Situatedness and learning: does environment impact second learning performance in a language-learning game?

Introduction

Immersive virtual environments (IVEs) are increasingly being explored for their use as educational aids. There are many reasons for their use, including motivational benefits (???), interaction and contextual feedback (???), and enabling collaboration (???). In computer aided language learning (CALL), another aspect is also being explored: situatedness. From both cognitive science (???) and applied linguistic (???) perspectives, how situated a learner feels in a contextually relevant environment is considered closely related to how they perform cognitively.

IVEs provide a much deeper sense of situatedness than previous forms of computer interaction (???), but how this relates to language learning is underexplored. Many examples of language IVE systems aim to recreate a contextually relevant learning environment (???), but few (if any) have explored the extent to which the relevance of the context provides benefits over an irrelevant alternative. Context is a wide term, covering everything from the viability of the interactions, to the perceived motivation for their use, to the environment’s graphic setting. Understanding how a contextually relevant IVE impact learning has important implications for learning system design going forward, as well as understanding how environments can be used to affect players in serious games, and to invoke outcomes such as incidental learning (???).

Beyond this, research into learning software design has also shown that unrealistic or impossible situations are beneficial for cognition, beyond which could be achieved in real-world settings (??? In VR boys learn best when the teacher is a drone – girls lean better from virtual Marie). The impact of a situated “hyper reality” (???) on language learning also underexplored.

We propose a comparative experiment that compares outcomes from a second language learning game set in contextually irrelevant, contextually relevant, and hyper-reality scenarios. We will monitor pre- and post-test performance, as well as covariables such as motivation, presence and cognitive load, to determine the impact of different types of situatedness on learner experience.

Experiment

Game set in Japanese ramen restaurant. You are the chef. Waiter gives you orders, you prepare ingredients.

  • Number
  • Food Item (Tonkotsu, Shoyu, Shio, Miso)
  • Ingredients (Spring Onion, Radish, Tofu, Noodle, Egg, Stock, Salt, Soy, Pork, Cabbage, Bean sprouts, Bamboo Shoots, Kamaboko)
  • Verb: Chop, Grate, Pour, Stir, Move, Serve
  • Feedback words: Tasty, Disgusting,

Three settings:

  • No customer, no waiter, no sous chef, no restaurant. Just actions in abstract space
  • Customer, waiter, sous chef, restaurant
  • Aliens & Robots. Space-restaurant.

Literature

In the search for ever more effective learning systems, researchers have begun exploring the use of immersive virtual environments (IVEs) as a tool for leveraging the benefits of high levels of situatedness. Situatedness is the theoretical position that cognition is intertwined with environmental, social, and cultural factors (Costello, 2014). It does not describe a different type of cognition, but posits that “situativity” is fundamental to all cognitive activity (Norman, 1993; Vera & Simon, 1993). Situatedness is closely linked with embodied theories of cognition: cognition is situated in the environment and we off-load cognitive work onto our environments (Wilson, 2002).

Although there is general agreement that situatedness contributes to learning (Brown, Collins, and Duguid 1989), there is still some discussion about what constitutes a situated learning experience, the aspects which contribute to the creation of a “situation”, the contribution of each of those factors, and whether these distinctions are even possible (Greeno, 1998). There are important considerations when constructing artificial simulations, both in the real world or inside IVEs. One view of situated learning (Lave 1988; Shor 1987) splits the instructional process into four areas:

  • content: the facts and processes of the task;
  • context: the situations, values, beliefs, and environmental cues by which the learner gains and masters content;
  • community: the group with which the learner will create and negotiate meaning of the situation;
  • participation: the process by which learners working together and with experts in a social organization solve problems related to everyday life circumstances

Context

Context is perhaps the most relevent aspect for investigations into learning in IVEs, as immersive environments offer a unique ability to recreate more beleivable situations (???) and provide enhanced, spatial and persuasive environment cues (???). But what constitutes a situated context? According to Rogoff, a situated learning context is “the … physical and conceptual structure as well as the purpose of the activity and the social milieu in which it is embedded.” (1984) Context, therefore, includes the general atmosphere and physical setting as well as concurrent “background” events (Ruth, 1992). The benefits of context stem from its ability to learners to “learn how to exploit contextual resources for their goals by looking for, recognizing, evaluating, and using information resources productively.” (Harley, 1993)

Sitauted context has further been broken down into two types: inter-context and intra-context (Rehm, 2003). Inter-context are the global contexts of the environment, such as language, setting, expected behaviour. They are not singular (e.g. “office”), but rather, an overlap of numerous aspects (e.g. an English-language meeting in a professional office environment). The intra-context is an individual's understanding of these interacting inter-context subsystems, and the way they understand how to act in that setting. In creating an immersive virtual learning enviroment, designers build an inter-context that, ideally, will promote a strong intra-contextual response inside the learner, and thus create a strong learning response. Roschelle and Clancey (1992) explain this relationship as: “representation and meaning are not prior to social and physical interaction but are constructed in activity” - something greatly hindered without a functional inter-context.

Authenticity

Most situated learning research attempts to establish their inter-context by closely recreating real-world environments where the target knowledge is typically used (Wilson, 1993). This “authentic” context cues the learner to situational resources and serves as an advance organizer for related problem-solving contexts (Choi, 1995). Authentic tasks are “coherent, meaningful, and purposeful activities that represent the ordinary practices of a culture” (Brown et al., 1989). They are often problem based (Choi), which means students are better able to gauge what they are learning and how to use it (Collins, 1993; Collins, Brown, & Newman, 1989). For language learning, this is particlarly useful, as task-based language learning is one of the predominant approaches to tuition (Ellis, 2003). Authentic environments also show added motivational potential, as students are more likely to become self-referenced and purposefully engaged by learners (Choi).

Authenticity is missing from most traditional classroom settings, where students are frequently given problems or tasks that are of little relevance and bear little meaning to an authentic setting, and therefore provide little inter-context (Newmann, 1991). Some approaches try to remedy this: language teachers have found that using language context-inducing props and prompts - realia - is useful for language tuition (Nelson, 1965).

Situated language learning

Language learning is widely considered as a learning process that is deeply related with situatedness. This has given rise to many explorations into situated teaching processes, particularly immersion learning programmes. Immersion learning programmes involve sending language students to live amongst and use their target language. Outcomes from the study abroad process are generally favourable, including positive of attitudes toward the community and the language, less anxiety when using the language, more intention to speak it (Desrochers &Gardner, 1981), increased participation in language program (Ingram, 2005), gains in the test scores on grammar and reading (Freed, 1990) and contributions to fluency and naturalness of speech (Coleman, 1997).

However, Stein (1997) found that despite increase fluency, immersion students also showed gaps in grammatical accuracy. This sounds like a predictable outcome from highly situated, task-based learning, as language is prioritised as a tool used to accomplish an action (and therefore meaning tobe understood), rather than to acheive perfect grammatical expression.

Video game context

It is important to stress that context, and authenticity, do not nesessiarily mean the creation of realistic simulations of real world situations. The aspects of context: physical structure, conceptual structure purposeful activity and the social milieu, can be contextually-relevent for a situation in digital and fictional experiences, especially evident in massively multiplayer online role-playing games (Sylvén & Sundqvist). There are many examples of incidental second language learning occuring through interactions with World of Warcraft (Rama 2012; Zheng 2012; Maior). This could be because, although the fantasy setting is not realistic, it is authentic - there is a “realistic problem-solving processes” and a “cognitive realism” (Smith 1986) of the situated tasks. Smith argues that it is the cognitive authenticity, rather than a physical authenticity (1987), which is important to situated learning. For example, in learning tennis, it does not matter if a simulated tennis ball has the correct weight if you attempt to pick it up in an IVE, but that you believe that the tennis ball will act correctly when you provide the correct contextual input.

Generally, games with a focus on situatedness have been shown beneficial for learning (Vogel et al. 2006; Dawley & Dede), and that more realistic 3D games perform better than 2D ones, due to “more accurately simulated and authentic learning environments” (Liao & Chen, 2007). Critical and reflective analysis of VR tools and their underlying concepts shows that 3D environments are ideal for language learning by providing an situated, learner-centered learning environment (Schwienhorst, 2002).

For IVEs, there is evidence that belief and behaviour concerning a simulated environment is similar to that of a real situated environment (Heydarian, Carneiro & Gerber, 2015). Evidence has shown that these environments are engaging and convincing enough to create anxiety in participants (Powell, Chalmers), to successfully teach and encourage the correct deployment of cultural gestures (Cheng, Yang, Andersen, 2017), break down pragmatic instruction barriers in the foreign-language classroom (Sykes, 2012), encourage communicative competence in second language tuition (Johnson, Valente, 2009), help memorise foreign language words (Ebert, Gupta, Makedon, 2016) and gestures (Vazquez, 2018), as well as provide engaging language learning experience (Chang, Sheldon, Si and Hand, 2012).

While we have seen IVEs leveraging situatedness to provide more engaging learning and emotional experiences, and situatedness is strongly considered beneficial for learning outcomes, there are few comparative studies controling for a situatedness variable in IVEs. We believe that to truly understand the interplay of situatedness and IVEs, this variable needs to be explored experimentally.

Beyond reality & novelty

IVEs allow us to more faithfully and engagingly recreate situated environments, but they can also do more: VR is to go beyond what is real, it is more than simulation, it is also creation, allowing us to step out of the bounds of reality and experience paradigms that are otherwise impossible (Slater, 2016) - they can allow us to tailor environment to amplify learning outcomes. A powerful example of this is when (Bailenson 2008) broke the rules of spatial proximity that exist in physical space, to place all students in a classroom at the center of the teacher's field of view (compared to the periphery) and by being closer to the teacher (compared to farther away). This led to stronger learning outcomes across the group then a non-augmented control.

One potential method for going beyond reality is to encorporate the novelty learning effect into the environment, as there is a well-evidenced link between novelty and memorisation (Bunzeck, Düzel, Neuron, 2006).

Inside studies of novelty and memorisation, there is much discussion about designing to enagage the positive benefits of novelty, without leading to distraction or alienation. There is some evidence that experiences that share a commonality with past experiences ('common novelty') provide the best learning outcomes (Duszkiewicz, McNamara, Takeuchi, 2019). The benefits of a commonality has also been noted when learning in IVEs, with objects needing to be consistent with a learning 'schema' (e.g. the environment they are presented in) for memorisation to be more efficient (Mania, Robinson, Brandt, 2005).

The benefits of novelty on learning also extend beyond the time of encoding. Ballarini et al found that, as long as a novel experience happens within a few hours of a learning occasion, memorisation is improved (Ballarini, Martínez, Perez, Moncada, 2013).

By incorporating novelty into this research experiment, we can understand if it can help us answer one of the ultimate questions of IVE learning: how do we improve on real-world tuition?