A Study of the Impact of Virtual Reality Technology on Immersive Language Learning in Higher Education English Education

The experiment selected 122 students from University H as the subjects, who were from different majors and all of them took English courses in the same semester. Before the course started, the staff did not inform the participants about the content of this experiment, and after the experiment was finished, 122 questionnaires were collected. The experimental subjects were randomly assigned to four groups according to the two spatial abilities and teaching methods, and each group of experiments was conducted separately and independently of each other, in which one group (high spatial ability) and two groups (low spatial ability) were the experimental groups using virtual reality technology, and three groups (high spatial ability) and four groups (high spatial ability) were the ordinary groups with traditional teaching.

The mediating effect reflects the influence effect played by variable M in the path of influence of independent variable X on dependent variable Y. The mediating effect path is shown in Figure 1.

Figure 1.

Analysis of mesomeric effect Path

The test model for the mediation effect is as follows: 1Y=cX+e1$$Y = cX + {e_1}$$ 2M=aX+e2$$M = aX + {e_2}$$ 3Y=c′X+bM+e3$$Y = c\prime X + bM + {e_3}$$

The coefficient c, coefficient a, coefficient b and coefficient c′ reflect the effect of the independent variable on the dependent variable under the different models above.

This experimental immersive VR is realized by a panoramic VR all-in-one machine, which restores the English learning scene through panoramic roaming, obtains an immersive learning experience, and allows students to complete the study and learning activities without leaving home. Immersive VR devices can be used in different modes of use (free viewing by students, unified control by teachers) according to the teaching purpose, and in order to let the learners focus on the course content, this experiment adopts the unified control mode of teachers [24]. Specific process: the teacher opens the central control system, selects the joint broadcast mode, organizes the students to receive the VR helmet, and after ensuring that all the students are successfully connected to the terminal equipment, the teacher selects the English teaching resources to play. Students in the process of experiencing, through the helmet on the function buttons on the virtual scene of things to operate, after selecting a thing, there will be a relevant introduction to the voice or text in the way of the screen for students to understand and learn, in the completion of the visit to the study, the students return the helmet, the teacher closes the system.

A complete English course teaching is divided into two parts, the theoretical learning process is carried out in the traditional classroom, and the teaching is mainly completed through textbook lectures and PPT presentations. Figure 2 shows the teaching design framework.VR practice teaching is carried out after the theoretical learning process, as a useful supplement to traditional classroom teaching, and its main purpose is to strengthen learning, promote the understanding of the learning material, and stimulate emotions in a deeper way. This paper is aimed at testing the role of VR practical teaching in promoting learning effectiveness. Therefore, the experiment focuses on the VR practical teaching component.

Figure 2.

Teaching design framework

At the end of the VR experimental study, the subjects were brought to the post-test area next to the experiment, firstly, the experimental staff distributed questionnaires to the subjects and introduced the precautions for filling in the questionnaires, then the subjects were guided by the experimental staff to fill in the questionnaires, including the questionnaire of the English experimental test, the questionnaire of the cognitive load and the questionnaire of the motivation to learn; finally, the questionnaires were retrieved and carefully checked each questionnaire to make sure that the key information hadn’t been missed, and all the All the questionnaires were required to be completed within a specified period of time, and the subjects were given snacks as a reward after completing the questionnaires.

The results of the learning effect test for high and low spatial ability in the experimental group and the general group teaching style respectively are shown in Table 2 below. From the table, it can be learned that the learning effect test results of high spatial ability learners are 22.51±5.11, and the learning effect test results of the experimental group are 23.34±4.77, while the learning effect test results of low spatial ability learners are 16.45±4.02, and the learning effect test results of the experimental group are 19.53±4.58;When the experimental subjects were in the general group, the learning effect test scores of the high spatial ability learners were 22.51±5.11, and the learning effect test scores of the low spatial ability learners were 16.45±4.02; when the experimental subjects were in the experimental group, the learning effect test scores of the high spatial ability learners were 23.34±4.77, and the learning effect test scores of the low spatial ability learners were 19.53±4.58.

The results of the ANOVA of teaching style (ordinary group vs. experimental group) and learners’ spatial ability (high vs. low) are shown in Table 3. The results in the table show that the main effect of teaching style was significant, F(3,156)=16.28, p<0.01, np²=0.091, and the experimental group’s score on the language learning effect test was significantly higher than the ordinary group’s score on the language learning effect test; the main effect of spatial competence in was significant, F(3,156)=37.33, p<0.01, np²=0.188, and high spatial ability learners scored significantly higher than low spatial ability learners scored on the Language Learning Effectiveness Test (LLET); the interaction between teaching style and spatial ability learners on LLET scores was significant, F=13.67, p<0.01, np² = 0.079.

The interaction analysis between teaching style (experimental group vs. ordinary group) and learners’ spatial ability (high vs. low) is shown in Figure 5. For the low spatial ability learners, when the teaching styles were presented in the general group and in the experimental group respectively, the difference in learning effects was significant, F(1,156) = 35.21, p<0.01, np² = 0.290. The experimental group presented significantly better language learning effects (21.14±3.67) than those of the general group (16.21±3.56).

For the low spatial ability learners, when the learning materials were presented in the general group and in the experimental group, respectively, there was a significant difference in the learning effects, F(1, 156) = 35.43, p<0.01, np² = 0.290. The experimental group’s language learning effect (21.15±3.61) was significantly better than the learning effect of the general group (16.36±4.08).

For high spatial ability learners, the difference in the learning effect was not significant F(1, 156) = 0.05, p = 0.867 when the teaching style was presented in the ordinary group and in the experimental group respectively. The language learning effect presented in the experimental group (22.34 ± 4.08) was not significantly higher than the learning effect in the ordinary group (22.12 ± 3.87).

Figure 5.

Interactive diagram of teaching and spatial ability

The correlation between the use of virtual reality technology, spatial ability, cognitive load and learning effect was analyzed, and Spearman’s method was chosen to obtain the correlationcoefficients between the variables, as shown in Table 4, which shows that it can be found that: the use of virtual reality technology has a significant positive correlation with the learning effect (r=0.415, p<0.01), a significant positive correlation with the cognitive load (r=0.153, p<0.01), a significant positive correlation between cognitive load and learning outcomes (r=0.467, p<0.01), and spatial ability is significantly correlated with the use of virtual reality technology, cognitive load and learning outcomes.

The independent variables in this study are categorical variables, so it is necessary to virtualize the independent variables, with group as the independent variable X (assigned the value of experimental group = 1, general group = 0), language learning effect as the dependent variable Y, and cognitive load as the mediator variable M.

The results of the test are shown in Table 5, and the data indicate that the use of virtual reality technology has a significant predictive effect on language learning outcomes (B=0.77, t=4.79, p<0.001), and when the mediating variable cognitive load is put in, the direct predictive effect of the use of virtual reality technology on the language learning outcomes of students remains significant (B=0.59, t=3.8, p<0.001). The negative predictive effect of the use of virtual reality technology on students’ cognitive load of learning was significant (B=-0.63, t=-3.2, p<0.01), as was the negative predictive effect of students’ cognitive load on their language learning outcomes (B=-0.37, t=-3.69, p<0.01). In addition, the upper and lower 95% confidence intervals of the direct effect of the use of virtual reality technology on students’ language learning outcomes were (0.28, 0.99), and the upper and lower 95% confidence intervals of the mediating effect of students’ cognitive load between the use of virtual reality technology and language learning outcomes were (0.03, 0.40), both of which did not contain 0, indicating that the use of virtual reality technology not only directly predicts language learning outcomes, but also can directly predict language learning effects, but also can predict their learning effects through the mediation of students’ cognitive load, and this direct and mediated effects account for 79% and 21% of the total effect, respectively. This result reflects that cognitive load has a partially mediating role in the effect of the use of virtual reality technology on language learning outcomes, and Hypothesis 2 is valid.

The mediator model with moderation was tested controlling for the gender variable, with a sample size of 122, with group as the independent variable X (assigned as experimental group = 1, general group = 0), learning effect as the dependent variable Y, cognitive load as the mediator variable M, and spatial ability level (spatial ability test score) as the moderating variable W at the 95% confidence interval.

The test results are shown in Table 6, the data show that after putting the spatial ability level into the model, the interaction term between the use of virtual reality technology and the spatial ability level does not have a significant predictive effect on students’ cognitive load (B=0.32, t=1.48, p>0.05), which indicates that the students’ spatial ability level does not modulate the predictive effect of the use of virtual reality technology on the cognitive load, and combined with the two-factor ANOVA results, hypothesis 3 is not valid.

In addition, the results of the moderating effect of spatial ability on the use of realistic technology on learning outcomes through cognitive load are shown in Table 7, when spatial ability is at a low (M-1SD) or average level (M), the upper and lower bounds of the 95% confidence intervals are (2.22, 14.43) and (1.08, 11.24), respectively, and neither of them contains 0. Cognitive load has a mediating role in the effect of the use of virtual reality technology on the learning outcomes has a mediating role. That is to say, for students with lower spatial ability levels, they show lower cognitive load when using virtual reality technology, which in turn shows better learning effects; while when the spatial ability level is higher (M+1SD), the upper and lower limits of the 95% confidence interval are (-2.55, 8.45), which contain 0, suggesting that the cognitive load does not have a mediating role. In other words, the use of virtual reality technology does not affect the cognitive load of students with higher levels of spatial ability in language learning. At all three levels of spatial ability, the mediating effect of cognitive load in the relationship between the use of virtual reality technology and language learning effects also tends to decrease, i.e., as the level of spatial ability of the learners increases, the virtual reality technology is less likely to enhance the language learning effects of the learners by decreasing their cognitive load.