A systematic review of foreign language learning with immersive technologies (2001-2020)

Rebecca M. Hein; Carolin Wienrich; Marc E. Latoschik; Rebecca M. Hein; Carolin Wienrich; Marc E. Latoschik

doi:10.3934/electreng.2021007

AIMS Electronics and Electrical Engineering

2021, Volume 5, Issue 2: 117-145. doi: 10.3934/electreng.2021007

Previous Article Next Article

Review Topical Sections

A systematic review of foreign language learning with immersive technologies (2001-2020)

1.
Human-Computer Interaction, Julius-Maximilians-Universität Würzburg, Am Hubland, D-97074 Würzburg
2.
Human-Technique Systems, Julius-Maximilians-Universität Würzburg, Oswald-Külpe-Weg 82, D-97074 Würzburg

Received: 30 January 2021 Accepted: 12 April 2021 Published: 16 April 2021

This study provides a systematic literature review of research (2001–2020) in the field of teaching and learning a foreign language and intercultural learning using immersive technologies. Based on 2507 sources, 54 articles were selected according to a predefined selection criteria. The review is aimed at providing information about which immersive interventions are being used for foreign language learning and teaching and where potential research gaps exist. The papers were analyzed and coded according to the following categories: (1) investigation form and education level, (2) degree of immersion, and technology used, (3) predictors, and (4) criterions. The review identified key research findings relating the use of immersive technologies for learning and teaching a foreign language and intercultural learning at cognitive, affective, and conative levels. The findings revealed research gaps in the area of teachers as a target group, and virtual reality (VR) as a fully immersive intervention form. Furthermore, the studies reviewed rarely examined behavior, and implicit measurements related to inter- and trans-cultural learning and teaching. Inter- and transcultural learning and teaching especially is an underrepresented investigation subject. Finally, concrete suggestions for future research are given. The systematic review contributes to the challenge of interdisciplinary cooperation between pedagogy, foreign language didactics, and Human-Computer Interaction to achieve innovative teaching-learning formats and a successful digital transformation.

Keywords:

Citation: Rebecca M. Hein, Carolin Wienrich, Marc E. Latoschik. A systematic review of foreign language learning with immersive technologies (2001-2020)[J]. AIMS Electronics and Electrical Engineering, 2021, 5(2): 117-145. doi: 10.3934/electreng.2021007

Related Papers:

[1]	Zongying Feng, Guoqiang Tan . Dynamic event-triggered $H_{\infty}$ control for neural networks with sensor saturations and stochastic deception attacks. Electronic Research Archive, 2025, 33(3): 1267-1284. doi: 10.3934/era.2025056
[2]	Yawei Liu, Guangyin Cui, Chen Gao . Event-triggered synchronization control for neural networks against DoS attacks. Electronic Research Archive, 2025, 33(1): 121-141. doi: 10.3934/era.2025007
[3]	Xingyue Liu, Kaibo Shi, Yiqian Tang, Lin Tang, Youhua Wei, Yingjun Han . A novel adaptive event-triggered reliable $H_\infty$ control approach for networked control systems with actuator faults. Electronic Research Archive, 2023, 31(4): 1840-1862. doi: 10.3934/era.2023095
[4]	Chao Ma, Hang Gao, Wei Wu . Adaptive learning nonsynchronous control of nonlinear hidden Markov jump systems with limited mode information. Electronic Research Archive, 2023, 31(11): 6746-6762. doi: 10.3934/era.2023340
[5]	Hangyu Hu, Fan Wu, Xiaowei Xie, Qiang Wei, Xuemeng Zhai, Guangmin Hu . Critical node identification in network cascading failure based on load percolation. Electronic Research Archive, 2023, 31(3): 1524-1542. doi: 10.3934/era.2023077
[6]	Chengbo Yi, Jiayi Cai, Rui Guo . Synchronization of a class of nonlinear multiple neural networks with delays via a dynamic event-triggered impulsive control strategy. Electronic Research Archive, 2024, 32(7): 4581-4603. doi: 10.3934/era.2024208
[7]	Liping Fan, Pengju Yang . Load forecasting of microgrid based on an adaptive cuckoo search optimization improved neural network. Electronic Research Archive, 2024, 32(11): 6364-6378. doi: 10.3934/era.2024296
[8]	Ramalingam Sakthivel, Palanisamy Selvaraj, Oh-Min Kwon, Seong-Gon Choi, Rathinasamy Sakthivel . Robust memory control design for semi-Markovian jump systems with cyber attacks. Electronic Research Archive, 2023, 31(12): 7496-7510. doi: 10.3934/era.2023378
[9]	Nacera Mazouz, Ahmed Bengermikh, Abdelhamid Midoun . Dynamic design and optimization of a power system DC/DC converter using peak current mode control. Electronic Research Archive, 2025, 33(4): 1968-1997. doi: 10.3934/era.2025088
[10]	Yejin Yang, Miao Ye, Qiuxiang Jiang, Peng Wen . A novel node selection method for wireless distributed edge storage based on SDN and a maldistributed decision model. Electronic Research Archive, 2024, 32(2): 1160-1190. doi: 10.3934/era.2024056

Abstract

1. Introduction

The COVID-19 pandemic poses a significant challenge to global public health. In the past two years, SARS-CoV-2 has spread to five continents resulting in more than six million deaths (by 14 September 2022) [1]. Cambridge University estimated that the pandemic would cost around 82 trillion dollars to the global economy over five years [2]. Effective and reliable prevention and treatment methods are urgently demanded to defeat the virus effectively. As of 12 September 2022, more than 12 billion vaccine doses have been administered worldwide [1]. By the end of 2021, eight anti-SARS-CoV-2 drugs based on therapeutic neutralizing antibodies (nAbs) have been approved under Emergency Use Authorization (EUA) by the US Food and Drug Administration (FDA) and/or European Agency of Medicines (EMA) [3]. However, the virus is evolving continuously, giving rise to numerous variants [4]. Some variants (such as Alpha, Beta, Gamma, Delta, and Omicron) with enhanced virulence are defined as variants of concern (VOCs) by the World Health Organization (WHO) [5]. Recently, Omicron has evolved into multiple sub-variants: BA.2, BA.3, BA.4, and BA.5 [5]. The viral variants have, in turn, raised concerns about the effectiveness of the currently available vaccines for coronavirus [6]–[8]. Many studies have reported the high immune evasion ability (IEA) of the Omicron variant and its sub-variants [9]–[15]. Therefore, it is crucial to systematically analyze the effect of mutations in the spike protein on antibody binding in order to guide the design of broad-spectrum antibodies.

The receptor-binding domain (RBD) of the SARS-CoV-2 spike glycoprotein is the target of most current clinical antibodies [3],[16]. The spike protein is a homotrimer located on the surface of the virus membrane that mediates entry of the virus into the host cells via interaction between the receptor binding motif (RBM) of RBD and the human angiotensin converting enzyme 2 (ACE2) receptor [17]. Typically, RBD is buried on the surface of the spike protein. In the presence of ACE2, the conformation of the spike protein shifts to an opening state [18]. This conformational shift brings RBD to the “up” conformational state, and the RBM is exposed on its top surface for ACE2 binding [17]. Most antibodies target the epitopes that overlap with the RBM and interfere with viral infection by blocking the binding of ACE2 to the spike protein [16]. In addition, some antibodies also bind to the side surface of RBD [3],[16]. Mutations in the viral variants weaken the interactions between antibodies and their target epitopes on the variants, thus allowing the immune evasion of the virus. The Omicron variant exhibits low binding affinity to most of the currently available SARS-CoV-2 neutralizing antibodies [9],[12].

Computational biology provides powerful tools to study the binding affinity of the spike protein to various kinds of antibodies based on an energetic perspective. Our previous studies explored the structure/energy basis of spike-ACE2 interactions [19] and the effects of mutations on receptor binding in some variants [20]. We also predicted important mutations, such as N501Y, Q493R, and Q498R, appearing in the subsequent VOCs [19],[20]. These works confirmed the validity of our approach.

In this study, we first constructed the structural models of the spike-antibody complexes based on the experimental structures. Then, a systematic binding free energy (ΔG_binding) and binding free energy change (ΔΔG_binding) were assessed to evaluate the differences in the affinity of eight nAbs (AZD1061, AZD8895, CT-P59, LY-COV555, LY-COV016, REGN10933, REGN10987, and S309) to the spike proteins of eight viral variants (Alpha, Beta, Gamma, Delta, Omicron BA.1, Omicron BA.2, Omicron BA.3, and Omicron BA.4; the spike proteins of BA.4 and BA.5 harbored the same mutations). Through the analysis of ΔΔG_binding values, we designed a scoring method to assess the IEA of viral variants. Next, we analyzed the differences in ΔG_binding and ΔΔG_binding during the binding of the ACE2 receptor with two patient-derived antibodies (P22A and 510A5) and an artificially designed antibody mimic (AHB2). We also discussed the possible directions for future antibody design. Our results could help provide valuable insights into developing more effective therapeutic methods for the eradication of SARS-CoV-2 and its variants.

2. Materials and methods

2.1. Modeling the complex structures

The wild-type spike/ACE2 complex structure was determined by high-resolution cryogenic electron microscopy (cryo-EM) (PDB ID: 7DF4) [18]. Two mutations of the Omicron variant (N679K and P681H) locate in the fragment that is omitted from the experimental structure. Then the missing fragment was repaired using Modeller [21]. In addition, structural data showed that the RBD of the spike protein of the Omicron variant does not exhibit any significant conformational difference from that of the wild-type [22]. Therefore, we used the wild-type structure as the first-order approximation to predict the effects of mutations in the spike protein. The mutations of the eight variants were introduced into wild-type spike protein using PyMOL. Information about the variants of SARS-CoV-2 used in this work is shown in Table 2. After removing the ACE2 receptor from the complex structure, we obtained the isolated spike protein structure. The structures of antibodies and their interactive models were extracted from known structures, which are as follows:

Table 1. Information about antibodies.

	Name	PDB ID	Released Date	Epitope regions	Source
Neutralizing antibodies	AZD1061	7L7E	2021-09-01	Top	Neutralized wild-type SARS-CoV-2 virus, the wild type [23],[24]
	AZD8895	7L7D	2021-09-01	Top	Neutralized wild-type SARS-CoV-2 virus, the wild type [23],[24]
	CT-P59	7CM4	2021-01-20	Top	The peripheral blood mononuclear cells of a SARS-CoV-2 convalescent patient, the wild type [25],[26]
	LY-CoV016	7C01	2020-05-27	Top	a convalescent COVID-19 patient from China, the wild type [27]
	LY-CoV555	7KMG	2021-01-27	Top	Isolated from a convalescent COVID-19 patient from North America, unknown type [28]
	REGN10933	6XDG	2020-06-24	Top	Regeneron's VelocImmune® human antibody mouse platform, the wild type [29]
	REGN10987	6XDG	2020-06-24	Top	B cells of human donor previously infected with SARS-CoV-2, the wild type [29]
	S309	7TN0	2022-02-02	Side	Isolated from a subject who recovered from a SARS-CoV infection in 2013 [30],[31]
Patient-derived antibodies	P22A	7CHS	2021-05-19	Top	Isolated from SARS-CoV-2 infected patients, the wild type [32]–[34]
	510A5	7WS7	2022-06-01	Side	COVID-19 convalescent blood samples collected within a 2-month window post discharge, unknown type [35],[36]
Artificially designed antibody mimic	AHB2	7UHB	2022-06-08	Top

| Show Table

DownLoad: CSV

Table 2. Information about SARS-CoV-2 variants.

WHO Label	PANGO Lineage	Date first detected	GenBank Accession NO.
Alpha	B.1.1.7	2020-09	OV054768.1
Beta	B.1.351	2020-10	OX003129.1
Gamma	P.1	2020-12	OX000832.1
Delta	B.1.617.2	2020-12	OK091006.1
Omicron	BA.1	2021-11	OX315743.1
Omicron	BA.2	2021-11	OX315675.1
Omicron	BA.4	2022-02	OP093374.1

| Show Table

DownLoad: CSV

2.2. Calculating the binding free energy change

The all-atom structures were converted into CG representation and subjected to extensive relaxation (11000 steps, 0.0001 ps step-size) under the temperature of 50 K. During relaxation, one structure was obtained every 1000 steps. Finally, we got a conformational trajectory comprising 11 structures. All these structures were used for energy evaluation. Our CG model, which was adapted from previous studies [37]–[39], focused on the precise treatment of the electrostatic charges and was sensitive to the charge distribution of the protein ionized groups. Hence, before energy evaluation, a Monte Carlo proton transfer (MCPT) method [38] was used to determine the charge states of the residues in each structure. During MCPT, protons were “jumped” between ionizable residues, and a standard Metropolis criterion was utilized to calculate the acceptance probability. The total CG folding free energy ( $Δ G_{f o l d}$ ) was calculated according to the following formula:

$Δ G_{f o l d} = Δ G_{m a i n} + Δ G_{s i d e} + Δ G_{m a i n - s i d e} = c_{1} Δ G_{s i d e}^{v d w} + c_{2} Δ G_{s o l v}^{C G} + c_{3} Δ G_{H B}^{C G} + Δ G_{s i d e}^{e l e c} + Δ G_{s i d e}^{p o l a r} + Δ G_{s i d e}^{h y d} + Δ G_{m a i n - s i d e}^{e l e c} + Δ G_{m a i n - s i d e}^{v d w}$ (1)

In this formula, the CG folding free energy ( $Δ G_{f o l d}$ ) consists of 3 parts: the main chain free energy ( $Δ G_{m a i n}$ ), the side chain free energy ( $Δ G_{s i d e}$ ), and the free energy of main-side interactions ( $Δ G_{m a i n - s i d e}$ ). These three parts can also be divided into 8 terms: side chain van der Waals energy ( $Δ G_{s i d e}^{v d w}$ ), main chain solvation energy ( $Δ G_{s o l v}^{C G}$ ), main chain hydrogen bond energy ( $Δ G_{H B}^{C G}$ ), side chain electrostatic energy ( $Δ G_{s i d e}^{e l e c}$ ), side chain polar energy ( $Δ G_{s i d e}^{p o l a r}$ ), side chain hydrophobic energy ( $Δ G_{s o d e}^{h y d}$ ), main chain/side chain electrostatic energy ( $Δ G_{m a i n - s i d e}^{e l e c}$ ), and main chain/side chain van der Waals energy ( $Δ_{m a i n - s i d e}^{v d w}$ ). The scaling coefficients c₁, c₂, and c₃ were set as 0.10, 0.25, and 0.15, respectively. The energy reached equilibrium after 10000 steps due to the large size of the protein. Therefore, the average energy of the last two structures was used as the final energy. All relative calculations were performed using the Molaris-XG package [40],[41].

The binding energies for spike-antibody complexes ( $Δ G_{b i n d i n g}$ ) were calculated using the following formula:

$Δ G_{b i n d i n g} = G_{s p i k e - a n t i b o d y} - G_{s p i k e} - G_{a n t i b o d y}$ (2)

In this formula, the terms on the right represent the CG folding free energy of the spike-antibody complex ( $G_{s p i k e - a n t i b o d y}$ ), the CG folding free energy of the isolated spike ( $G_{s p i k e}$ ), and the CG folding free energy of isolated antibody ( $G_{a n t i b o d y}$ ), respectively.

To calculate the binding free energy change ( $Δ Δ G_{b i n d i n g}$ ) of the variant spike-antibody complex, the following formula was used:

$Δ Δ G_{b i n d i n g} = Δ G_{b i n d i n g_{v a r i a n t}} - Δ G_{b i n d i n g_{w i l d t y p e}}$ (3)

In this formula, the terms on the right represent the binding free energy of the variant spike to an antibody ( $Δ G_{b i n d i n g_{v a r i a n t}}$ ) and the binding free energy of the wild type spike to the same antibody ( $Δ G_{b i n d i n g_{w i l d t y p e}}$ ), respectively. The lower the $Δ Δ G_{b i n d i n g}$ , the higher the stability.

2.3. Calculating the normalized score

The eight variants were considered as the set V, and the eight antibodies were considered as the set A. For each pair of a variant and an antibody, we calculated ΔΔG_binding according to the method described in Section 4.2. All ΔΔG_binding values were considered as the set $G = {Δ Δ G_{i j} | i \in V, j \in A}$ . Each ΔΔG_binding value was normalized using the following formula:

$N o r m a l i z e (Δ Δ G_{b i n d i n g}) = \frac{Δ Δ G_{b i n d i n g} - M i n (G)}{M a x (G) - M i n (G)}$ (4)

In this formula, the terms on the right represent the value would be normalized (ΔΔG_binding), the minimum value of all ΔΔG_binding values (Min(G)), and the maximum value of all ΔΔG_binding values (Max(G)).

The IEA score of each variant (S_i) was calculated using the following formula:

$S_{i} = \frac{\sum_{j}^{j \in A} N o r m a l i z e (Δ Δ G_{b i n d i n g})}{8}, i \in V$ (5)

In this formula, the terms on the right represent each variant belong to the set V (i) and each antibody belong to the set A (j). The lower the S_i, the weaker the immune escape ability.

2.4. Analyzing the interaction surface

The interfaces between antibodies and the spike protein were analyzed using PDBePISA [42].

3. Results

3.1. Assessing the immune evasion ability of variants

The eight nAbs selected for this study recognize different antigenic sites in the RBD of the spike protein of SARS-CoV-2 (Figure 1A). These antibodies can be divided into two groups based on whether or not they can block the spike protein's binding to ACE2 [43]. The first groups comprised seven nAbs (AZD1061, AZD8895, CT-P59, LY-COV555, LY-COV016, REGN10933, and REGN10987) binding to the epitopes that overlap with the RBM of the top site of RBD (Figure 1A), which is also the primary binding site of the ACE2 receptor. The virus cannot be recognized and activated in the presence of these antibodies, thus losing the ability to infect human cells [16]. The second group consisted of the S309 antibody that binds to the side sites of the RBD (Figure 1A). S309 targets non-RBM sites and hence does not block the binding of the spike protein to the ACE2 receptor. Therefore, the S309 binding is not dependent on the conformation of the spike protein and is hypothesized to access all three epitopes of the spike trimer and sterically shield the engagement of ACE2 to block the viral infection [16].

Figure 1. (A) The eight nAbs bind to the spike protein with different poses; (B) Mutations in the SARS-CoV-2 spike of each variant, and the mutated sites are highlighted by colors; (C) The binding free energy change (ΔΔG_binding) of each combination of variant and antibody; (D) The normalized score of the IEA of variants, the error bar represents the variance in the values of a variant.

DownLoad: Full-Size Img PowerPoint

The mutations in SARS-CoV-2 variants result in changes in the binding affinities of their spike proteins to the eight nAbs. The eight chosen variants' spike proteins have mutations that were primarily clustered in the RBM region (Figure 1B), which might enhance the binding affinity of their spike proteins to the ACE2 receptor [20],[44]. Meanwhile, such mutations have been shown to weaken the spike-antibody interaction and interfere with antibody recognition, thus conferring IEA to viral variants [9]–[11],[43]–[47]. Therefore, the ΔΔG_binding induced by the mutations on the spike protein can be used to determine the IEA of variants.

We calculated the binding free energies of the wild-type (ΔG_binding-wt) spike and each variant spike protein (ΔG_{binding-variant}) with each therapeutic nAb. Then, we calculated the binding free energy change (ΔΔG_binding = ΔG_{binding-variant} - ΔG_binding-wt, see Methods). In total, the more recent the variants of the spike protein are, the lower binding ability they have to nAbs (Figure 1C). In at least one variant, a reduction in the binding affinity of the nAbs belonging to the first group to the RBD was observed (Figure 1C). For instance, we noticed a mild decrease in the binding affinity of RGEN10933 to the alpha variant, but increases in the other seven variants (Figure 1C). However, the binding affinities of the spike proteins of all eight variants to S309 were enhanced compared to the wild-type spike. (Figure 1C, Table S1). This finding might be attributed to the fact that the binding epitopes of S309 are evolutionarily conserved across several sarbecoviruses, including SARS-CoV [43],[47]. Our findings that most nAbs show weaker binding affinity to the Omicron variants except for S309 were consistent with recent studies [10],[11]. To intuitively measure the IEA of viral variants, we designed a scoring system based on the ΔΔG_binding calculations. The ΔΔG_binding value of each variant-antibody combination was normalized into the range [10],[11] by the min-max feature scaling method. The IEA score assigned to a variant was the average of its eight normalized values (corresponding to eight antibodies). Our findings showed that the recently evolved sub-lineages of Omicron, BA.3 and BA.4, possess the most robust ability to evade antibody-induced neutralization (Figure 1D, Table S2). According to the variance of all ΔΔG_binding values of a variant (the error bar in Figure 1D), we find the IEA score is sensitive to a single value. For instance, the variance of BA.3 is lower than that of BA.4. Combined with the ΔΔG_binding calculations, we find that the BA.3 weakens the affinity with most of the nAbs, while the BA.4 exhibits the most prominent reduction in binding affinity to LY-COV555 in all antibodies, which might be contributed to that fact that BA.4 has an additional L452R mutation located on the interface between LY-COV555 and spike (Figure 1C, Figure S1). The substitution of Leucine (uncharged and small side chain) with Arginine (positive charged and large side chain) may disrupt the electrostatic interaction or result in steric hindrance to affect the binding between LY-COV555 and the spike (Figure S1). The IEA scores of these variants indicated that the IEA of SARS-CoV-2 is reinforced in the newly emerging variants (Figure 1D) , which is consistent with the trend researchers observed [48]. Thus, the IEA scoring system can also be used to assess the immune escape capability of new viral variants that might evolve in the future.

3.2. Analyzing the binding affinity of the spike proteins to P22A and 510A5

To further investigate the differences between the top and side epitopes of the spike RBD, we analyzed the binding energies of the spike proteins complexed with two patient-derived antibodies, P22A and 510A5. The P22A antibody was isolated from a patient infected with the early Wuhan strain (wild-type) [49],[50]. P22A can potently neutralize the wild-type spike [49],[50], but so far, no data has shown its ability to bind to the spike protein of variants. P22A binds to the top side of RBD (Figure 2A) and causes massive spatial clashes with ACE2 (approximately 2000 Å³) [49],[50]. The 510A5 antibody was isolated from a patient infected with an unknown type of SARS-CoV-2 (not reported in the original paper), but exhibiting binding affinity to the wild-type, Delta, and Omicron variants [51]. It can bind to both the top and side sites of the RBD of the wild-type spike protein but can only bind to the side sites in the Omicron variant (Figure 2A), resulting in clashes with ACE2 [51].

Figure 2. (A) The two patient-derived antibodies (P22A and 510A5) bind to different sites on the spike protein; (B) The binding free energy change (ΔΔG_binding) for each combination of variant and antibody; (C, D) The residues contribute to the formation of hydrogen bonds when antibodies bind to spike; spike residues are shown in green, antibody residues are shown in yellow, and the hydrogen bonds are shown in cyan.

DownLoad: Full-Size Img PowerPoint

We calculated the ΔG_binding and the ΔΔG_binding values for WT and all variants. Despite the P22A antibody being produced by induction of wild-type virus particles, it exhibits lower ΔG_binding than the 510A5 for all eight variants (Figure S2). This finding might be ascribable to the difference between the top and side binding sites of RBD. Therefore, we performed the interface analysis of the antibody-spike complex structures with respect to these two antibodies. Our results showed that P22A has a larger interface area (1102.1 Å² vs. 664.7 Å²), interacts with more spike residues (41 vs. 18), and forms more hydrogen bonds (22 vs. 7) than 510A5 (Figure 2C, D, Table S3). The ΔΔG_binding values showed that the binding affinity of 510A5 to the spike protein is enhanced for all variants. This result was comparable to that observed for S309 (Figure 1C). Many studies have reported the high immune evasion ability of the Omicron variant and its sub-variants. Therefore, we expected that the omicron BA.1, BA.2, BA.3, and BA.4 should have poor binding affinity to P22A compared to the wild type. However, our simulation results show that the omicron BA.1, BA.2, and BA.3 exhibit increased binding affinity to P22A, except for the omicron BA.4 (1.79 kcal/mol, Figure 2B) which presents a prominent decrease in binding affinity. These results may indicate that, for antibody P22A, not all the Omicron sub-variants exhibit high immune escape ability to antibody P22A, but the BA.4 is the most transmissible virus variant. The interface analysis showed that, compared to the wild-type, the spike-510A5 interface area increases while the spike-P22A interface area decreases for the Omicron BA.4 variant (Table S3). In the case of 510A5, the increase in interface area can be credited to the N440K mutation located on the side site (area increases from 137.9 to 148.6 Å²) and the residue N439 that is near N440K (area increases from 8.9 to 13.8 Å²). These findings indicated that the epitopes on the side sites of RBD are more resistant to the enhanced IEA of viral variants.

3.3. Analyzing the binding affinity of the spike protein to AHB2 and ACE2

Except for nAbs, engineered ACE2 traps, such as protein mimics, are also promising therapeutic tools to counter SARS-CoV-2 [52],[53]. Baker et al. designed a lot of minibinder proteins as ACE2 traps to block the interaction between the spike protein and ACE2 [54],[55]. To assess the effectiveness of minibinders in counteracting immune escape of viral variants, it is best to compare the binding affinities of them with that of ACE2. We performed structural and energy analysis of AHB2, which is representative of these minibinder proteins. The binding mode and the helical structure of AHB2 that interacts with the RBD domain of the spike protein are similar to those of ACE2 (Figure 3A, B). However, despite being an ACE2 mimic, the main RBD-interacting helix of AHB2 has low sequence similarity with ACE2 (Figure 3B). We then calculated the ΔG_binding of the AHB2 or ACE2-spike protein complex. Our results showed that the binding affinities of wild-type and all variants to ACE2 were much stronger than those to AHB2 (Figure 3C). The finding might be attributed to the fact that ACE2 is a natural receptor that is more prone to bind to spike. The area of the ACE2-spike interface was larger than that of the AHB2-spike interface (Figure 3F). We also found a substantial enhancement in the binding affinity of ACE2 to the spike protein of more recently evolved variants (Figure 3C, D), which might explain the high transmissibility of the Delta and Omicron variants [20],[22],[56]–[58]. However, the variants exhibited a higher binding affinity to AHB2 (Figure 3C, D), implying that AHB2 could resist the neutralization potency reduction induced by mutations in the spike protein. This finding was consistent with previous experimental results showing that AHB2 can effectively neutralize the Delta and Omicron BA.1 variants [55]. In addition, our results also showed that AHB2 exhibited improved affinity to other sub-lineages of the Omicron variants, that is, BA.2, BA.3, and BA.4 (Figure 3D). We identified the residues of the spike protein of Omicron BA.4 that participates in the interaction with AHB2 and ACE2. Most residues that involved in the complex formation were identical, proving that AHB2 binds to the same sites as ACE2 (Figure 3E). Although the area of the ACE2-spike interface is larger, five additional spike residues interact with AHB2 (Figure 3E, Table S4). We speculated that AHB2 resisted the neutralization potency reduction of variants by interacting with as many spike residues as possible. In summary, these results highlighted the potential of ACE2 mimics as antiviral therapeutic tools to mitigate the rapidly evolving pandemic.

Figure 3. (A) The structures and interaction modes of AHB2 and ACE2; (B) The structure and sequence alignments of the main RBD-interacting helix of the two proteins; (C) The binding free energy (ΔG_binding) of each combination of variants and the two proteins; (D) The binding free energy change (ΔΔG_binding) of each combination of variants and the two proteins; (E, F) The key residues and interaction surface of the spike protein contributed to the binding of AHB2 and ACE2.

DownLoad: Full-Size Img PowerPoint

4. Discussion and conclusions

Using structural modeling and computational biology methods, we systematically analyzed the binding affinity of the spike proteins of eight SARS-CoV-2 variants to three groups of antibodies: monoclonal nAbs for therapeutic, antibodies isolated from patients who were infected with the virus, and artificially designed antibody mimics. The immune evasion of newly emerging viral variants is a significant issue at present, with concrete manifestation in the reduced binding affinity of mutant spike proteins to antibodies. We designed a scoring system based on the ΔΔG_binding values to evaluate the IEA of variants. This method can also be applied to evaluate an antibody's broad-spectrum effectiveness, thus playing an important role in designing antibodies targeting the spike protein.

The antibodies selected for this study bound to either the top or the side sites of the RBD of spike protein, neutralizing the coronavirus in multiple ways [16]. As the top sites overlap with the RBM, the antibodies bind to this site directly block the interactions between the spike protein and the ACE2 receptor. Our results showed that the antibodies bound to the top sites exhibited a higher binding affinity than those that bound to the side sites. This might be attributed to the larger interaction interface and the formation of more hydrogen bonds during the interactions with the top sites. However, mutations also tend to aggregate at the top sites. Subsequently, the antibodies that bind to the top sites are more susceptible to variant evasion than those that bind to the side sites. Since the side sites are more evolutionarily conserved, the antibodies that target these sites show comparable binding affinity to the spike proteins across different variants.

Our results reflect a comprehensive conclusion that the Omicron sub-variants have higher immune escape ability than other variants of SARS-CoV-2. Actually, the binding affinity and the interaction mode between the spike protein and antibodies are more complicated than we discussed in this work. As we discussed in section 3.2, to the specific antibody P22A, the early Omicron sub-variants (BA.1 and BA.2) exhibit stronger binding affinities than the alpha, beta, gamma, and delta variants. These antibodies may have some important interaction modes that could help us develop new therapeutic antibodies. In this study, we only calculated the binding energies and interaction modes of two specific antibodies, P22A and 510A5. More representative antibodies should also be included in our studies in the future, which may give rise to a new avenue for designing effective and stable antibodies.

Steric effects triggered by antibodies play an important role in blocking the binding of ACE2 to the spike protein. Previous studies have shown that the 510A5 antibody cannot effectively neutralize the Omicron spike protein, even though it binds to the side sites of its RBD [51]. When binding to the side sites, this antibody cannot generate sufficient steric hindrance to block the interaction between ACE2 and the spike protein [51]. These findings suggest that side-site targeting antibodies might fail to block ACE2-spike interaction, although the viral variant does not readily evade their binding. Considering the advantages and drawbacks of the top and side sites, exploiting the combined advantages conferred by the antibodies that bind to these two types of sites could lead to the development of better therapeutic approaches.

In addition to the top and side sites of RBD, antibodies targeting the fusion peptides of the spike protein have also been reported in recent studies [59],[60]. After binding to the ACE2 receptor, the spike protein is divided into two parts, S1 and S2, by the membrane enzyme transmembrane serine protease 2 (TMPRSS2) or endosomal cathepsins [61],[62]. Then, the S1 subunit is shed, and the fusion peptide is exposed, leading to the insertion of the fusion peptide into the cell membrane, which results in viral fusion [61],[62]. Some motifs of the fusion peptide are highly conserved. Thus, the antibodies targeting these motifs can neutralize several viral variants [59],[60]. Therefore, fusion peptides can also be used as an essential epitope in developing effective antibodies. In future studies, we will analyze the binding affinities of the existing antibodies that target the fusion peptide.

Overall, our study presented a systematic analysis of the binding affinity of various antibodies to the spike proteins of various variants. We compared the effects of two types of RBD binding sites on antibody binding and neutralization evasion of viral variants. Our findings can be used to design more effective broad-spectrum antibodies or mimics against viral agents.

References

[1]	Ahn SJ, Bostick J, Ogle E, et al. (2016) Experiencing nature: Embodying animals in immersive virtual environments increases inclusion of nature in self and involvement with nature. Journal of Computer-Mediated Communication 21: 399–419. doi: 10.1111/jcc4.12173
[2]	Banakou D, Groten R, Slater M (2013) Illusory ownership of a virtual child body causes overestimation of object sizes and implicit attitude changes. Proceedings of the National Academy of Sciences 110: 12846–12851. doi: 10.1073/pnas.1306779110
[3]	Barreira J, Bessa M, Pereira LC, et al. (2012) MOW: Augmented Reality game to learn words in different languages: Case study: Learning English names of animals in elementary school. In 7th Iberian conference on information systems and technologies (CISTI 2012). 1–6. IEEE.
[4]	Barrett MD (2011) Intercultural competence. EWC Statement Series 2: 23–27.
[5]	Bazzaza MW, Alzubaidi M, Zemerly MJ, et al. (2016) Impact of smart immersive mobile learning in language literacy education. In 2016 IEEE Global Engineering Education Conference (EDUCON). 443–447. IEEE.
[6]	Berns A, Mota JM, Ruiz-Rube I, et al. (2018) Exploring the potential of a 360 video application for foreign language learning. In Proceedings of the Sixth International Conference on Technological Ecosystems for Enhancing Multiculturality. 776–780.
[7]	Blell G, Doff S (2014) Mehrsprachigkeit und Mehrkulturalität: Einführung in das Thema. Zeitschrift für Interkulturellen Fremdsprachenunterricht 19.
[8]	Bond M, Marín VI, Dolch C, et al. (2018) Digital transformation in German higher education: student and teacher perceptions and usage of digital media. Int J Educ Technol H 15: 1–20. doi: 10.1186/s41239-017-0083-9
[9]	Buhl H, Winter R (2009) Full virtualization – BISE's contribution to a vision. Business and Information Systems Engineering 1: 133–136. doi: 10.1007/s12599-008-0023-2
[10]	Byram M, Doyé P (1999) Intercultural competence and foreign language learning in the primary school. The teaching of modern foreign languages in the primary school, 138–151.
[11]	Chabot S, Drozdal J, Peveler M, et al. (2020) A collaborative, immersive language learning environment using augmented panoramic imagery. In 2020 6th International Conference of the Immersive Learning Research Network (iLRN), 225–229.
[12]	Chang YJ, Chen CH, Huang WT, et al. (2011) Investigating students' perceived satisfaction, behavioral intention, and effectiveness of English learning using augmented reality. In 2011 IEEE International Conference on Multimedia and Expo. 1–6.
[13]	Chen G, Starosta W (1999) A review of the concept of intercultural awareness. Human Communication 2: 27–54.
[14]	Chen CP, Wang CH (2015) The effects of learning style on mobile augmented-reality-facilitated English vocabulary learning. In 2015 2nd International Conference on Information Science and Security (ICISS). 1–4.
[15]	Chen YL (2016) The effects of virtual reality learning environment on student cognitive and linguistic development. The Asia-Pacific Education Researcher 25: 637–646. doi: 10.1007/s40299-016-0293-2
[16]	Chen SY, Hung CY, Chang YC, et al. (2018) A study on integrating augmented reality technology and game-based learning model to improve motivation and effectiveness of learning English vocabulary. In 2018 1st International Cognitive Cities Conference (IC3). 24–27.
[17]	Chen MP, Wang LC, Zou D, et al. (2020) Effects of captions and English proficiency on learning effectiveness, motivation and attitude in augmented-reality-enhanced theme-based contextualized EFL learning. Computer Assisted Language Learning, 1–31.
[18]	Chen MRA, Hwang GJ (2020) Effects of experiencing authentic contexts on English speaking performances, anxiety and motivation of EFL students with different cognitive styles. Interactive Learning Environments, 1–21.
[19]	Chen Y, Smith TJ, York CS, et al. (2020) Google Earth Virtual Reality and expository writing for young English Learners from a Funds of Knowledge perspective. Computer Assisted Language Learning 33: 1–25. doi: 10.1080/09588221.2018.1544151
[20]	Chen CH (2020) AR videos as scaffolding to foster students' learning achievements and motivation in EFL learning. British Journal of Educational Technology 51: 657–672. doi: 10.1111/bjet.12902
[21]	Chew SW, Jhu JY, Chen NS (2018) The effect of learning English idioms using scaffolding strategy through situated learning supported by augmented reality. In 2018 IEEE 18th International Conference on Advanced Learning Technologies (ICALT). 390–394.
[22]	Chung LY (2011) Using avatars to enhance active learning: Integration of virtual reality tools into college English curriculum. In The 16th North-East Asia Symposium on Nano, Information Technology and Reliability. 29–33.
[23]	Chung LY (2012) Virtual Reality in college English curriculum: Case study of integrating second life in freshman English course. In 2012 26th International Conference on Advanced Information Networking and Applications Workshops. 250–253.
[24]	Cooper C, Varley-Campbell J, Booth A, et al. (2018) Systematic review identifies six metrics and one method for assessing literature search effectiveness but no consensus on appropriate use. J clin epidemiol 99: 53–63. doi: 10.1016/j.jclinepi.2018.02.025
[25]	Cornillie F, Clarebout G, Desmet P (2012) Between learning and playing? Exploring learners' perceptions of corrective feedback in an immersive game for English pragmatics. ReCALL: Journal of Eurocall 24: 257–278. doi: 10.1017/S0958344012000146
[26]	Dalim CSC, Dey A, Piumsomboon T, et al. (2016) TeachAR: An interactive augmented reality tool for teaching basic English to non-native children. In 2016 IEEE International Symposium on Mixed and Augmented Reality (ISMAR-Adjunct). 82–86.
[27]	De Freitas S (2006) Learning in immersive worlds: A review of game-based learning. 2–60.
[28]	De Grove F, Van Looy J, Courtois C (2010) Towards a serious game experience model: Validation, extension and adaptation of the GEQ for use in an educational context. In Playability and player experience. 10: 47–61. Breda University of Applied Sciences.
[29]	Draxler F, Labrie A, Schmidt A, et al. (2020) Augmented reality to enable users in learning case grammar from their real-world interactions. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems. 1–12.
[30]	Eisenmann M, Grimm N, Volkmann L (2010) 92. Teaching the new English cultures and literatures. English and American Studies in German 2009: 164–166. doi: 10.1515/9783484431225.164
[31]	Eisenmann M (2019) Teaching English: Differentiation and Individualisation. utb GmbH.
[32]	Fast-Berglund Å, Gong LL (2018) Testing and validating extended reality (xR) technologies in manufacturing. Procedia Manuf 25: 31–38. doi: 10.1016/j.promfg.2018.06.054
[33]	Fernández SS, Pozzo MI (2017) Intercultural competence in synchronous communication between native and non-native speakers of Spanish. Language Learning in Higher Education 7: 109–135. doi: 10.1515/cercles-2017-0003
[34]	Garzon J, Acevedo J (2019) Meta-analysis of the impact of Augmented Reality on students' learning gains. Educational Research Review 27: 244–260. doi: 10.1016/j.edurev.2019.04.001
[35]	Gelsomini M, Leonardi G, Garzotto F (2020) Embodied learning in immersive smart spaces. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems. 1–14.
[36]	Giraudeau P, Olry A, Roo JS, et al. (2019) CARDS: a mixed-reality system for collaborative learning at school. In Proceedings of the 2019 ACM International Conference on Interactive Surfaces and Spaces. 55–64.
[37]	Grünewald A (2016) Üben und Übungen im Fremdsprachenunterricht. Üben und Übungen beim Fremdsprachenlernen: Perspektiven und Konzepte für Unterricht und Forschung. Arbeitspapiere der 36. Frühjahrskonferenz zur Erforschung des Fremdsprachenunterrichts, 84.
[38]	Guillén-Nieto V, Aleson-Carbonell M (2012) Serious games and learning effectiveness: The case of It'sa Deal! Computers and Education 58: 435–448.
[39]	Hamilton D, McKechnie J, Edgerton E (2020) Immersive virtual reality as a pedagogical tool in education: a systematic literature review of quantitative learning outcomes and experimental design. Journal of Computer Education 8: 1–32. doi: 10.1007/s40692-020-00169-2
[40]	Hammer M (2012) The intercultural development inventory: A new frontier in assessment and development of intercultural competence. Sterling, VA: Stylus Publishing. In M. Vande Berg, R. M. Paige, and K. H. Lou (Eds.). Student learning abroad, 115–136.
[41]	Hao KC, Lee LC (2019) The development and evaluation of an educational game integrating augmented reality, ARCS model, and types of games for English experiment learning: an analysis of learning. Interactive Learning Environments, 1–14.
[42]	Hassani K, Nahvi A, Ahmadi A (2016) Design and implementation of an intelligent virtual environment for improving speaking and listening skills. Interactive Learning Environments 24: 252–271. doi: 10.1080/10494820.2013.846265
[43]	He J, Ren J, Zhu G, et al. (2014) Mobile-based AR application helps to promote EFL children's vocabulary study. In 2014 IEEE 14th International Conference on Advanced Learning Technologies. 431–433.
[44]	Herrera F, Bailenson J, Weisz E, et al. (2018) Building long-term empathy: A large-scale comparison of traditional and virtual reality perspective-taking. PLOS ONE 13: e0204494. doi: 10.1371/journal.pone.0204494
[45]	Ho SC, Hsieh SW, Sun PC, et al. (2017) To activate English learning: Listen and speak in real life context with an AR featured u-learning system. Journal of Educational Technology and Society 20: 176–187.
[46]	Hsieh M (2016) Development and evaluation of a mobile AR assisted learning system for English learning. 2016 International Conference on Applied System Innovation (ICASI), Okinawa, 1-4.
[47]	Hsieh MC (2016) Teachers' and students' perceptions toward augmented reality materials. In 2016 5th IIAI International Congress on Advanced Applied Informatics (IIAI-AAI). 1180–1181.
[48]	Hsu TC (2019) Effects of gender and different augmented reality learning systems on English vocabulary learning of elementary school students. Universal Access in the Information Society 18: 315–325. doi: 10.1007/s10209-017-0593-1
[49]	Huang X, Han G, He J, et al. (2018) Design and Application of a VR English Learning Game Based on the APT Model. In 2018 Seventh International Conference of Educational Innovation through Technology (EITT). 68–72.
[50]	Hung HC, Young SSC (2015) An investigation of game-embedded handheld devices to enhance English learning. Journal of Educational Computing Research 52: 548–567. doi: 10.1177/0735633115571922
[51]	Ibrahim A, Huynh B, Downey J, et al. (2018) Arbis pictus: A study of vocabulary learning with augmented reality. IEEE T Vis Comput Gr 24: 2867–2874. doi: 10.1109/TVCG.2018.2868568
[52]	Ji S, Li K, Zou L (2019) The Effect of 360-Degree Video Authentic Materials on EFL Learners' Listening Comprehension. In 2019 International Joint Conference on Information, Media and Engineering (IJCIME). 288–293.
[53]	Johnson-Glenberg MC, Birchfield DA, Tolentino L, et al. (2014) Collaborative embodied learning in mixed reality motion-capture environments: Two science studies. Journal of Educational Psychology 106: 86. doi: 10.1037/a0034008
[54]	Khatoony S (2019) An Innovative Teaching with Serious Games through Virtual Reality Assisted Language Learning. In 2019 International Serious Games Symposium (ISGS). 100–108.
[55]	Kincheloe Joe L (2008) Critical Pedagogy Primer 2nd edition. English: New York, NY: Peter Lang.
[56]	Küçük S, Yylmaz RM, Göktap Y (2014) Augmented reality for learning English: Achievement, attitude and cognitive load levels of students. Education and Science/Egitim ve Bilim 39.
[57]	Küster L (2014) Zur Einführung in den Themenschwerpunkt. Fremdsprachen lehren und lernen 43: 2.
[58]	Lan YJ (2015) Contextual EFL learning in a 3D virtual environment. Language Learning and Technology 19: 16–31.
[59]	Lee K, Kweon SO, Lee S, et al. (2014) POSTECH immersive English study (POMY): Dialog-based language learning game. IEICE T Inf Syst 97: 1830–1841. doi: 10.1587/transinf.E97.D.1830
[60]	Lee SM, Park M (2020) Reconceptualization of the context in language learning with a location-based AR app. Computer Assisted Language Learning 33: 936–959. doi: 10.1080/09588221.2019.1602545
[61]	Leyva F, Plummer CJ (2015) National Institute for Health and Care Excellence 2014 guidance on cardiac implantable electronic devices: health economics reloaded.
[62]	Li KC, Tsai CW, Chen CT, et al. (2015) The design of immersive English learning environment using augmented reality. In 2015 8th International Conference on Ubi-Media Computing (UMEDIA). 174-179.
[63]	Liaw ML (2019) EFL learners' intercultural communication in an open social virtual environment. Journal of Educational Technology and Society 22: 38–55.
[64]	Liou HC (2012) The roles of Second Life in a college computer-assisted language learning (CALL) course in Taiwan, ROC. Computer Assisted Language Learning 25: 365–382. doi: 10.1080/09588221.2011.597766
[65]	Liu IF, Chen MC, Sun YS, et al. (2010) Extending the TAM model to explore the factors that affect intention to use an online learning community. Computers and education 54: 600–610. doi: 10.1016/j.compedu.2009.09.009
[66]	Liu E, Liu C, Yang Y, et al. (2018) Design and implementation of an augmented reality application with an English Learning Lesson. In 2018 IEEE International Conference on Teaching, Assessment, and Learning for Engineering (TALE) 494–499.
[67]	Lorenzo CM, Lezcano L, Alonso SS (2013) Language Learning in Educational Virtual Worlds-a TAM Based Assessment. J UCS 19: 1615–1637.
[68]	Matveev AV (2002) The perception of intercultural communication competence by American and Russian managers with experience on multicultural teams (Doctoral dissertation, Ohio University).
[69]	Milgram P, Kishino F (1994) A taxonomy of mixed reality visual displays. IEICE T Inf Syst 77: 1321–1329.
[70]	Moher D, Liberati A, Tetzlaff J, et al. (2011) Bevorzugte Report Items für systematische Übersichten und Meta-Analysen: Das PRISMA-Statement. DMW-Deutsche Medizinische Wochenschrift 136: e9–e15. doi: 10.1055/s-0031-1272982
[71]	Neumeier P (2005) A closer look at blended learning–parameters for designing a blended learning environment for language teaching and learning. ReCALL: the Journal of EUROCALL 17: 163. doi: 10.1017/S0958344005000224
[72]	Oberdörfer S, Latoschik ME (2019) Predicting learning effects of computer games using the Gamified Knowledge Encoding Model. Entertain Comput 32: 100315. doi: 10.1016/j.entcom.2019.100315
[73]	Oberdörfer S, Elsässer A, Schraudt D, et al. (2020) Horst-The teaching frog: learning the anatomy of a frog using tangible AR. In Proceedings of the Conference on Mensch und Computer. 303–307.
[74]	Peck T, Seinfeld S, Aglioti S, et al. (2013) Putting yourself in the skin of a black avatar reduces implicit racial bias. Consciousness and cognition 22: 779–787. doi: 10.1016/j.concog.2013.04.016
[75]	Qu C, Ling Y, Heynderickx I, et al. (2015) Virtual bystanders in a language lesson: examining the effect of social evaluation, vicarious experience, cognitive consistency and praising on students' beliefs, self-efficacy and anxiety in a virtual reality environment. PloS one 10: e0125279. doi: 10.1371/journal.pone.0125279
[76]	Quintín E, Sanz C, Zangara A (2016) The impact of role-playing games through Second Life on the oral practice of linguistic and discursive sub-competences in English. In 2016 International Conference on Collaboration Technologies and Systems (CTS). 148–155.
[77]	Ratan R, Beyea D, Li B, et al. (2020) Avatar characteristics induce 'users' behavioral conformity with small-to-medium effect sizes: A meta-analysis of the proteus effect. Media Psychology 23: 651–675. doi: 10.1080/15213269.2019.1623698
[78]	Redondo B, Cózar-Gutiérrez R, González-Calero JA, et al. (2020) Integration of augmented reality in the teaching of English as a foreign language in early childhood education. Early Childhood Education Journal 48: 147–155. doi: 10.1007/s10643-019-00999-5
[79]	Ripka G, Grafe S, Latoschik ME (2020) Preservice Teachers' encounter with Social VR–Exploring Virtual Teaching and Learning Processes in Initial Teacher Education. In SITE Interactive Conference. 549–562. Association for the Advancement of Computing in Education (AACE).
[80]	Sherman WR, Craig AB (2003) Understanding virtual reality. San Francisco, CA: Morgan Kauffman.
[81]	Shih YC, Yang MT (2008) A collaborative virtual environment for situated language learning using VEC3D. Educational Technology and Society 11: 56–68.
[82]	Shih YC (2015) A virtual walk through London: Culture learning through a cultural immersion experience. Computer Assisted Language Learning 28: 407–428. doi: 10.1080/09588221.2013.851703
[83]	Skarbez R, Frederick PB, Mary CW (2018) Immersion and Coherence in a Stressful Virtual Environment. In Proceedings of the 24th Acm Symposium on Virtual Reality Software and Technology. 1–11.
[84]	Slater M, Wilbur S (1997) A framework for immersive virtual environments (FIVE): Speculations on the role of presence in virtual environments. Presence: Teleoperators and Virtual Environments 6: 603–616. doi: 10.1162/pres.1997.6.6.603
[85]	Slater M (1999) Measuring presence: A response to the Witmer and Singer presence questionnaire. Presence 8: 560–565. doi: 10.1162/105474699566477
[86]	Slater M (2003) A note on presence terminology. Presence connect 3: 1–5.
[87]	Sanchez-Vives MV, Mel S (2005) From presence to consciousness through virtual reality. Nat Rev Neurosci 6: 332–339. doi: 10.1038/nrn1651
[88]	Slater M (2009) Place illusion and plausibility can lead to realistic behaviour in immersive virtual environments. Philos T R Soc B 364: 3549–3557. doi: 10.1098/rstb.2009.0138
[89]	Tulodziecki G, Grafe S (2012) Approaches to learning with media and media literacy education–trends and current situation in Germany. Journal of Media Literacy Education 4: 5.
[90]	Tulodziecki G, Grafe S, Herzig B (2019) Medienbildung in Schule und Unterricht: Grundlagen und Beispiele. UTB GmbH.
[91]	Vate-U-Lan P (2012) An augmented reality 3d pop-up book: the development of a multimedia project for English language teaching. In 2012 IEEE International Conference on Multimedia and Expo. 890–895.
[92]	Vedadi S, Abdullah ZB, Cheok AD (2019) The Effects of Multi-Sensory Augmented Reality on Students' Motivation in English Language Learning. In 2019 IEEE Global Engineering Education Conference (EDUCON). 1079–1086.
[93]	Wang CX, Calandra B, Hibbard ST, et al. (2012) Learning effects of an experimental EFL program in Second Life. Educational Technology Research and Development 60: 943–961. doi: 10.1007/s11423-012-9259-0
[94]	Wang YF, Petrina S, Feng F (2017) VILLAGE—V irtual I mmersive L anguage L earning and G aming E nvironment: Immersion and presence. British Journal of Educational Technology 48: 431–450. doi: 10.1111/bjet.12388
[95]	Wienrich C, Johanna G (2020) AppRaiseVR–An Evaluation Framework for Immersive Experiences. I-Com 19: 103–121. doi: 10.1515/icom-2020-0008
[96]	Wienrich C, Döllinger NI, Hein R (2020) Mind the Gap: A Framework (BehaveFIT) Guiding The Use of Immersive Technologies in Behavior Change Processes. arXiv preprint arXiv: 2012.10912.
[97]	Wienrich C, Eisenmann M, Latoschik ME, et al. (2020) CoTeach - Connected Teacher Education. VRinSight Greenpaper 53–55.
[98]	Witmer BG, Singer MJ (1998) Measuring presence in virtual environments: A presence questionnaire. Presence 7: 225–240. doi: 10.1162/105474698565686
[99]	Wohlgenannt I, Simons A, Stieglitz S (2020) Virtual Reality. Business and Information Systems Engineering 5: 455–461. doi: 10.1007/s12599-020-00658-9
[100]	Wu MH (2019) The applications and effects of learning English through augmented reality: A case study of Pokémon go. Computer Assisted Language Learning, 1–35.
[101]	Yang MT, Liao WC (2014) Computer-assisted culture learning in an online augmented reality environment based on free-hand gesture interaction. IEEE T Learn Technol 7: 107–117. doi: 10.1109/TLT.2014.2307297
[102]	Yeh HC, Tseng SS, Heng L (2020) Enhancing EFL students' intracultural learning through virtual reality. Interactive Learning Environments, 1–10.
[103]	Zhang X, Zhou M (2019) Interventions to promote 'learners' intercultural competence: A meta-analysis. International Journal of Intercultural Relations 71: 31–47. doi: 10.1016/j.ijintrel.2019.04.006

Reader Comments

Your name:*

Email:*
© 2021 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)