Unformatted Attachment Preview
de Freitas, S. (2018). Are Games Effective Learning Tools? A Review of Educational Games. Educational Technology &
Society, 21 (2), 74–84.
Are Games Effective Learning Tools? A Review of Educational Games
Sara de Freitas
Birkbeck College, University of London, Malet Street, London, United Kingdom // firstname.lastname@example.org
(Submitted December 7, 2016; Accepted January 9, 2017)
The literature around the use, efficacy and design of educational games and game-based learning
approaches has been building up gradually and in phases, across different disciplines and in an ad hoc way.
This has been problematic in a number of ways and resulted in fragmented literature and inconsistent
referencing patterns between different sub-disciplines and countries. This is mainly because no distinct
single-disciplinary perspective has emerged because of: the cross-disciplinary nature of educational games,
a reliance on single-disciplinary contexts for studies, changing terminologies in different contexts and the
use of multi-methodological approaches. Distinct perspectives from education science, game science,
neuroscience and information science have deepened our understanding of play and games. This research
has become more quantitative, rigorous and nuanced as a result of more studies focused upon therapeutic
health applications of games, the serious games research movement and more efficacy and comparative
studies that examine and quantify utility.
Educational games, Serious games, Game science, Neuroscience and games
Defining efficacy in educational contexts can be challenging due to the range of variables involved in different
learning contexts. Additionally, there are disciplinary restraints that have traditionally meant that crossdisciplinary approaches to data collection and analysis have been broadly discouraged. However to understand
education, and in particular questions around efficacy, necessarily we need to adopt more cross-disciplinary
approaches. As an example, research emerging from education science is being supplemented by findings from
computer science (e.g., interfaces and interactivity), neuroscience (e.g., brain function and activity) and
information science (e.g., analytics and user-modelling). Notably these include findings from computer science
which allow us to consider usability improvements and human-computer interactions (e.g., Barr et al., 2007),
findings from neuroscience which provide a greater understanding of how games impact our brain plasticity
(e.g., Bavelier et al., 2012; Kühn et al., 2011; Kühn et al., 2014) and approaches that use analytics in games as a
replacement for assessment (e.g., Serrano-Laguna et al., 2012). Together, these findings help provide a broader
understanding of how we can model learning experiences in digital, data-rich game environments, and tell us
more about how we learn.
The review found that “game science” is emerging as a new term to replace “serious games” which has been a
significant term for the game studies research community for the last ten years. Similar to “serious games”, the
new term aims to link game studies to a greater scientific capability which has the potential to help us model and
better understand: the learning behaviours of individuals and groups in game environments, learning design
through the metaphor of game design and how games and play work to help people learn.
Establishing the efficacy of games and learning is a complicated endeavour. It needs to be kept within a wider
context of understanding how we learn. So how game science fits into the wider disciplinary framework is a
critical consideration. When viewed from this educational perspective, the notion of “game science” is part of
the field we might call, “education science” and due to its digital nature it is often placed within the subdiscipline of Technology-Enhanced Learning (TEL). However, clearly there is important work to be found across
a range of different areas including: human-computer interaction (e.g., Barr et al., 2007), health education and
research (e.g., Papastergiou, 2009), neuroscience research (e.g., Kühn et al., 2011; Colzato et al., 2013; Lewis,
2013; Kühn et al., 2014), and across other literature such as business and management (e.g., Pasin & Giroux,
2011), school education (e.g., Hainey et al., 2016), advertising and marketing (e.g., Terlutter & Capella, 2013),
military training and simulations (e.g., Hassain et al., 2012), environmental awareness-raising (RebolledoMendez et al., 2009), therapy training (Horne-Moyer et al., 2014), teacher training (e.g., Kenny & McDaniel,
2011) and emergency-response training (e.g., Chen et al., 2008). One challenge with the literature so scattered is
that not all researchers acknowledge the breadth of the area and range of applications, and therefore miss vital
academic contributions by looking too narrowly at the literature-base. The situation is exacerbated by rifts
between US and European research in serious games and between simulation and games literatures, and often
you can see researchers will completely ignore critical papers from one “side” of the Atlantic or the other leading
ISSN 1436-4522 (online) and 1176-3647 (print). This article of the Journal of Educational Technology & Society is available under Creative Commons CC-BY-ND-NC
3.0 license (https://creativecommons.org/licenses/by-nc-nd/3.0/). For further queries, please contact Journal Editors at email@example.com
to misunderstandings and incomplete starting points. Despite a number of special issues on games crossing
several fields, the continuation of fragmentation of the field has again happened with the split between
researchers in serious games and the new area of “gamification”. Gamification here is used to mean the
application of elements of game-mechanics and/or game-design techniques. To attempt to bring the literatures
closer together and to attempt to harmonize some of the terminology, this paper aims to map out the potential
new ground for learning as evidenced in the sub-field of technology enhanced learning that defines game-based
To overcome these significant disciplinary challenges, this paper seeks to outline some of the major
contributions of the field from different disciplines over time and synthesise these using an integrative approach
to a broader education science perspective. The aim is to problematize the current scope of Education Studies
and to reposition game science more critically within this educational context and perspective.
Methodology of literature review
This article has adopted a “grounded theory” approach used over a number of years to assess the main themes
emerging from the fields that touch on educational games. The method used included a semi-systematic review
process with a single-coder, wide literature searches across databases using keywords to collect high impact and
cited articles and is supplemented with a journal hand-search. Keywords included educational games, serious
games, learning games, web-based games and digital games. Once key texts were identified from the literature
search, these were grouped into disciplinary perspectives. The emerging perspectives of education science, game
science/studies, neuroscience and information science were distilled and key articles identified were included in
The Review: A recent history of game science
Wave 1: What are games?
Some of the earliest work in the field of game science focused upon, changing definitions and nominations of
educational games. For example, work that outlined classifications of games, typologies and ontologies was
found in the early literature (e.g., Caillois & Barash, 1961; Sutton-Smith & Roberts, 1971). While the earlier
work focused upon structuralist perspectives upon educational games as consistent with the trend for semiotics
and structuralist analysis, the theme re-emerged later on in the more recent literature as a theme of consideration
(e.g., Elverdam & Aarseth, 2007; Kamii & DeVries, 1980; Salen & Zimmerman, 2004). But the more consistent
theme of poststructuralism and postmodernist perspectives necessarily focused more upon notions of play than
structure also in line with constructivist and qualitative studies.
It is perhaps ironic that constructivist approaches to learning have become so associated with qualitative
approaches as although the work does focus upon individual construction of meaning, the social constructivism
of Vygotsky (1980) and others does propose learning in social groups as a central component of learning. But
here a split between the American and other literatures can be noted as a de-emphasis of social learning and a
greater focus upon Skinnerism and behaviourist approaches as consistent with the individualism and competition
of the American ideal. The mode of bringing education theory together with an American individualist twist and
its bringing into the paradigm of psychology jointly ensured that the more social focus emphasised by Russian
theorist Vygotsky did not become the dominant discourse. The legacy of this can also be seen in the more general
sparseness of social learning theory and was compounded by difficulties with researching and analysing group
work, a trend that is partially being reversed in studies such as Star where collaboration rather than competition
techniques are emerging (Star, 2015).
Wave 2: The serious games movement
Negative publicity around violence in games, in particular entertainment games have attracted popular attention.
The robust evidence of games causing violence has overall been inconclusive (e.g., Elson & Ferguson, 2015) –
but nonetheless the distinction between games for entertainment and games for non-entertainment was a major
driver for why the “serious games movement” occurred in the early 2000s (Blumberg et al., 2013). However,
once non-entertainment games could be demonstrably “taken seriously” for purposes such as military training
and health education and therapy then the research field gained greater credibility.
Early “serious games” titles, such as America’s Army, have set the bar high in terms of the budget ($33 million
invested up until 2015 in all titles). Although small budgets next to entertainment games, (e.g., $265 million for
Grand Theft Auto 5), America’s Army is still considered one of the best exemplars of a serious game today.
Having been first published in 2002, it has 13 million registered players who have played 260 million hours.
Developed by the US Moves Institute to solve the recruitment problems of the US Army, unfortunately the game
has proved to be more of an oddity than a trend. Few large budget serious games have been developed since
2002, and those that have been commissioned have not always enjoyed longevity of support once piloting phases
have concluded, e.g., Code of Everand (Dunwell et al., 2014). During this period, although relatively
disconnected from the mainstream games literature, the “serious games movement” did gain important
contributions from game studies, such as a deeper understanding of the mechanisms of competition as a design
component (Cagiltay et al., 2015), how to balance entertainment principles of fun with instructional design and
the need to integrate teams of developers, writers and instructional designers.
Wave 3: Technology-enhanced learning perspectives: Out of the wilderness?
The next phase of focus upon educational games borrowed heavily from technology-enhanced learning
approaches. There, a focus upon verification and validation of online learning and e-learning was leading to a
wide range of comparative learning studies. Again studies were often lacking in robust methodologies, but were
beginning to seek a more scientific basis for analysing the efficacy of learning techniques. This approach was
driven-out of concerns about the quality of learning in online settings and studies were often more utilityfocused. While the early studies had attempted to group games in typologies and genres, these studies focused
upon comparisons with other e-learning formats and against traditional learning measures (e.g., Knight et al.,
Out of this work, a movement to understand game design emerged, how could games be designed for different
learning contexts? Could commercial off-the-shelf (COTS) games be used? These questions led to a range of
studies of games in educational contexts and collections of case studies (e.g., Kim et al., 2009; Michael & Chen,
2005; Prensky, 2005; Shute et al., 2009). This phase of research was dominated by educational perspectives.
However, there were significant difficulties with uptake of games in educational contexts. As Simon EgenfeldtNielsen outlined in his thesis (Egenfeldt-Nielsen, 2005), games did not fit into the one-hour lessons, into the
single disciplinary focus or into the single-teacher model of traditional learning. Games were disruptive, they
demanded greater changes to the traditional delivery and infrastructure of education in schools, colleges and
universities. Beyond traditional learning paradigms (see Table 1), game-based approaches required: crossdisciplinarity, longer class durations, mixed student groups, social learning and team-teaching models to come
into place to really capitalise on the merits of the game and gameplay as learning approaches (de Freitas, 2014).
Four disciplinary perspectives from the literature
While it is difficult to be too prescriptive with the time periods, the research does seem to fall broadly into four
broad disciplinary categories: education science including theory and practice studies and using elements of
pedagogy and psychology, game science contextualised through technology enhanced learning, neuroscience that
have focused upon brain-function and plasticity and information science-driven studies that focus more upon
data analytics and behavioural modelling. The following sections outline these perspectives (see summary in
Education science perspective on educational games
Major contributions to understanding learning formed early theoretical and developmental approaches to
learning. Through understanding learning as cognitive and developmental sets of processes, theorists and
educationalists, such as Jean Piaget, defined ages and stages of development associated with “normal patterns of
development” (Piaget, 1971). But Piaget also understood the importance of play in learning (Piaget, 2013). Play
has been a theme of the work around games necessarily, but has not been a well-understood aspect of learning.
More recent play research by Jean Twenge and others shows how important and developmental play is to
learning (e.g., Campbell & Twenge, 2015; Chudacoff, 2007; de Freitas, 2014; Gray, 2011; Twenge & Campbell,
In the light of the internet, broadened connectivity and mobile access to online educational content, there has
been a de-emphasis on content and curriculum and a sharpened focus upon digital literacy and 21st century skills.
Employability for the changing global employment market presents new needs for graduates and students
(Harlow & Bowman, 2016). The move to a more utilitarian position, driven by education via web-based
technologies and digitisation, has reworked how we deliver a university education and even challenged what the
role of the university is (Sugden, 2013).
Table 1. Comparing the traditional, new learning and future learning approaches
Traditional paradigm of learning
New learning paradigm
Challenge and activity-led learning
Student developed pedagogy
Tutor-led learning delivery
Artificial Intelligence (AI)
Classroom and lecture hall focus
Any-time, anywhere learning
Seamless lifelong learning
Formative assessment / Peer
No assessments / levelling,
points and awards
Age and stage
Competency and personalised
Unique learning patterns
Traditional curriculum e.g.,
New curriculum e.g., 21st century
Hidden curriculum e.g.
literacy and numeracy
personalised skills and
Blended work and learning
In the author’s recent work, she articulates this disruption as a “new learning” paradigm. One that focuses upon
problem-, challenge- and active pedagogy, peer learning and is competency-based and personalised (de Freitas,
2014). This differs from the traditional modes of curriculum-based and tutor-led approaches. With the work on
games we can begin to see the rudiments of what the author calls a “future learning’” paradigm, which advances
to student-led approaches where adaptive learning is scaffolded through AI bots, assessment gives way to inbuilt levelling-up and the curriculum is hidden (See Table 1).
Game science perspective upon educational games
One of the main stated inhibitors to uptake of educational games and approaches was the lack of robust scientific
and evidence-based research. The first randomised and pragmatic randomised controlled trials (RCTs/PCTs)
started in the late 2000s. One of the early trials was undertaken by Knight et al. (2010), focusing upon a
comparison between traditional and game-based approaches in emergency response training. Arnab et al.
undertook an RCT which considered a serious game in a classroom setting. Miller and Robertson undertook an
RCT on educational benefits of games consoles in classrooms (Miller & Robertson, 2011). While Star
considered a randomised control trial for gamification in StarQuest to identify cooperative and competitive
design elements in university students (Star, 2015). Arbogast et al. (2014) were examining the use of an
educational game for road crossing in their recent study.
Unsurprisingly most recently RCTs involving games have focused upon health and medical conditions including
patients with weight conditions (e.g., Ahola et al., 2013; Maddison et al., 2011; Siervo et al., 2013; Straker et al.,
2011; Straker et al., 2013). Fung et al. (2012) considered the use of the Wii Fit for knee rehabilitation. Foss et al.
(2013) used their randomised control trial to discover effective use of the i-Bit which is a novel binocular device
which uses games and videos to improve patients with a lazy eye. Picherri et al. (2012) looked at the impact of a
dance game upon gait. Another popular area for study was the impact of games upon the elderly. An interesting
study by Nouchi et al. (2012) explored the positive impact on executive and processing speeds on the elderly of
brain training games in their study. While Mayas et al. (2014) explored the plasticity of the brain in the elderly
after non-violent game play. A study on Wii Fit games for patient’s living with Parkinson’s disease was
undertaken recently by Pompeu et al. (2012); and one looking at improvements from gameplay with Diabetes
sufferers (Kempf & Martin, 2013). Allam et al. (2015) in their RCT on gamification in an online intervention for
Rheumatoid Arthritis Patients found that “physical activity increased over time for patients having access to
social support sections plus gaming (unstandardised beta coefficient β = 3.39, p = .02).” Patients were also more
empowered and used services less as a result.
In addition to more quantitative studies such as RCTs/PCTs, meta reviews have offered important research
contributions to overcoming the prevalence of different disciplinary perspectives. Often these reviews have been
cross-disciplinary in scope and dimension, single topic-focused, centred-upon comparative studies or in support
of game design improvements. While there was a large group of studies done on violence in games (e.g.,
Anderson & Bushman, 2001; Anderson et al., 2010), these studies do not ...