Immersive virtual reality (VR) is increasingly used as a tool for vocational training, especially safety critical vocations such as mine rescuers’ safety and aviation safety where real-world training is often too complicated, costly, or risky (Pedram et al, 2020, Buttussi & Chittaro, 2017). VR has triggered innovative changes in education by allowing learners to acquire skills through repetition and practice in virtual learning systems. VR can provide real-time visualisation and interaction within a virtual world that closely resembles a real world (Chuan et al, 2018). This allows students to practise procedural skills and partake in decision making in an authentic and safe environment.

Very often, these VR learning environments require the use of head-mounted devices which are costly, fragile and not suitable for use over extended training sessions. Henceforth, the development of low-cost, high-resolution personal computers has made it feasible to offer 3D VR learning in school settings through desktop VR learning systems (Huang et al., 2016).

The delivery of VR through head-mounted devices are used in most VR systems to create a sense of immersion for users. On the other hand, desktop VR simulates a real environment or 3D representation of an abstract concept on the screen, which allows learners to interact with the virtual environment using a keyboard mouse, or other navigation/control devices (Merchant et al., 2012). Head-mounted VR offers a more immersive learning experience but the provision of head-mounted devices for every student is a challenge, especially for big classrooms and home-based learning. VR made accessible on any PC or laptop gives students the opportunity to practice and gain mastery of air traffic control skills at their own pace. This could be useful in vocations such as air traffic control where costly equipment are needed to train learners in procedural knowledge and skills.

To date, there are limited studies on the effects of different types of displays on procedural knowledge and skills acquisition in air traffic control. In a study conducted on the effects of different types of virtual reality display on presence and learning in an aircraft safety training scenario (Buttussi & Chittaro, 2017), the types of VR display (desktop VR vs head-mounted VR with different field of view and degree of freedom) affected users’ sense of presence but did not significantly affect self-efficacy. According to Makransky and Lilleholt (2018), immersive VR provided significantly higher presence than desktop VR, a strong positive predictor of both motivation and enjoyment, which in turn positively predicted perceived learning effectiveness.

Hence, this study aims to compare differences in perceived learning effectiveness between VR Head-mounted Display (VR-HMD) and VR Desktop Display (VR-DD) for procedural training of air traffic control. In addition, the study examines the extent to which the variables, i.e. perceived ease of use, perceived usefulness and sense of presence will influence students’ perceived learning when using different VR displays.

The following research questions were formulated to guide the data analysis in this study: 

1. To what extent does each predictor variable (perceived ease of use, perceived usefulness, sense of presence) correlate with perceived learning?

2. Are there significant differences on perceived ease of use, perceived usefulness, sense of presence and perceived learning between VR-HMD and VR-DD?

3. To what extent do perceived ease of use, perceived usefulness and sense of presence significantly predict perceived learning between VR-HMD and VR-DD?

This study employs a cross-sectional quantitative research design using an online questionnaire. Data were collected from 76 final-year students undertaking Airside Operations and Air Traffic Management module in the polytechnic. The intact group was split into two groups of students using different VR systems to acquire their knowledge and skills on aircraft take-off and landing procedure, i.e. VR-HMD (N=33) and VR-DD (N=43). A self-report questionnaire was administered to the participants after 3 weeks of learning with the VR systems. Self-report questionnaire was intended to measure students’ perceived ease of use, perceived usefulness, sense of presence and perceived learning. The scales on perceived ease of use and perceived usefulness were adapted from the Technology Acceptance Model (Davis et al., 1989) while sense of presence and perceived learning were adapted from Makransky & Petersen (2019) and Lee et al (2010) respectively. A total of 16 items with a 5-point Likert scale ranging from strongly disagree (1) to strongly agree (5) were included in the questionnaire. Demographic data such as gender, age and prior experience with VR systems were collected for the purpose of reporting the profiles of the participants. Statistical analyses were performed using IBM SPSS version 24.0. Descriptive statistics, correlations, Independent-samples T-test, and regression analyses were employed in the data analysis. In addition, the regression model was tested using hierarchical linear modelling.

Preliminary analysis showed that all the study variables had mean ratings, ranged between 4.05 and 4.25, above 4.00 on a 5-point Likert scale. This indicates favorable responses from the participants pertaining to evaluation of the study variables. The Cronbach’s alpha coefficient which is a measure of internal consistency reliability for the 4 variables ranged between .91 and .96, all above the threshold value of .70 (Nunnally & Bernstein, 1994). No multicollinearity issues were found in the regression analyses as the variance inflation factor (VIF) values, ranged between 1.77 and 2.69, were significantly lower than the recommended value of 10.0 (Hair et al., 2009). For RQ1, all the study variables are positively and significantly correlated with perceived learning (.76 ≤ r ≤ .80, p< 0.01). In terms of RQ2, perceived usefulness (M= 4.50, SD = .50) in VR-HMD is significantly higher than VR-DD (M = 4.06, SD = 1.02) with p<.05 at medium effect size. Regarding RQ3, perceived ease of use (β = .38, p < .01) and perceived usefulness (β = .42, p < .01) were the significant predictors of perceived learning in VR-DD. However, for VR-HMD, perceived ease of use and sense of presence were significant predictors of perceived learning with sense of presence being the higher significant predictor (β = .56, p < .001), followed by perceived ease of use (β = .29, p < .1). Sense of presence contributed to 15% of variance in perceived learning for VR-HMD but is not a significant determinant of perceived learning for VR-DD. In conclusion, this study provided insights on the variables which impact perceived learning between VR-HMD and VR-DD. The results of the study will provide insights for educators to help them better understand the factors influencing students’ perceived learning, which could aid in the instructional design of VR learning contents as well as choice of technology use. The implications of the findings, limitations of study and future research will be discussed in the presentation. *Special thanks to Mr Mathivaanan S/O Vedaraj from the School of Engineering for his contribution in developing VR content for both VR-HMD and VR-DD and collecting data for this study.

virtual reality, head-mounted display, desktop display, perceived learning