catching the ball and simulating a pass to a teammate in both static and dynamic settings. Third, in the basket-shooting scene, the user must attempt shots at the basket from diverse angles and ranges, requiring the user to approach closer or farther, based on the difficulty presented. Lastly, all the learned skills are integrated in the “Final Game” scene, where the user must apply the knowledge acquired in the three main scenes previously described in a simulated game scenario.

The current study utilized children as experts for both content and face validity assessments. For content validity, their understanding of basketball and wheelchair basketball, as well as their familiarity with new technologies, was considered. In the face validity study, their expertise in new technologies was evaluated, considering their experience playing video games (VanDeventer and White, 2002) and the daily usage of these technologies (Hague and Payton, 2010; Plowman et al., 2012).

In both content validity and face validity studies, the number of participating children exceeded the number of experts usually recommended in studies with adults (Lynn, 1985). The value for content validity is of 20 experts, which suggests that the random inter-judge agreement could decrease. In our study and many other studies, the arbitrary limit proposed by Lynn (1985) has surpassed, with as many as 30 experts involved in some cases (Handage and Chander, 2021; Yadav et al., 2021). In the case of the content validity study conducted with children aged between 11 and 18 years, we agree with the proposal by Stewart et al. (2005), which suggests that the panel of experts should be expanded to avoid linguistic, conceptual, and contextual comprehension issues. Furthermore, the level of attention or concentration is influenced by the maturity of neural circuits that determine the level of attention.

The same holds true for the face validity study, which included 395 children aged between 7 and 12 years. The enlargement of the sample size was consistent with the recommendations of other authors (Stewart et al., 2005). To ensure the representativeness of the sample, it is crucial to include individuals from the target age group and gender. As cognitive and affective development can vary significantly between males and females, it is essential to account for these differences. The 395 participants were divided into two main groups: those aged 7–9 years and those aged 10–12 years. The sample was further divided into 10 subgroups, each containing no more than 60 cases, based on the Age × Gender interaction.

In contrast to studies that examine age and gender variables independently in the context of virtual reality for children (Bart et al., 2008; Marlenga et al., 2017; Schwebel et al., 2008), our study considers the Age × Gender interaction effect. This approach enabled us to examine the fluctuations in the critical values of I-FVI that depend on both variables more precisely.

The Virtual Reality Content Validity Questionnaire and the SSQ were employed to analyze the level of agreement regarding immersion, presence, physical and psychological fidelity, degree of realism, and the side effects associated with the use of the program BSR-RV 2^®. Notably, no significant differences were observed between the critical values of CVR, unless in the grade of realism, in which differences were observed between content experts (wheelchair basketball players) and lay experts (conventional basketball players) aligning with studies with experts and novices in the area of creativity (Kaufman et al., 2008). The findings of our study indicate that there are some variations in the scores obtained by the two groups in the Virtual Reality Content Validity Questionnaire, which we attribute to their level of familiarity with basketball. It is possible that some of the tasks in the program, such as chair movement, ball passing, and basket shooting, were not entirely new to some participants.

Regarding the SSQ, the CVR data showed a low incidence of cybersickness, with a high agreement on the low incidence of symptoms. The results revealed a virtual absence of general discomfort, salivation, vertigo, and burping, along with a low incidence of headache, difficulty in focusing, or eyestrain. Furthermore, the low intergroup differences in the SSQ may be explained by the results of Huppert et al. (2019), who reported that the degree of cybersickness is similar or decreases in the age group–12–16 years and older.

Overall, the CVR values were found to be lower than 0.42 in some instances, which necessitated the adjustment of the sound and music in the different scenes, as well as the instructions and adaptations related to the operation of the controllers to enhance the intuitiveness of the simulation. This is consistent with I-CVI values, although the magnitude of the critical values may be misleading in certain cases.

Step 5 involved conducting a face validity study that involved selecting 395 children aged 7–12 years as lay experts to assess the test (Nevo, 1985). The overall face validity indices, including the presence and psychological fidelity indices, varied. The critical values for the impact score were high for enjoyment and positive activation, whereas the levels of intrinsic motivation and absorption were moderate. However, the negative activation critical value was low, since most participants did not show this type of reaction.

The impact of the Virtual Reality program’s duration and number of trials on non-aesthetic domains, such as absorption and intrinsic motivation is likely to be reflected in the results obtained. To increase absorption, it is necessary to have longer sessions. The same can be said for GAMEX’s measure of intrinsic motivation, in which lower values were associated with achieving a higher degree of autonomy and sense of control. Given that the BSR-RV 2^® program included scenes of sporting activities with varying levels of difficulty, such as shooting a basket or moving a chair, a greater number of trials and longer duration would have been necessary to achieve a higher degree of control.

Although both the I-FVI and Impact Score indices are presented in this study, there is currently no consensus on the ideal index for assessing the face validity of a virtual tool, as there are studies that use the I-FVI proposed by Yusoff (2019b) as a reference (Alvarez-Lopez et al., 2020; Baharuddin et al., 2024; Mohamad Marzuki et al., 2018), whereas others use the impact score proposed by other authors (Abdollahipour et al., 2016; Bek et al., 2009; DunnGalvin et al., 2008; Lacasse et al., 2002; Zamanzadeh et al., 2015). However, other studies have established the mean between groups of experts to certify their validity (Alsalamah et al., 2017; Alzahrani et al., 2013; Chuan et al., 2023; Evans et al., 2015). Given this lack of consensus, the present study has decided to show both values, despite the fact that these values have been determined with a lower sample size in previous studies (Yusoff, 2019b; Zamanzadeh et al., 2015). Even so, the vast majority of the values presented in the face validity study were still acceptable. Other studies, however, have simply established the mean between groups of experts to certify their validity (Alsalamah et al., 2017; Alzahrani et al., 2013; Chuan et al., 2023; Evans et al., 2015).

Regarding the cybersickness scale SSQ, there were very few differences between T1 and T2 in terms of the symptoms, except for headache, difficulty focusing, difficulty concentrating, and blurred vision. These symptoms showed a significant decrease within the absence and mild range. However, the values obtained on the SSQ in both the I-FVI and impact score were higher than acceptable in all items. It is important to note that the neurovegetative symptomatology was practically absent in all cases.

The issue with this study lies in the analysis of the results based on the I-FVI on the GAMEX scale, which yielded values below the accepted threshold of 0.79, as proposed by Yusoff (2019b). Upon examining the studies that contributed to this figure, it is evident that the majority of them involved adult participants. The situation changes when analyzing studies involving children of a similar age and subject matter to our own study. In research on the validity of non-virtual pictorial instruments for games or sports involving children, the face validity scores are typically lower than 0.79. This is true in studies assessing sports physical competencies with samples larger than 200 cases (Lopes et al., 2016), as well as in studies on performance in aquatic physical activities (Morgado et al., 2023), movement actions (Moulton et al., 2019), and serious games (Sim et al., 2015). The same trend is observed in virtual reality studies with scores below 4 on a 5-point Likert scale, such as the study by Marlenga et al. (2017), Schwebel et al. (2008), and Veerman et al. (2024).

In certain studies, when the values were lower than anticipated (I-FVI <0.80; Kappa <0.70), some items that violated these principles had to be revised or retained (Schilling et al., 2007). This discussion has also been raised in other academic disciplines such as marketing, where the percentage of experts required to agree on retaining an item can vary significantly, ranging from 60% to 80% (Hardesty and Bearden, 2004).

To determine whether the outcomes were influenced by the sample size and to adhere to the criteria suggested for evaluating the construct validity of the GAMEX (Eppmann et al., 2018). The results reveal that the use of small sample sizes (N = 30; N = 60; N = 90) do not detect some differences in the degree of agreement, especially on cognitive variables (e.g., absorption) when considering the Age × Gender interaction. However, this issue is resolved with larger sample sizes (N = 300; N = 360). This result is crucial as it calls into question the traditional reliance on non-parametric samples to determine face validity with children. Therefore, the greater intra- and intergroup variability in their results suggests that enlarging the sample size is necessary to avoid inconsistencies and identify the developmental subgroups.

The validity scores obtained from the group of child experts were likely influenced by the unique psychological processes associated with each developmental stage. As a result, it was deemed necessary to examine the values derived from the SSQ and GAMEX using methods that focused on both inter- and intra-group differences.

When we examined the critical values (I-FVI and Impact Score) in relative terms, we observed variability in the age-gender interaction, particularly in absorption and negative activation. The comparison between the 7–9- and 10–12-year-old groups revealed slight inter-group and intra-group discrepancies. Overall, the 7–9-year-old group exhibited a higher degree of absorption (items 8–12), sense of dominance (Item 24), and negative activation (Items 18 and 19). The level of agreement was higher in 7-year-old females and 7- and 8-year-old males in terms of absorption, reaching close to 73%. This is significant because absorption is linked to the level of tonic activation and attention. Children in this age group displayed greater engrossment or preoccupation with the objects and images in the BSR-RV 2^® program, often disregarding other aspects of their surroundings (“Playing the game made me forget where I am”) and, in some cases, exhibiting dissociative signs.

Dissociation that is not pathological can result in experiences where children become fully engaged and focused, leading to a temporary loss of awareness and decreased monitoring of internal emotions and external activities (Eisen and Lynn, 2001). Children who score higher in this area (between 7–8 years old) exhibit characteristics of the pre-operational stage of cognitive development, a period marked by egocentrism that can make it challenging to distinguish between one’s own thoughts and external reality (e.g., images and movies). Our findings align with those of Beatini et al. (2024), which suggest that perspective-taking and changes in egocentrism occur around the age of nine. As such, we propose that absorption may represent typical manifestations of infantile cognitive structures and a normative regulatory strategy, which may be influenced by neurobiological development in 7–9 to year-olds. For instance, in a study using virtual reality by Baumgartner et al. (2008), prefrontal brain activation was less pronounced in the 8–11 age group.

Few studies have examined the impact of virtual reality on child development from a comparative perspective. Our research aligns with the findings of Baughan et al. (2024), which link high absorption in VR to a lack of attention to internal sensations and external interruptions. However, they did not explore the relationship between absorption and age. In contrast, Nakayama et al. (2020) observed that children aged 5–8 years were more susceptible to problems related to video games than those aged 10 years.

Two sessions were conducted to measure cybersickness using the SSQ, separated by a 1-week interval. This approach was used to control for the effects of phasic activity or fluctuations over time that may occur spontaneously or in response to an event. These fluctuations are changes in the autonomic nervous system that were assessed through symptoms associated with cybersickness and negative emotions, such as “jittery” and “frenzied.” The agreement on negative activation in the GAMEX questionnaire was higher for both females and males at ages 7 and 8 years than for the 9–11 age group. Significant intra-group variations were observed.

The findings from the SSQ displayed few inconsistencies between the two trials (trial 1 and trial 2) for the age groups of 7–9-year-olds and 10–12-year-olds. However, an analysis of the effects of age, gender, and trial revealed minor differences in the level of agreement for symptoms such as eyestrain, difficulty focusing and concentrating, vertigo, and blurred vision. Children aged 7–8 years had a higher incidence of these symptoms. The comparative analysis by Age × Gender interaction showed irregular results, demonstrating a significant decrease in agreement from the age of 9 years onwards, both in males and females. Our results align with those of other studies (Baumgartner et al., 2006; Huppert et al., 2019; Nakayama et al., 2020), which found that the age group with the highest incidence of side effects was 8 years, corroborating the data presented in our study. The reduced awareness of bodily signals in the age group of 9–11 years is similar to the results obtained in studies with adults who played videogames, which showed various forms of reduced awareness of bodily signals (Swinkels et al., 2021), and with children over 10 years old (Nakayama et al., 2020).

Generally, the findings do not support the developmental range suggested by Staudinger et al. (2003) for 7–9-year-olds and 10–12-year-olds, as the face validity values indicate greater homogeneity between the two age groups. Specifically, the highest values of I-FVI in phasic activation (cybersickness) and tonic activation (degree of absorption) suggest a critical moment at age 7–8 in males, whereas a similar trend is observed in females of the same age range, although at a lower intensity. These findings are consistent with those of studies that examine the impact of gender on video game use (Martucci et al., 2023).

Depending on age and gender, psychological transitions from sensation to perception, perception to thought, and thought to action exhibit variations. Physical sensations generally do not interfere with the children’s evaluation of the scenes, except for a small subgroup of 7–8-year-old males, possibly linked to the low incidence of adverse effects (e.g., nausea, dizziness). Additionally, the limited concern associated with these effects may have contributed to improved attention levels, particularly in 10–11-year-old males. The children’s identification with the scenes appears to have been enhanced by their enjoyment of the immersive environment, with a stronger relationship observed in 10–11-year-old males. This, in turn, may have positively influenced higher levels of absorption and intrinsic motivation among the children.