Physical Education and Sport Pedagogy

ISSN: 1740-8989 (Print) 1742-5786 (Online) Journal homepage: http://www.tandfonline.com/loi/cpes20

How does TGfU work?: examining the relationship

between learning design in TGfU and a nonlinear
pedagogy
Clara Wee Keat Tan , Jia Yi Chow & Keith Davids
To cite this article: Clara Wee Keat Tan , Jia Yi Chow & Keith Davids (2012) How does TGfU
work?: examining the relationship between learning design in TGfU and a nonlinear pedagogy,
Physical Education and Sport Pedagogy, 17:4, 331-348, DOI: 10.1080/17408989.2011.582486
To link to this article: http://dx.doi.org/10.1080/17408989.2011.582486

Published online: 17 Nov 2011.

Submit your article to this journal

Article views: 2965

View related articles

Citing articles: 9 View citing articles

Full Terms & Conditions of access and use can be found at

http://www.tandfonline.com/action/journalInformation?journalCode=cpes20
Download by: [Universiti Pendidikan Sultan Idris]

Date: 29 September 2015, At: 00:38

Physical Education and Sport Pedagogy

Vol. 17, No. 4, September 2012, 331 348

How does TGfU work?: examining the relationship between

learning design in TGfU and a nonlinear pedagogy
Clara Wee Keat Tana , Jia Yi Chowa and Keith Davidsb
Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

National Institute of Education, Physical Education and Sports Science, Nanyang Technological
University, Singapore; bSchool of Human Movement Studies, Queensland University of
Technology, Brisbane, Australia
(Received 24 June 2010; nal version received 9 March 2011)
Background: In the last few decades, conceptions about teaching and learning in
physical education have evolved from a teacher-centred approach to a more studentcentred approach where learners are encouraged to develop problem-solving skills,
critical thinking and autonomy of thought. A popular model advocating this approach
in physical education, Teaching Games for Understanding (TGfU), has attracted
widespread attention. Although advocates of TGfU have provided some empirical
and anecdotal evidence to support the tactical over technical approach to games
teaching, recent work has highlighted that to date, the question Does TGfU work?
has remained largely unanswered. Therefore, there is a need to research the intuitive
assumptions about how students learn to play games and to understand how the
TGfU approach might work for games teaching and learning.
Purpose: The purpose of this paper is to provide insights to further our understanding of
the possible processes underpinning the pedagogical principles of TGfU in games
teaching. In this regard, we outline how a Nonlinear Pedagogy approach could
provide a theoretical rationale to explain how the principles of TGfU might support
learning design for games teaching. To achieve this aim, we examined the viability of
the four key pedagogical principles of the TGfU model and highlighted the
theoretical and practical implications of Nonlinear Pedagogy, considered with some
empirical evidence from the motor learning literature.
Findings: The theoretical ideas emanating from an ecological dynamics perspective,
such as constraints manipulation, importance of maintaining information-movement
coupling and harnessing movement variability, can underpin a Nonlinear Pedagogy
approach. It has been proposed that research evidence from the motor learning
literature can provide a suitable theoretical grounding to support the viability of the
four main pedagogical principles of the TGfU model (i.e., sampling, tactical
complexity, representation and exaggeration) and can contribute insights to the
possible processes of TGfU in games teaching.
Summary: A Nonlinear Pedagogy approach has the potential to provide researchers and
physical educators with an understanding of the theoretical and practical work on TGfU,
in association with its pedagogical principles. Understanding the underlying processes
linked to the key pedagogical principles in learning design is critical for addressing
pedagogy and curriculum concerns in physical education to enhance student learning.
The ideas raised in this paper provided some rationale for the efcacy of the model,
and also a platform for researchers and practitioners to more effectively engage
students using the TGfU model.

Corresponding author. Email: clara.tan@nie.edu.sg

ISSN 1740-8989 print/ISSN 1742-5786 online

# 2012 Association for Physical Education
http://dx.doi.org/10.1080/17408989.2011.582486
http://www.tandfonline.com

332

C.W.K. Tan et al.

Keywords: TGfU; pedagogical principles; nonlinear pedagogy; learning design;
physical education

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

Introduction
In the last few decades, conceptions about teaching and learning in physical education have
changed. The perspective of the teaching-learning process has evolved from a teachercentred approach to a more student-centred approach where students are encouraged to
develop problem-solving skills, critical thinking and autonomy of thought (e.g., Lee
2003; Richard and Wallian 2005). Constructivist approaches in physical education have
conceptualised students as active learners, with individual needs, and teachers as facilitators
in physical education (Lee 2003). Consequently, there has been a shift in research focus
from the process/product paradigm, subject-matter knowledge and assessment modalities, to questions about how learners individually construct their knowledge in relation
to their learning environment (Richard and Wallian 2005).
One particular constructivist model in physical education, Teaching Games for Understanding (TGfU) (Bunker and Thorpe 1982; Thorpe and Bunker 1989; Thorpe, Bunker, and
Almond 1984), has become popular, attracting widespread attention. TGfU is a game- and
learner-centred model to learning games. The original six-step TGfU model, coupled with
four fundamental pedagogical principles (sampling, representation, exaggeration and tactical complexity), was presented as a curriculum model for developing decision making and
skill performance in games (Thorpe and Bunker 1989; Thorpe, Bunker, and Almond 1984).
In recent years, the TGfU model has been simplied by Grifn, Mitchell, and Oslin (1997)
to a three-stage model that focuses on three essential lesson components of the model (i.e.,
game form, tactical awareness and skill execution) for practitioners when teaching for tactical awareness and skill acquisition. There have also been attempts by Holt, Strean, and
Bengoechea (2002) and Kirk and MacPhail (2002) to clarify the models conceptual underpinnings (Mitchell 2003). To summarise, the TGfU model has shifted games learning from
a traditional highly structured, technique-based emphasis with a focus on technical development or knowledge content, to a more student-based approach that teaches both tactics
and skills in small-game learning contexts (Bunker and Thorpe 1982; Thorpe and
Bunker 1989).
Some advocates of TGfU (e.g., Berkowitz 1996; Booth 1983; Burrows 1986) have
tended to rely on anecdotal evidence to support the tactical over a technical approach to
games teaching. However, TGfU research, which has remained somewhat xated on comparing its relative benets compared to more traditional methods (i.e., teaching tactics or
techniques rst) to determine which is superior, has provided somewhat equivocal ndings.
Although some studies (e.g., Mitchell, Grifn, and Oslin 1995; Turner and Martinek 1999)
have reported that the tactical approach resulted in better performance outcome measures
for some aspects of skill execution and cognitive processes, other work has found both
approaches to be equally effective with few signicant differences between them (e.g.,
French et al. 1996; Mitchell et al. 1995; Turner 1996; Turner and Martinek 1992).
In recent years, researchers have been challenged to go beyond the dualist debate concerning technical vs. tactical approaches, to examine and understand the processes underlying games learning and teaching, more generally. As argued by Rink (2001, 123), when
you spend all of your effort proving that a particular kind of teaching is better than another
kind of teaching, you limit what you can learn about the very complex teaching/learning
process. While previous TGfU work highlighted that the model, similar to other models
such as Sport Education, has been interpreted and adopted in many and varied ways,

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

Physical Education and Sport Pedagogy

333

Grifn and Patton (2005) noted that the TGfU model and associated pedagogical principles
have only recently begun to receive some much needed critical evaluation, close examination and re-conceptualisation. To this end, recent work on TGfU (e.g., Butler, Grifn,
and Nastasi 2003; Grifn and Patton 2005; Rink 2001) has stressed the importance of
examining and understanding the processes involved in learning with that method.
A key issue noted by Rink (2001) was the lack of discussion about the viability of
problem-solving methodologies and inadequate attention paid to the validity of assumptions about what children learn and how they learn it. Recent work has highlighted the
absence from current discourse of support for assumptions about learning design in
TGfU (e.g., Butler et al. 2008; Grifn and Patton 2005). In this sense, it could be argued
that analyses of TGfU processes have tended to have a stronger operational, rather than
theoretical focus. Although TGfU has been predicated on a constructivist theory of knowledge and learning, more empirical support is needed to support the somewhat intuitive
assumptions that the approach works for games teaching and learning (Butler et al.
2008; Grifn, Brooker, and Patton 2005; Grifn and Patton 2005). In fact, Butler et al.
(2008) have proposed that to date, Thorpe and Bunkers (1986) question to researchers,
Does TGfU work? has remained largely unanswered. However, it is clear that previous
work on TGfU has provided some responses to Thorpe and Bunkers (1986) question
Does TGfU work? It could be argued that the notion of whether TGfU works or not is
perhaps dependent on the agreement of the desired outcomes of the approach (e.g., affective
outcomes such as more enjoyment or performance outcome such as skill acquisition). It is
apparent that TGfU, similar to other models, can be used with varying agendas. As such, it
is important for this variation to be considered when addressing whether the approach
works or not.
In recent years, several researchers (e.g., Butler et al. 2008; Grifn et al. 2005; Grifn
and Patton 2005) have suggested a number of potential theoretical perspectives to examine
the assumptions of the TGfU model and to further our understanding of the processes
involved in TGfU. Three possible theoretical frameworks, namely achievement goal
theory, information processing and situated learning, have been proposed as a relevant
grounding for examining the efcacy of TGfU as a pedagogical approach (Grifn,
Brooker, and Patton 2005). However, these proposed theoretical explanations have also
drawn some criticism. For example, while Chow et al. (2007) acknowledged that these
three theoretical frameworks have provided some relevant theoretical support for the efcacy of TGfU, they also drew attention to some shortcomings of these theoretical
frameworks.
With regard to achievement goal theory, Chow et al. (2007) commented that this explanation seemed to provide a limited picture, from a psychological and affective perspective,
to further our understanding of the underlying theoretical processes of TGfU. A signicant
criticism was that the theory lacked an explanation of the acquisition of appropriate movement and decision-making skills during games teaching. With regards to empirical work on
TGfU that have been shaped by information processing approaches (e.g., French et al.
1996; Turner 1996; Turner and Martinek 1999), Chow et al. (2007) noted that these
studies have provided some relevant insights into the acquisition of knowledge.
However, a perceived limitation was that they may not have provided an accurate picture
of how decision-making skills occur and develop in TGfU. More specically, work from
the information-processing approaches have tended to emphasize an examination of
learning at a micro-level, by focusing to a great extent on the construction of knowledge
by the learner without taking into consideration the important dynamic environmental interactions that occur in most learning programmes (Chow et al. 2007; Kelso 1995; Thelen

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

334

C.W.K. Tan et al.

1995; Van Gelder and Port 1995). Recent work using situated learning perspectives (e.g.,
McNeil et al. 2004; Rovegno, Nevett, Brock and Babiarz 2001; Wright et al. 2005) has permitted TGfU to be examined at a macro-level by taking into consideration the multitude of
physical, cognitive, social and environmental factors present in the learning environment.
While research from this perspective has provided a viable description of learning processes
by considering learner-environment interactions, there seemed to be a lack of information
on how learning or goal-directed behaviour could result from such interactions (for detailed
discussion see Chow et al. 2007).
Here we attempt to show how recent advances in the motor learning literature from an
ecological dynamics perspective might build on past theoretical analyses of TGfU, to contribute a suitable theoretical grounding to advance understanding of learning design in
TGfU by emphasising the role of learner environment interactions (Chow et al. 2007;
Chow and Tan 2009; Davids, Button, and Bennett 2008; Renshaw et al. 2010). Ecological
dynamics has provided relevant insights and understanding of learning processes in sport
and physical education (e.g., see Chow et al. 2007; Renshaw et al. 2010; Rink 2001;
Rovegno, Nevett, and Babiarz 2001; Chow et al. 2011). Within that framework, a constraints-led perspective has been proposed for understanding skill acquisition and game
play. In a recent issue of PESP, Renshaw et al. (2010) provided an overview of a constraints-led perspective and discussed how it might support a nonlinear pedagogical
approach, in which learners are considered to exemplify nonlinear dynamical systems, exhibiting key characteristics such as stability and instabilities in behaviour and the capacity to
exploit inherent self-organisation processes during learning. It was proposed that learning
design in physical education could be based on learners emergent capacity to uniquely
satisfy task, environment and performer constraints.
The current paper aims to draw the links between theoretical ideas underpinning Nonlinear Pedagogy and learning design in TGfU. To achieve this aim, it extends Renshaw
et al.s (2010) discussion to further understanding of the processes that might underpin
the four key pedagogical principles (sampling, tactical complexity, representation and
exaggeration) of TGfU in games teaching. Through analysis of the relationship between
theoretical ideas of nonlinear pedagogy and pedagogical principles in TGfU, we examine
insights on learning design in physical education with respect to games teaching. For
example, there have been attempts to apply the pedagogical principles to provide developmentally appropriate teaching materials and resources for curriculum development (e.g.,
progressions of activities, modied games, and frameworks for games) to enhance
student learning (e.g., Mandigo and Anderson 2003; Mitchell, Oslin, and Grifn 2003,
2006). Holt, Strean, and Bengoechea (2002) have argued that the four fundamental pedagogical principles, associated with TGfU, need to be more widely considered in the physical education literature, since there is generally a lack of empirical research to support the
viability of these key pedagogical principles in learning design. Indeed, more work on the
pedagogical principles needs to be undertaken in order to guide the practice of planning
games curriculum (Butler et al. 2008). We start by providing an overview of theoretical
ideas underpinning a nonlinear pedagogical approach and relevant empirical ndings
before discussing the relationship of key concepts with the pedagogical principles of TGfU.
A nonlinear pedagogical approach to teaching and learning in PE
Elsewhere the value of considering learners in physical education to exemplify nonlinear
dynamical systems in nature has been addressed (e.g., see Davids, Chow, and Shuttleworth
2005; Chow et al. 2007). These theoretical analyses revealed ubiquitous characteristics of

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

Physical Education and Sport Pedagogy

335

nonlinear dynamical systems, such as weather systems, global nancial markets, colonies
of ants and schools of sh, and neurobiological systems, including the capacity for periods
where such systems experience stability and instability, transitions from one pattern of behaviour to another when perturbed, and inherent self-adjustment processes under constraints (Chow et al. 2009). A key role in nonlinear pedagogy is played by the
constraints that each individual learner needs to satisfy during learning and performance
of games skills (see extensive work in physical education by Rovegno and colleagues
e.g., Rovegno 2006; Rovegno and Kirk 1995; Rovegno et al. 2001).
According to Newells constraints model (1986, 1996), learners in physical education
face a number of constraints which shape the acquisition of new coordination patterns
and movement behaviours (Newell 1996). These constraints can generally be classied
into three distinct categories: performer, environment and task constraints (Newell 1986).
Performer constraints refer to the specic structural and functional characteristics of
learners and include factors related to their physical, physiological, emotive and psychological disposition (Newell 1986). For example, learners tness levels and technical abilities, as well as psychological factors such as motivation to learn, may inuence how they
solve particular task problems or acquire unique movement behaviours that are individualspecic. Environmental constraints refer to physical factors such as visual and auditory
information surrounding the performer (e.g., the amount of light or level of noise in the
learning environment, ambient temperature and altitude or practice surfaces). Other
environmental constraints incorporate social inuences such as peer groups, societal and
cultural expectations in the community, family support and availability of facilities; these
factors are of particular relevance for young learners as their motor performance is often
strongly inuenced by the presence of critical group members such as the teacher or
peers. Task constraints comprise the goal of the task, rules of the activity, the learning
location, and the implements or equipment used during the learning experience. Task constraints are typically controlled by a teacher in a professional role related to learning design.
An important point is that constraints in these three key categories do not act independently of each other. Instead, the emergence of movement patterns in learners occurs
within an embodied framework where the performer, environment and task play signicant
roles in shaping the development of movement outcomes and the acquisition of skill
(Chow et al. 2007; Newell 1996; Renshaw et al. 2010). Specically, the dynamic interactions
among the conuence of constraints in the learning situation direct the learner to seek out
functional behaviours that are likely to achieve behavioural goals during learning. These
dynamic interactions are a key phenomenon of nonlinear systems and learners certainly
behave as complex neurobiological systems (Chow et al. 2009), as they search for stable
and functional movement solutions present among different possibilities for behaving
(Newell 1986).
In a nonlinear pedagogical approach, the ideas of self-adjustment under interacting constraints, the interwoven relationship between information and movement, as well as the role
of movement variability in adapting to changing task and environmental constraints, are relevant for our understanding of learning design in TGfU. We elucidate these key ideas in the
following sections, before exemplifying how they might explain the main pedagogical principles underpinning the TGfU approach.
Inherent information guides movement
From a nonlinear pedagogical perspective, knowledge about a movement or tactical
decision to be made is not constructed based on the existence of some internal

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

336

C.W.K. Tan et al.

representations located at higher levels of the human movement system (see Davids,
Button, and Bennett 2008). Instead, the importance of learner-environment interactions is
emphasised. With practice (i.e., game play or skill practice), a learner is able to increasingly
couple the information available in the practice environment to the actions required to
achieve a specic task goal.
An important factor to consider is the relationship between the information available in
specic performance contexts and learners actions. It has been argued that biological organisms, including humans, are surrounded by huge arrays of energy ows that can act as information sources to support goal-directed behaviour, including decision making, planning and
the organisation of actions. In order to access this information (e.g., through sight, sound,
touch), learners need to act. The relationship between information and action supports the recognition that humans must perceive in order to move, but must also move in order to perceive
(Gibson 1979). Learners can attune their movements to critical sources of information during
practice to establish strong information-movement couplings (i.e., they can form a strong
relationship between the information required for movement and movement needed to generate more information) to guide their behaviours.
This potential for information-movement coupling should inform learning design in physical education. Separating information and movement in order to help learners manage the
learning environment is a weakness in traditional teaching techniques. For example, in physical education, teaching the run-up separately from an action like long jumping or throwing a
javelin, or the tennis or volleyball serving action separately from the toss, is a fallacy which
encourages information-movement decoupling. Such a teaching approach is commonly
known as task decomposition. Instead, task simplication, where the difculty level of the
performing action is reduced while keeping the critical temporal-spatial relationship intact
should be encouraged (see Renshaw et al. 2010). The key point here is that task simplication, rather than decomposition, allows information to guide movement and in turn, movement produces the necessary information for further action.
Research in movement science has provided empirical support for the concept of information-movement coupling. For example, Bootsma and van Wieringen (1990) examined
participants executing the table-tennis forehand stroke and found that while the initiation of
the stroke was quite variable, the point of bat to ball contact was generally very consistent.
Continual guidance of action by perceptual information was suggested to be present as participants performed compensatory movement execution to allow for precise direction as the
arm moves towards the point of ball-bat contact. Similar support for the importance of
information and movement coupling in the generation and control of movement has also
been observed in volleyball serves (Temprado et al. 1997) and one-handed catching
(Button et al. 2002).
From the perspective of task decomposition, even the use of ball projection machines in
practising games skills such as batting in cricket has come under close scrutiny since
advance information from a bowlers actions is eliminated in such learning environments
(see Pinder, Renshaw, and Davids 2009). It needs to be understood that such learning
designs skew the nature of information-movement couplings formed by learners so that
they are predicated on ball ight information only, without reference to advance information (pre-ball release) from a bowlers actions (Pinder, Renshaw, and Davids 2009).
Shaping the development of movement outcomes through movement variability
In the development of movement outcomes related to games teaching, recent work suggests
that learning is more than merely observing a permanent behavioural change. It is also

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

Physical Education and Sport Pedagogy

337

about adaptive change or behaviour change in relation to a specic task goal (see Chow
et al. 2007; Renshaw et al. 2010). Although movement variability (often seen as a sign
of inconsistency in movement performance outcomes) has traditionally been viewed as
noise in the central nervous system, it is now clear that movement pattern variability
can actually play a functional role in helping learners adapt to dynamic performance
environments (Davids, Bennett, and Newell 2006; Riley and Turvey 2002).
A nonlinear pedagogical approach views movement variability as an intrinsic feature of
skilled motor performance since it provides the exibility required to adapt to complex
dynamic sport environments (Davids, Button, and Bennett 2008). From a pedagogical perspective, movement variability facilitates free exploration in specic performance contexts
(see Renshaw et al. 2010). While a movement itself may not be repeated in an identical
manner, it is important for learners to be allowed to explore different ways to achieve a
similar outcome.
For example, research has shown that it is impossible to kick a ball in exactly the same
way with every attempt since there will always be minute differences in limb movement
speeds or joint angle motion, although the performance outcome could be the same (i.e.,
accurately hitting a pre-assigned target) (see Chow et al. 2006). Even among expert
javelin throwers, Schollhorn (1998) found that there were no common movement pattern
and that variability in movement was inherent, showing that it was impossible for even
elite performers to repeat exactly the same movement between throwing trials. This
feature of repetition without repetition (Bernstein 1967) in human movement systems
actually provides learners with the capacity to invent novel ways to solve typical performance problems. Importantly, variability in movement is critical to allow learners to explore
and make changes to their movement behaviours to progressively meet their assigned task
goals in learning situations.
Some recent work in the motor learning literature has shown that movement variability
is a critical phenomenon for change in movement behaviours to occur. For example, Chow,
Davids, Button, and Rein (2008) reported how adult male novices in a soccer kicking task
demonstrated high levels of kicking pattern variability between the transition from one preferred kicking pattern to a new preferred kicking pattern over four weeks of practice (570
kicking trials per learner). It was also determined that the least successful learners demonstrated the least amount of kicking pattern variability and showed very little exploration in
kicking patterns during the practice phase.
In relation to team game performance contexts, Araujo and colleagues (2004) found that
in a 1 v 1 basketball dribbling game, successful dribblers demonstrated increased movement variability nearer to the defender when they effectively dribbled past the defender
towards the basket. Attackers with relatively small amounts of movement variability exhibited less success in the 1 v 1 task. In a recent study, Schollhorn et al. (2009) discussed ideas
relating to how variability can be incorporated into skill practices. Particularly, it was
pointed out that different practice conditions of variability can lead to different amounts
of learning. Ideas like variable practice, where different variations of the same skill can
be presented (e.g., serving to the right, centre and left in tennis) can result in higher contextual interference (i.e., interference that occurs as a consequence of performing different
tasks or skills within a practice session [Battig 1979]) as compared to, for example, constant
practice schedules. Notably, greater interference could have a benecial learning effect
during retention and transfer contexts (Brady 1998). Interestingly, Schollhorn et al.
(2009) emphasised the need for differentiated contextual interference for different learners.
More specically, they suggested that children or novice learners should be exposed to
practice schedules with lower levels of contextual interference while adults or more

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

338

C.W.K. Tan et al.

skilled learners should be presented with practice schedules with higher contextual interference. So, even though variability may be functional, the amount of variability designed into
practice conditions needs to be adjusted to suit different learners, possibly at various learning stages.
Newells model of motor learning (1985) identied that various stages of learning
(Coordination, Control and Skill stage) could be a suitable platform to explore how teaching
of motor skills can be adjusted for learners of various skill levels. This model of learning,
which has been closely associated with the nonlinear pedagogical approach (see Chow et al.
2007; Renshaw et al. 2010), will be further elucidated in subsequent sections to highlight its
relevance to the pedagogical principles underpinning the TGfU approach.
In summary, recent research suggests that educators need to view movement variability
as an integral process in learning and acquiring effective movement patterns specic to a
task goal. A nonlinear pedagogical approach to teaching games skills provides learners
with opportunities to explore and nd movement solutions within a set of specic constraints (especially task constraint) designed by physical educators. The key role of physical
educators, however, is not to simply allow freedom of play during lessons, but rather to
carefully manipulate task constraints through rigorous learning design to guide learners in
adapting their movements to overcome specic movement challenges in modied games.
A nonlinear pedagogy and learning design in TGfU
As discussed earlier, key ideas in nonlinear pedagogy can offer new conceptual insights into
understanding learning design issues in TGfU (Chow et al. 2007). Similar to a situated
learning perspective, nonlinear pedagogy adequately captures the rich range of diverse constraints in skills learning and participation in games present in physical education. This
approach emphasises the nature of the learner-environment interactions through the use
of modied games to better situate learning within the context of the learning environment.
In this section we outline how a nonlinear pedagogy can provide insights on learning
designs incorporating the four main pedagogical principles of the TGfU approach (i.e.,
sampling, task complexity, representation and exaggeration), providing some rationale
for the efcacy of the model.
Sampling
The principle of sampling is based on the premise that games selected for learning should
offer a variety of experiences with possibilities to show similarities and differences between
apparently dissimilar and similar games, respectively (Thorpe, Bunker, and Almond 1984;
Thorpe and Bunker 1989). Systems of games classication (e.g., Almond 1986; Ellis 1983)
have been devised to illustrate where the similarities lie. These systems have proved valuable in facilitating the integration of games with similar tactical possibilities within the
sampling procedure (Thorpe, Bunker, and Almond 1984; Thorpe and Bunker 1989).
Specically, games within the same category have common tactical elements and
employ strikingly similar strategies to achieve similar goals. These elements or strategies,
when understood by students, can be transferred from one game to another, within the same
games category. To provide students with a deeper tactical learning and improved game performance, Mitchell, Oslin, and Grifn (2006) added that teachers should select from within
rather than across games categories, by identifying similarities among games within each
category. For example, football and eld hockey, both invasion/territorial games, share
similar characteristics even though the specic rules and equipment used in both games

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

Physical Education and Sport Pedagogy

339

differ. Students who understand tactics related to the principle of advancement into the
opposing teams playing area to score a goal in football, for example, are likely to be
able to transfer that understanding to eld hockey game play. As such, sampling from
different types of games can expose students to a variety of game forms and experiences
to help them learn to transfer their learning from one game to another and gain greater
understanding of game play in general (Thorpe et al. 1984; Thorpe and Bunker 1989).
The principle of sampling has received support from research literature on transfer of
learning. Findings from different educational areas (e.g., literacy, technology, motor skill
acquisition) as well as anecdotal evidence from practitioners support the views that understanding of game characteristics can transfer positively between tactically similar games
(e.g., Beard 1993; Dan Ota and Vickers 1998; McAloon 1994). However, as noted by
Mitchell, Oslin, and Grifn (2006), there are only a few studies that have investigated
and supported the transfer of understanding in physical activity (e.g., Martin 2004; Mitchell
and Oslin 1999).
In nonlinear pedagogy the principle of sampling can be explained with respect to the
relationship between the dynamics of a specic task (to be learned) and the individual learners existing intrinsic dynamics. Intrinsic dynamics can be broadly dened as the current
dispositions for performance in a specic physical activity in each individual that are
shaped by important constraints such as genes, developmental experiences and learning
(Davids, Button, and Bennett 2008). For example, good games players need functional
genetic propensity (e.g., for power or speed endurance), appropriate developmental experiences during growth and maturation and task-specic learning in a particular sport. These
constraints shape the intrinsic dynamics of an individual so that they become a stable part of
an individuals performance capacity.
Where the task dynamics for two games are similar (e.g., futsal and football or rugby
league and rugby union), they will lead to cooperation with the existing intrinsic dynamics
of an individual learner seeking to transfer between the games. Positive transfer can occur
because the intrinsic dynamics of the learner can support learning and performance in both
games. Where the task dynamics of a to-be-learned game (e.g., badminton) differ from the
intrinsic dynamics of a learner (e.g., who has learned to play tennis to a high level of performance), there will be competition between the two sets of dynamics. The nature of the
relationship between task dynamics and intrinsic dynamics of learners captures the potential
for transfer of movement behaviours across games. When critical elements of relevant task
constraints (e.g., rules of the game, use of equipment and playing area) within two games
are similar (e.g., invasion games such as eld hockey and football), a stable platform is provided for movement behaviours to be positively transferred from one game to another.
In this way, the pedagogical principle of sampling in TGfU is relevant and effective in
learning design for exposing students to similar tactical characteristics within different
games to lead learners to harness their existing intrinsic dynamics to facilitate successful
learning and performance. Such ideas are congruent with the suggestion by Mitchell,
Oslin, and Grifn (2006) that, in planning programme design in games teaching, teachers
should select from within rather than across games categories to support deeper tactical
learning and improved game performance. Practitioners need to sample key common features within the different categories based on the games classication systems, so that positive transfer of learning can occur from one game to another as learners harness their
existing system intrinsic dynamics to positive effect.
Another useful example of sampling task dynamics in learning design to harness existing
system dynamics occurs in the acquisition of some techniques in different game categories
to support transfer across these games categories. For example, the overhead throwing action

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

340

C.W.K. Tan et al.

can form the basis of movement patterns for performing various specic technical skills in
different game categories. If the intrinsic dynamics of an individual are stable with reference
to the overhead throwing action, it can be used as a basis for learning other related technical
movements such as the volleyball overhead serve, the tennis serve and the badminton overhead smash. However, the specic execution of these different movements will still need to be
further adjusted and rened for the specic performance contexts, with the overhead throwing
action as the foundational movement pattern. It has been highlighted that the closer the controlled environment is to the context of the game, the greater the likelihood of effective transfer taking place (Rink, French, and Graham 1996). Nevertheless, substantial empirical
evidence needs to be collected to determine the extent of transfer that can occur across
game categories (Rink, French, and Graham 1996).
Tactical complexity
The pedagogical principle of complexity in TGfU involves designing and matching the
game forms to the developmental level of the student, so that the tactical problems presented may not be too complex for the learners to understand. As highlighted by Thorpe,
Bunker, and Almond (1984) and Thorpe and Bunker (1989), it is logical to start with
simpler games if our objective is to ensure that children or novice players understand the
games they play; that is, games with less tactical complexity should be taught rst before
games with greater complexity.
Designing a progression of game forms with increasing tactical complexity can occur at
two levels, within and across the game categories. Within the game categories, Mitchell
et al. (2003, 2006) proposed levels of game complexity (i.e., levels I, II and III) with
increasing tactical complexities. Complexity can also be perceived across categories. Of
the four game categories, target games considered less complex, are usually recommended
to be taught rst, followed by net/ barrier or elding games, and then invasion/ territorial
games (e.g., Thorpe, Bunker, and Almond 1984; Thorpe and Bunker 1989; Werner, Thorpe
and Bunker 1996).
The sequence of selecting to teach less complex games rst (both within and across the
games categories) is relevant to the idea of accommodating task complexity to different skill
levels and constitutes a key aspect of learning design in physical education. Newells (1985)
model of learning, mentioned earlier in the discussion on variability, has application here.
The model proposed that there are three stages of learning: Coordination, Control and Skill
stages. In the early stage of learning (Coordination stage), a movement skill challenges a
learner to assemble an approximate movement behaviour to the target pattern. The
learner seeks to harness the existing individual (intrinsic dynamics) to nd stable and preferred movement solutions to specic motor problems. The successful search for a functional coordination pattern allows performance of the task to a basic level, as the learner
assembles component relations between relevant parts of the body. Because of the
pattern assembly challenge, at the Coordination stage, it may be more useful to introduce
less complex games involving a smaller number of task and environmental constraints
like target games or net/barrier games. For example, less complex target games (e.g.,
archery, dart throwing) which require simpler movement patterns involving fewer limb segments, are less demanding for individuals to assemble. Such simpler movements could be
more functional at the early stage of learning to produce goal-directed behaviour.
Performers at the next stage of learning, the Control stage, can exibly adapt a stable
movement pattern to approximately t changing performance environments (Newell
1985). Chow, Davids, Button, and Koh (2008) observed how skilled soccer players were

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

Physical Education and Sport Pedagogy

341

able to effectively vary foot speed at ball contact to chip a ball towards different target
positions. A higher foot speed at ball contact was used when the target was located
further away. However, novices were not able to vary foot speed at ball contact functionally
when target positions were altered during kicking trials. A plausible explanation could be
that skilled players were able to use reactive forces in the different limb segments during the
kicking action to allow an effective kicking pattern to emerge compared to learners at the
Coordination stage where the aim is to acquire a basic movement pattern. This view
suggests that, at the Control stage, it is important to organise more complex games
involving elding (e.g., softball or cricket) or invasion games (e.g., basketball or netball)
where learners are required to execute multi-articular movements involving many limb
segments and are faced with increased amounts of information to manage the numerous
movement possibilities.
At the Skill stage of learning, more complex manipulations of task constraints can be
incorporated to raise the difculty of game play to help learners rene and demonstrate
already well-established skills in varied game situations. For example, the size of play
areas could be reduced to increase the time-space demands in small-sided games, challenging skilled performers to raise their perceptual-action attunement for successful task completion. Certainly, invasion games could provide ample opportunities to effectively engage
learners at the Skill stage of learning.
Increasing task complexity can be achieved by manipulating task constraints within the
same game category or the same sport. Different learners may progress at different rates in
acquiring game tactical knowledge/technical skills. Physical educators need to adjust the
complexity of the learning task appropriately to adequately challenge learners to achieve
success. At the Skill stage of learning, learners can vary a movement pattern in an
energy-efcient manner to t changing circumstances in dynamic environments (Davids,
Button, and Bennett 2008). Physical educators teaching a class of skilled learners need
to provide situated games with higher levels of complexity in terms of concepts and technical skills to suitably challenge these learners. Advance tactical decisions and knowledge
have to be presented in constrained games to provide opportunities for skilled learners to
explore movement solutions in those contexts. Useful constraints to be added to these
small-sided games include changes to space, target areas, equipment, player numbers
involved and introduction of articial rules. For example, the teacher could limit the
number of ball touches per player or time spent in an attacking zone.
Representation
Representation involves developing modied, mini-games that contain the same tactical
structure of the adult game (Thorpe and Bunker 1989; Thorpe, Bunker and Almond
1984). The premise of its effectiveness is that games are developed and modied to the
extent where the tactical intricacies of the adult game are retained but simplied and
adapted to suit the learners (e.g., children or novices) size, age, and ability for successful
participation (Thorpe and Bunker 1989; Thorpe, Bunker, and Almond 1984). The aim of
representation is for learners to experience opportunities for developing tactical awareness,
making appropriate decisions, and practicing skills in manageable practice environments.
In a nonlinear pedagogy, the relevant use of rules and equipment can provide a learning
situation that is appropriate for different stages of learning (Handford et al. 1997). For
example, in net games such as volleyball, an adult 6 v 6 version of the game may not be
as valuable a learning experience for many less skilled learners. They typically do not
have the positional awareness or skills to benet from this representation of the full game.

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

342

C.W.K. Tan et al.

However, a mini, simplied 3 v 3 game, perhaps initially allowing an additional

bounce between volley passes by the same team, is better suited to the lower skill levels of
the beginning learners, while maintaining the tactical structure of the adult form of the game.
The key point about the pedagogical principle of representation is to keep the information-movement coupling of the structured game or practice relevant such that it is still
representative of the actual game. This goal can be achieved through the process of task
simplication and physical educators should avoid the error of task decomposition in
seeking to enhance representation (Davids, Button, and Bennett 2008).
Undoubtedly, teachers or coaches should modify games that provide learners with
access to the perceptual information available in the performance context which can be
closely coupled with relevant actions. If the relationship between information and movement is weak after a task modication, it becomes challenging for teachers and students
to nd associations between the modied game and the full game form represented in
the TGfU approach. For example, if the use of a modied racket distorts the movement
information so that the learner adopts a highly unique stance or movement pattern for a
tennis forehand drive, then the modication is not representative and risks becoming less
relevant for the learner when transferred to an actual tennis game (where other variations
of the drive is required). In essence, representation needs to be built on a close relationship
between the dynamics of a modied task and the task dynamics involved in a full game. If
this close t is not maintained, the dynamics of the practice task and required performance
task may compete, rather than cooperate during learning. Nevertheless, more empirical
research in relation to transfer across representative game contexts needs to be undertaken
to properly determine its relevance for learning.
Additionally, the dynamics of a game are uid and variable in actual game contexts. As
such, practice task constraints should aim to mimic the actual performance environments as
much as possible. Traditionally, the focus of many coaches is to provide repetitive drills
during practice tasks to train athletes/ players to be as consistent as possible, repeating
a movement continually but devoid of how the skill might be functionally adapted for performance in dynamic game settings. Such drills tend to be too static and generally lack representation of how real game situations could be actualised. In a nonlinear pedagogy, the
infusion of variability requires learners to experience different movement parameters
during the learning process all the time. This principle forces the learner to explore a
larger performance solution space which will actually be more relevant and effective
when required to satisfy the ecological constraints of a real game context. As mentioned
earlier, the role of variability in modied games is to enhance learners exibility and adaptability, which are critical in satisfying performance constraints of real games.
Exaggeration
Closely related to the pedagogical principle of representation, Thorpe and Bunker (1989)
and Thorpe, Bunker, and Almond (1984) highlighted that mini-games developed should
also pose tactical problems that can be solved by children. For this process to occur,
games should maintain the primary rules of the full game, although elements of the
games can be modied to exaggerate a tactical idea to be explored. In this respect, exaggeration involves frequent manipulation of task constraints, a key strategic aim in a nonlinear
pedagogy, to provide relevant learning environments. This TGfU principle involves changing the rules of the game to overstate or emphasise a specic tactical problem by making
the to-be-taught concept obvious to the learners (albeit implicitly), thus constraining them
towards learning about a particular pattern of play. A well-established example of

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

Physical Education and Sport Pedagogy

343

exaggeration in net games is the use of a long and narrow court in practice; here, the shape
of the court makes it obvious for learners to direct their shots to the front and back of the
court. The perceived information from the task constraints (long narrow court), together
with the intentions of the performer, will accentuate the role of high clears and drop
shots in play. Exaggeration through task manipulation can also come in the form of
game rules. For example, in a modied game of football where more points are awarded
for consecutive passes than actually scoring a goal, the team in possession would be encouraged to maintain control of the ball. Here, behavioural pattern of keeping possession is precipitated through simple adjustment to game rules.
In a nonlinear pedagogy, a functional behaviour to be acquired (e.g., hitting long or
short, or maintaining ball possession) emerges based on the task constraints designed in
the modied game setting. The focus of this perspective highlights the role of manipulating
task constraints in learning design to exaggerate a specic tactical play to encourage and
shape learners behaviours without explicit instructions being involved. This process
avoids the over-use of verbal information which encourages explicit learning in favour
of the subtle manipulation of task constraints to encourage implicit learning. Conceptually,
the theoretical rationale in nonlinear pedagogy for this process involves the construction of
appropriate boundaries through the manipulation of task constraints. This pedagogical
activity facilitates learners to search within a narrower area in the perceptual motor workspace where functional movement solutions for task problems can be found and established.
This constrained approach also contrasts with the idea of free play where, a lack of exaggeration will warrant a blind search for movement solutions by the learners, resulting in
more time needed and enhanced health and safety risks.
Moreover, exaggeration in situational games can direct learners to the relevant information-movement relationships that are critical for successful game performance. This
point was exemplied by Chow, Davids, Button, and Koh (2008) who found that learners
can explore and discover individualised kicking patterns when challenged to kick a ball
over a height barrier in the absence of explicit instructions on kicking technique. The
height of the bar, distance of the target in the kicking task and the video presentation of
the expected outcome (i.e., learners were shown the accurate landing of the ball in an appropriately weighted manner only) all exaggerated the relevant information-movement
relationship to assist learners in their search for functional kicking solutions. Without exaggeration, learners become less attuned to the key environmental parameters that are pertinent for the generation of functional movement solutions. This aspect of learning design
is especially critical at the Control stage of learning (see Newell 1985) where learners
need to be exposed to more exploratory search in learning to nd associations between
their movement patterns and the performance context.
Learners exploration of functional movement solutions could occur where high levels
of movement variability are encouraged. Functional movement variability, as mentioned
previously, is a positive opportunity for learners to determine and differentiate between successful and less successful movement solutions in these game contexts. With the principle
of exaggeration, more opportunities are provided for learners to search for functional movement solutions within a controlled practice scenario where the tactical concepts to be
taught are clearly exaggerated.
Conclusion: future research and implications for practitioners
In this paper we have proposed how a nonlinear pedagogy can offer new conceptual
insights and could provide a theoretical rationale for explaining the operational principles

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

344

C.W.K. Tan et al.

of TGfU in learning design of games teaching. To achieve this aim, we examined the viability of the four key pedagogical principles of the TGfU model. We highlighted how the
theoretical and practical implications of nonlinear pedagogy, considered with some empirical evidence from the recent motor learning literature, to provide an understanding of how
TGfU might function. Our analysis originally stemmed from the need to provide an insight
to Thorpe and Bunkers (1986) question: Does TGfU work? Following that, the discussion on how does TGfU work? provided analysis of the pedagogical principles that underpin TGfU, with an emphasis on how they might be supported by a nonlinear pedagogical
perspective.
From a research standpoint, it is clear that there is a need for further empirical studies to
validate the proposal of a nonlinear pedagogy framework as a suitable theoretical grounding
to support learning design in TGfU. Considerations for future research include empirical
studies focusing on the role of small-sided games, looking at different categories of
games, conducting school-based learning studies, examining performance outcomes as
well as affective parameters, examining the learning design of TGfU beyond the four pedagogical principles, and also examining individual changes (that is, intra-individual analysis coupled with group analysis).
Additionally, some specic questions that might guide future research in TGfU from a
nonlinear pedagogical perspective include: What is the relative effectiveness of teacher guidance that focuses on the achievement of performance outcomes rather than on the production of a specic movement pattern? What are the affective (e.g., motivation) and
physical (e.g., activity level) consequences of this approach to teaching game skills? Is a
nonlinear pedagogy relevant for promoting enjoyment and learning games skills in
groups of low-skilled individuals? Does movement pattern variability emerge prior to the
acquisition of new preferred movement patterns, rather like uctuations prior to a phase
transition in a complex system?
The ideas raised in this paper could also inform practitioners to more effectively engage
students using the TGfU approach. Application of a nonlinear pedagogical perspective to
further our understanding of the possible processes underpinning the four key pedagogical
principles (sampling, tactical complexity, representation and exaggeration) of TGfU in
games teaching has several key practical implications for physical educators. First, physical
educators can lead learners to harness existing system intrinsic dynamics to facilitate learning and performance by exposing students to tasks with similar tactical characteristics (e.g.,
running into space or providing options for a pass to keep possession) within different
games in the same category (e.g., invasion games such as eld hockey and football). By
sampling key common features within specic categories, based on the games classication
systems, positive transfer of learning can occur from one game to another. Second, physical
educators need to plan for developmentally appropriate tasks by adjusting the complexity of
the task to adequately challenge learners to achieve success. This can be done by manipulating task constraints (e.g., changes to space, target areas, equipment, and number of
players involved) according to the learning stages of learners, as exemplied by
Newells (1985) model of motor learning. Third, physical educators should maintain the
functional information-movement couplings of the structured game or practice task such
that it is still representative of the actual performance environment (for example, a simplied 1v1 tennis game, using a lighter, shorter racket, a sponge ball, and a low net instead of
an adult 1v1 game). Practice task constraints should aim to mimic the actual performance
environment as much as possible by having learners experience variability (that is, different movement parameters) during the learning process all the time. Finally, physical educators could develop mini-games that are not only representative of the actual

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

Physical Education and Sport Pedagogy

345

performance environment, but should also pose tactical problems that appropriately challenge learners to explore functional solutions. Again, this aim can be achieved by subtle
manipulation of task constraints (e.g., by changing the rules of the game) to make the
to-be-taught concept obvious to the learners so as to encourage implicit learning about
a particular pattern of play (e.g., using a long narrow court for net games to accentuate
hitting long or short). In conclusion, attempting to understand the learning design of
TGfU from a nonlinear pedagogy approach might provide some valuable insights into
the underlying processes of the TGfU pedagogical principles. This research task might
allow researchers and physical educators to more effectively engage students and
provide developmentally appropriate teaching materials and resources for curriculum
development.

References
Almond, L. 1986. Reecting on themes: A games classication. In Rethinking games teaching, ed.
R.D. Thorpe, D.J. Bunker and L. Almond, 71 72. Loughborough, UK: Loughborough
University.
Araujo, D., K. Davids, S. Bennett, C. Button, and G. Chapman. 2004. Emergence of sport skills under
constraint. In Skill acquisition in sport: Research, theory and practice, ed. A.M. Williams and
N.J. Hodges, 409 33. London: Routledge.
Battig, W.F. 1979. The exibility of human memory. In Levels of processing in human memory, ed.
L.S. Cermak and F.I.M. Craik, 2344. Hillsdale, NJ: Erlbaum.
Beard, C.H. 1993. Transfer of computer skills from introductory computer courses. Journal of
Research in Computing in Education 25: 413 30.
Berkowitz, R.J. 1996. A practitioners journey: From skill to tactics. Journal of Physical Education,
Recreation and Dance 67: 44 45.
Bernstein, N.A. 1967. The control and regulation of movements. London: Pergamon Press.
Booth, K. 1983. An introduction to netball: An alternative approach. Bulletin of Physical Education
19: 27 31.
Bootsma, R.J., and P.W.C. van Wieringen. 1990. Timing and attacking forehand drive in table tennis.
Journal of Experimental Psychology: Human Perception 16: 21 29.
Brady, F. 1998. A theoretical and empirical review of the contextual interference effect and the learning of motor skills. Quest 50: 266 93.
Bunker, D., and R. Thorpe. 1982. A model for the teaching of games in secondary schools. Bulletin of
Physical Education 18: 58.
Burrows, L. 1986. A teachers reactions. In Rethinking games teaching, ed. R.D. Thorpe, D.J. Bunker
and L. Almond, 4552. Loughborough, UK: Loughborough University.
Butler, J., L. Grifn, and R. Nastasi, eds. 2003. Teaching games for understanding in physical education and sport. Reston, VA: National Association for Sport and Physical Education.
Butler, J., J. Oslin, S. Mitchell, and L. Grifn. 2008. The way forward for TGfU: Filling the chasm
between theory and practice. Physical and Health Education 74: 6 12.
Button, C., K. Davids, S.J. Bennett, and G.J.P. Savelsbergh. 2002. Anticipatory responses to perturbation of co-ordination in one-handed catching. Acta Psychologica 109: 75 93.
Chow, J.Y., K. Davids, C. Button, and M. Koh. 2006. Organization of motor system degrees of
freedom during the soccer chip: An analysis of skilled performance. International Journal of
Sport Psychology 37: 207 29.
Chow, J.Y., K. Davids, C. Button, and M. Koh. 2008. Coordination changes in a discrete multi-articular action as a function of practice. Acta Psychologica 127: 163 76.
Chow, J.Y., K. Davids, C. Button, and R. Rein. 2008. Dynamics of movement patterning in learning a
discrete multiarticular action. Motor Control 12: 21940.
Chow, J.Y., K. Davids, C. Button, R. Rein, R. Hristovski, and M. Koh. 2009. Dynamics of multiarticular coordination in neurobiological systems. Nonlinear Dynamics, Psychology and Life
Sciences 13: 2752.
Chow, J.Y., K. Davids, C. Button, R. Shuttleworth, I. Renshaw, and D. Araujo. 2007. The role of nonlinear pedagogy in physical education. Review of Educational Research 77: 25178.

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

346

C.W.K. Tan et al.

Chow, J.Y., K. Davids, R. Hristovski, D. Araujo, and P. Passos. 2011. Nonlinear Pedagogy: Learning
design for self-organizing neurobiological systems. New Ideas in Psychology 29, no. 2: 189 200.
Chow, J.Y., and C. Tan. 2009. Enhancing game play in PE and sport: A holistic approach. In
Perspectives on physical education and sports science in Singapore: An eye on the Youth
Olympic Games 2010 ed. N. Aplin, 106 17. Singapore: McGraw Hill.
Dan Ota, K., and J.N. Vickers. 1998. The effects of variable practice on the retention and transfer of
two volleyball skills in male club-level athletes. Journal of Sport and Exercise Psychology,
NASPSA Abstracts, Suppl. no. 20: S121.
Davids, K., S.J. Bennett, and K. Newell, eds. 2006. Movement system variability. Champaign, IL:
Human Kinetics.
Davids, K., C. Button, and S.J. Bennett. 2008. Coordination and control of movement in sport: An
ecological approach. Champaign, IL: Human Kinetics.
Davids, K., J.Y. Chow, and R. Shuttleworth. 2005. A constraints-based framework for nonlinear pedagogy in physical education. Journal of Physical Education, New Zealand 38: 1729.
Ellis, M. 1983. Similarities and differences in games: A system for classication. In Proceedings of
the AIESEP International Congress in Teaching Team Sports, 13742. Rome, Italy: COMI-Scula
dello Sport.
French, K.E., P.H. Werner, K. Taylor, K. Hussey, and J. Jones. 1996. The effects of a 6-week unit of
tactical, skill, or combined tactical and skill instruction on badminton performance of ninth-grade
students. Journal of Teaching in Physical Education 15: 43963.
Gibson, J.J. 1979. An ecological approach to visual perception. Boston, MA: Houghton-Mifin.
Grifn, L.L., R. Brooker, and K. Patton. 2005. Working towards legitimacy: Two decades of Teaching
Games for Understanding. Physical Education and Sport Pedagogy 10: 21323.
Grifn, L.L., S.A. Mitchell, and J.L. Oslin. 1997. Teaching sport concepts and skills: A tactical games
approach. Champaign, IL: Human Kinetics.
Grifn, L.L., and K. Patton. 2005. Two decades of teaching games for understanding: Looking at the
past, present, and future. In Teaching games for understanding: Theory, research and practice,
ed. L. Grifn and J. Butler, 117. Champaign, IL: Human Kinetics.
Handford, C., K. Davids, S. Bennett, and C. Button. 1997. Skill acquisition in sport: Some applications of an evolving practice ecology. Journal of Sports Sciences 15: 621 40.
Holt, N.L., W.B. Strean, and E.G. Bengoechea. 2002. Expanding the teaching games for understanding model: New avenues for future research and practice. Journal of Teaching in Physical
Education 21: 162 76.
Kelso, J.A.S. 1995. Dynamic patterns: The self-organization of brain and behaviour. Cambridge,
MA: MIT.
Kirk, D., and A. MacPhail. 2002. Teaching games for understanding and situated learning: Rethinking
the Bunker-Thorpe model. Journal of Teaching in Physical Education 21: 177 92.
Lee, A.M. 2003. How the eld evolved. In Student learning in physical education: Applying research
to enhance instruction, ed. S.J. Silverman and C.D. Ennis, 925. Champaign, IL: Human
Kinetics.
Mandigo, J.L., and A.T. Anderson. 2003. Using pedagogical principles in net/wall games to enhance
teaching effectiveness. Teaching Elementary Physical Education 14: 811.
Martin, R.J. 2004. An investigation of tactical transfer in invasion games. Poster presented at the
annual convention of the American Alliance for Health, Physical Education, Recreation and
Dance, in New Orleans.
McAloon, N.M. 1994. Connections (from the teachers desk). Journal of Reading 37: 698 9.
McNeill, M.C., J.M. Fry, S.C. Wright, W.K.C. Tan, K.S.S. Tan, and P.G. Schempp. 2004. In the local
context: Singaporean challenges to teaching games on practicum. Sport, Education and Society
9: 332.
Mitchell, S. 2003. Teaching net/wall games through tactical models. Teaching Elementary Physical
Education 14: 6 7.
Mitchell, S.A., L.L. Grifn, and J.L. Oslin. 1995. The effects of two instructional approaches on game
performance. Pedagogy in Practice: Teaching and coaching in physical education and sports 1:
3648.
Mitchell, S.A., and J.L. Oslin. 1999. An investigation of tactical understanding in net games.
European Journal of Physical Education 4: 162 72.
Mitchell, S.A., J.L. Oslin, and L.L. Grifn. 2003. Sport foundations for elementary physical education: A tactical games approach. Champaign, IL: Human Kinetics.

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

Physical Education and Sport Pedagogy

347

Mitchell, S.A., J.L. Oslin, and L.L. Grifn. 2006. Teaching sport concepts and skills: A tactical games
approach. 2nd ed. Champaign, IL: Human Kinetics.
Newell, K.M. 1985. Coordination, control and skill. In Differing perspectives in motor learning,
memory, and control, ed. D. Goodman, I. Franks and R.B. Wilberg, 295317. Amsterdam:
North-Holland.
Newell, K.M. 1986. Constraints on the development of coordination. In Motor development in children. Aspects of coordination and control, ed. M.G. Wade and H.T.A. Whiting, 34160.
Dordrecht, Netherlands: Martinus Nijhoff.
Newell, K.M. 1996. Change in movement and skill: Learning, retention and transfer. In Dexterity and
its development, ed. M.L. Latash and M.T. Turvey, 393430. Mahwah, NJ: Erlbaum.
Pinder, R.A., I. Renshaw, and K. Davids. 2009. Informationmovement coupling in
developing cricketers under changing ecological practice constraints. Human Movement
Science 28: 468 79.
Renshaw, I., J.Y. Chow, K. Davids, and J. Hammond. 2010. A constraints-led perspective to understanding skill acquisition and game play: A basis for integration of motor learning theory and
physical education praxis? Physical Education and Sport Pedagogy 15: 117 37.
Richard, J., and N. Wallian. 2005. Emphasizing student engagement in the construction of game performance. In Teaching games for understanding: Theory, research and practice, ed. L. Grifn and
J. Butler, 19 32. Champaign, IL: Human Kinetics.
Riley, M.A., and M.T. Turvey. 2002. Variability and determinism in motor behavior. Journal of Motor
Behavior 34: 99 125.
Rink, J.E. 2001. Investigating the assumptions of pedagogy. Journal of Teaching in Physical
Education 20: 11228.
Rink, J.E., K.E. French, and K.C. Graham. 1996. Implications for practice and research. Journal of
Teaching in Physical Education 15: 490 502.
Rovegno, I. 2006. Situated perspectives on learning. In The handbook of physical education, ed. D.
Kirk, D. Macdonald and M. OSullivan, 262 74. London: Sage.
Rovegno, I., and D. Kirk. 1995. Articulations and silences in socially critical work on physical education: Toward a broader agenda. Quest 47: 44774.
Rovegno, I., M. Nevett, and M. Babiarz. 2001. Learning and teaching invasion-game tactics in 4th
grade: Introduction and theoretical perspective. Journal of Teaching in Physical Education 20:
341 51.
Rovegno, I., M. Nevett, S. Brock, and M. Babiarz. 2001. Teaching and learning basic invasion-game
tactics in 4th grade: A descriptive study from situated and constraints theoretical perspectives.
Journal of Teaching in Physical Education 20: 37088.
Schollhorn, W.I. 1998. Systemdynamische betrachtung komplexer bewegungsmuster im lernprozess
[Dynamical systems view of complex patterns of movement in the learning process]. Frankfurt
am Main: Lang.
Schollhorn, W.I., G.M. Mayer-Kress, K.M. Newell, and M. Michelbrink. 2009. Time scales of adaptive behavior and motor learning in the presence of stochastic perturbations. Human Movement
Science 28, no. 3: 319 33.
Temprado, J., M. Della-Grasta, M. Farrell, and M. Laurent. 1997. A novice-expert comparison
of (intra-limb) coordination subserving the volleyball serve. Human Movement Science 16:
653 76.
Thelen, E. 1995. Motor development. A new synthesis. American Psychologist 50: 7995.
Thorpe, R., and D. Bunker. 1986. Is there a need to reect on our games teaching? In Rethinking
games teaching, ed. R. Thorpe, D. Bunker and L. Almond, 25 34. Loughborough, UK:
Loughborough University.
Thorpe, R., and D. Bunker. 1989. A changing focus in games teaching. In The place of physical education in schools. ed. L. Almond, 4271. London: Kogan Page.
Thorpe, R.D., D.J. Bunker, and L. Almond. 1984. A change in focus for the teaching of games. In
Sport pedagogy: Olympic Scientic Congress proceedings, ed. M. Pieron and G. Graham,
163 69. Champaign, IL: Human Kinetics.
Turner, A.P. 1996. Teaching for understanding: Myth or reality? Journal of Physical Education,
Recreation and Dance 67: 46 48.
Turner, A.P., and T.J. Martinek. 1992. A comparative analysis of two models for teaching games:
Technique approach and game-centred (tactical focus) approach. International Journal of
Physical Education 29: 15 31.

348

C.W.K. Tan et al.

Downloaded by [Universiti Pendidikan Sultan Idris] at 00:38 29 September 2015

Turner, A.P., and T.J. Martinek. 1999. An investigation into teaching games for understanding: Effects
on skill, knowledge and game play. Research Quarterly for Exercise and Sport 70: 286 96.
Van Gelder, T., and R.F. Port. 1995. Explorations in the dynamics of cognition. Mind as motion.
London: MIT Press.
Werner, P., R. Thorpe, and D. Bunker. 1996. Teaching games for understanding: Evolution of a model.
Journal of Physical Education, Recreation and Dance 67: 28 33.
Wright, S., M. McNeil, J. Fry, and J. Wang. 2005. Teaching teachers to play and teach games. Physical
Education and Sport Pedagogy 10: 61 82.