Improving police training from
a cognitive load perspective
Rebecca Mugford, Shevaun Corey and Craig Bennell
Department of Psychology, Carleton University, Ottawa, Canada
Purpose – The purpose of this paper is to present a theoretical framework, which describes how
police training programs can be developed in order to improve learning retention and the transfer
of skills to the work environment.
Design/methodology/approach – A brief review is provided that describes training strategies
stemming from Cognitive Load Theory (CLT), a well-established theory of instr uctional design. This is
followed by concrete examples of how to incorporate these strategies into police training programs.
Findings – The research reviewed in this paper consistently demonstrates that CLT-informed
training improves learning when compared to conventional training approaches and enhances the
transferability of skills.
Originality/value – Rarely have well-validated theories of instructional design, such as CLT, been
applied specifically to police training. Thus, this paper is valuable to instructional designers because it
provides an evidence-based approach to training development in the policing domain.
Keywords Cognitive Load Theory, Police training, Learning retention, Transfer of skills, Training,
Paper type Conceptual paper
Major advances have been made in the area of police training in recent years.
For example, in a number of countries, e-learning has become a common way of
delivering training courses to police officers in addition to the standard training
that continues to take place in the classroom and gymnasium (Canadian Police
Knowledge Network, 2008; European Police College (EPC), 2012; Interpol, 2012; Police
Sector Council (PSC), 2010). Indeed, between 2007 and 2009, officers across Canada
successfully completed over 8,100 e-learning courses on topics ranging from explosives
awareness to fatigue management (Canadian Police Knowledge Network, 2009).
Computer-based simulation training has also become a standard way of training
officers to perform effectively in different situations, including use-of-force encounters
(Bennell et al., 2007), driving exercises (Ross, 2009), and complex multi-agency
incidents ( Justice Institute of British Columbia ( JIBC), 2010).
While new training initiatives such as these undoubtedly have the potential to
benefit the policing community and enhance public safety, relatively little attention has
been paid, at least in the published literature, to how this training should be developed
and delivered in order to maximize its impact. Although some scholars have called for
theory-informed police training initiatives (e.g. Birzer, 2003; McCoy, 2006; Vodde, 2012;
Willis, 2010), what is currently absent in the police training literature are concrete,
empirically supported instructional strategies that can be incorporated into training to
promote the long-term retention and transfer of learned skills and knowledge. The
purpose of this paper is to address this gap by presenting a well-validated theory of
instructional design – Cognitive Load Theory (CLT) – and illustrating how some
specific instructional strategies stemming from this theory might be applied to
training in a policing context.
The current issue and full text archive of this journal is available at
Received 11 July 2011
Revised 16 May 2012
Accepted 22 May 2012
Policing: An International Journal of
Police Strategies & Management
Vol. 36 No. 2, 2013
rEmerald Group Publishing Limited
To accomplish our goal, the paper is divided into five main sections. First, a brief
overview of the strengths and weaknesses of current theoretical approaches to police
training is provided. Second, CLT is introduced as a theoretical framework that can
complement current theoretical approaches to police training. Third, a brief description
of various instructional design strategies stemming from CLT is provided.
Fourth, given the fact that the use of e-learning as a police training tool is rising on
a global scale (e.g. EPC, 2012; Interpol, 2012; PSC, 2010), examples of how CLT-based
training strategies can be used to enhance the quality of police training in an e-learning
environment are provided. Finally, aspects of CLT that require further research
1. Current approaches to police training
Throughout the past decade, a number of scholars have proposed changes to police
training that generally call for training programs to be designed with educational
theories in mind. For instance, a common recommendation in the police training
literature is to ensure that training initiatives are consistent with the principles
underlying adult learning (e.g. Birzer, 2003; Birzer and Tannehill, 2001; Glasgow
and Lepatski, 2012; McCoy, 2006; Palmiotto et al., 2000; Vodde, 2012; Werth, 2011).
Generally speaking, adult learning theories focus on the idea that adults are
independent learners and therefore respond to active, rather than passive, learning
initiatives. As such, adult learning environments should primarily be centred on the
student (i.e. andragogical) rather than the instructor (i.e. pedagogical) (Brookfield, 1986;
Knowles, 1980). For instance, rather than using a lecture-based format of instruction,
it is suggested that instructors of adults make use of learning activities that support
the autonomous nature of adult learners, such as group discussions and problem-
solving activities involving real-world case studies (Birzer, 2003; Knowles, 1980, 1990).
Moreover, since adults enter the learning environment with a great deal of previous
experience, instructors are also encouraged to integrate this experience into learning
tasks (Knowles, 1990).
Taking a student-centered approach to adult learning can also allow the many
different learning styles of adults to be accommodated in the training environment (e.g.
visual, auditory, read/write, and kinaesthetic/tactile) (Sprenger, 2003). As is the case
with other adult learners more generally (e.g. Bahadori et al., 2011; Breckler et al., 2009;
Johnson, 2009; Murphy et al., 2004), Landry (2011) recently found that the dominant
learning style of most officers in his sample was multimodal. Thus, using instructional
strategies that appeal to all learning styles is likely to be optimal. For visual learners,
this could be in the form of using PowerPoint and pictures for descriptions.
For auditory learners, small group discussions and the repetition of difficult
concepts are appealing. For read and write learners, case studies can be used. Lastly,
for kinaesthetic learners, role-playing and other hands-on activities may be best
In the policing literature, it has also been suggested that consideration of Bloom’s
Taxonomy of Educational Objectives (Bloom et al., 1956) is crucial when developing
police training programs (e.g. Cleveland, 2006; Vodde, 2009; Werth, 2011; Willis, 2010).
Briefly, Bloom’s Taxonomy is used to classify different learning objectives based
on three main domains – cognitive (i.e. knowledge) (Bloom et al., 1956), affective (i.e.
attitudes) (Krathwohl et al., 1964), and psychomotor (i.e. skills) (Simpson, 1972). Within
each of these domains, various levels of learning are defined. The learning that occurs
at the higher levels is dependent on having attained the relevant knowledge, skills,
and/or attitudes at the lower levels. The cognitive domain is the most frequently
cited, and is comprised of six cumulative levels of learning, beginning at the most basic
forms of learning (e.g. knowledge recall and comprehension) and progressing to
more intermediate (e.g. application of this knowledge) and finally, more complex
(e.g. analysis, synthesis, and evaluation) forms of learning (Bloom et al., 1956).
Given the dynamic problems faced by police officers while on-the-job, they are often
expected to perform at the highest level of Bloom’s Taxonomy (Cleveland, 2006; Werth,
2011). That is, not only are they expected to recall and understand a number of tasks,
but they are also required to analyse and evaluate a variety of problems on a daily
basis (Cleveland, 2006). As such, training needs to facilitate the development of these
higher-order thinking skills. As one way to develop these skills, problem-based
learning has been proposed by a number of police scholars (Barrows and Tamblyn,
1980). At the outset of a problem-based learning activity, trainees are given a real-life
problem and are expected to outline the key issues, identify available resources to
address the problem, and then generate an action plan for solving the problem
(Cleveland, 2006; Pitts et al., 2007; Werth, 2011). Problem-based learning activities are
believed to tap into the complex skills related to the higher levels of Bloom’s Taxonomy,
including the critical thinking and communication skills that are at the core of a police
officer’s everyday job responsibilities (Cleveland, 2006).
Although these theoretical orientations provide a promising general philosophy to
guide the development and delivery of police training programs, they do not always
offer tangible, empirically driven strategies concerning how curriculum should be
designed to achieve optimal learning conditions. Indeed, while Bloom has argued that
his original taxonomy should be modified to better fit the needs of learners within a
given field (Anderson et al., 2000), it is not always clear how the taxonomy should be
modified. Likewise, research has shown that although instructors are generally aware
of adult learning principles (i.e. they are taught the principles as a component of their
basic instructional techniques courses), they continue to rely on traditional pedagogical
forms of instruction (McCoy, 2006). This may be, at least in part, because police
instructors have not always been provided with specific training strategies that they
can implement, which support the adult learning paradigm.
We argue that the CLT framework may provide curriculum designers, and
ultimately instructors, with additional guidance in this respect. For instance, we
believe that research on CLT can provide answers to important questions that
frequently arise in training, such as how should instructional materials used in police
training be designed to ensure that learning outcomes are maximized? What are some
concrete strategies that can be used to promote active, rather than passive, learning?
How can police training be designed to facilitate the transition from the lower levels
of Bloom’s Taxonomy to the higher-order thinking skills of analysis, synthesis, and
Merely because an instructor adheres to adult learning principles does not mean
that learning will always prevail. This is because, regardless of the learning
philosophy presumed (i.e. pedagogy or andragogy) or the level of Bloom’s Taxonomy
that is targeted, the cognitive resources of the learner must also inform training
endeavours. We argue that CLT can further enhance these general theoretical
approaches by offering a number of practical techniques to ensure that the training
environment accommodates the cognitive limitations of trainees while also adhering to
the general principles of adult learning. Throughout the remainder of this paper, a
number of ways in which CLT can inform current theoretical approaches to police
training are provided. First, however, the basic concepts guiding CLT-informed
instruction must be outlined.
2. What is CLT?
Since its inception in the 1980s (e.g. Sweller, 1988), CLT has been primarily concerned
with how instructional materials should be designed and delivered in order to provide
trainees with the optimal conditions for learning. In other words, CLT focuses on the
most effective way to design training curriculum in order to positively impact
the learner. Training recommendations that are based on CLT draw on knowledge of
how the human cognitive system operates in an attempt to facilitate the development
of cognitive structures (i.e. schemas) on the part of the learner; structures that are
involved in the learning process, including the retention and transfer of knowledge and
skills (Ayres and Paas, 2009).
2.1 Memory systems and structures
The cognitive architecture that provides the foundation for CLT-based training
strategies consists of two related systems of human memory, including working
memory (WM) and long-term memory (LTM). More specifically, the central focus of
CLT is on the instructional design consequences of the limited ability of WM to process
information, coupled with the virtually unlimited capacity of LTM to store learned
information (Kirschner, 2002; Paas et al., 2003; Sweller, 1988). According to CLT, the
ability of these systems to process incoming information (e.g. during training)
and retrieve stored information (e.g. while on-the-job) is aided by the acquisition and
automation of cognitive structures referred to as schemas (Paas et al., 2003).
WM. WM is the memory system that briefly stores new, incoming information
while simultaneously manipulating this information in order for learning to occur
(Baddeley, 1992). The primary function of WM is to temporarily hold information that
is being actively, or consciously, processed (van Merrie
¨nboer and Ayres, 2005).
Any piece of information that needs to be learned (e.g. a word, a letter, a number) is
referred to as an element in WM (Sweller, 1994). WM can only hold between five and
nine elements at one time. Moreover, WM can only deal with actively processing two
to four related elements simultaneously (Groeger, 1997; Kirschner, 2002; Miller, 1956;
Paas et al., 2003).
LTM. Although our WM store is extremely limited, we are able to acquire and recall
an extensive amount of knowledge throughout our lifetime. This is due, in part, to the
existence of our LTM. The primary function of LTM is to store information when it is
not being consciously attended to, but is still essential for everyday understanding
(Bower, 1972). LTM is essentially unlimited; that is, it is able to store an immeasurable
amount of interrelated information that can (typically) be retrieved when needed
(Bower, 1972). The central concern of CLT is how training should be designed and
delivered to ensure that trainees can effectively process the to-be-learned information
in their limited WM, so as to facilitate the construction of cognitive schemas to be
stored in, and retrieved from, LTM (van Merrie
¨nboer and Sweller, 2005).
Schemas. Learning is often conceptualized as a change in the knowledge
structures that form LTM (Pawley et al., 2005). Changes to an individual’s LTM result
from the construction of cognitive schemas, which assemble incoming information
elements into understandable subgroups or chunks of information that hold the
same underlying purpose to the learner (Paas et al., 2003; Tennyson and Elmore, 1997).
Schemas range in complexity, from those involving only a few related concepts
(e.g. how to identify a type of narcotic) to higher order schemas that can encompass
many related pieces of information that have been acquired over time (e.g. the various
actions that an officer must undertake when seizing illegal substances) (Pawley et al.,
2005). A schema permits an individual to recognize situations as belonging to a
specific type requiring certain identifiable actions to reach an appropriate solution
(Paas, 1992). Schemas may also be used as a guide in new circumstances, fostering
transfer of knowledge to situations that the individual has not previously encountered
(Paas, 1992). For instance, a police officer may be confronted with aspects of a spousal
assault situation that he or she has never before handled, but their generalized
knowledge about domestic assault, embodied in a “domestic assault schema”, may
suggest a certain response.
In general, schemas have two related functions: storing and organizing information
in LTM and reducing the burden placed on WM (Pawley et al., 2005; Sweller et al.,
1998). Because of their organizational structure, any given schema in LTM can hold a
large amount of information without taxing an individual’s cognitive resources
(Kirschner, 2002). Similarly, because a complex schema can be dealt with as one
element when brought into WM, schemas serve to reduce the strain placed on the
learner by freeing up WM space (Sweller et al., 1998). If enough practice occurs,
schemas can also come to direct behaviour in a relatively unconscious fashion (i.e. they
can become automated; Sweller et al., 1998). Since automated behaviours can be
performed with little conscious effort, schema automation also serves to further reduce
WM load (van Merrie
¨nboer and Ayres, 2005).
Accordingly, CLT proposes that the primary goal of training programs should be to
promote schema acquisition and automation (Kirschner, 2002; Sweller et al., 1998).
Both processes free WM capacity that can then be devoted to making sense of new
instructional materials during training.
2.2 Defining cognitive load
In general, the term “cognitive load” refers to the mental effort experienced by a
learnerwhenperformingacertaintask(Clarket al., 2006). In order to achieve
optimal training results (i.e. for schema acquisition and automation to occur) CLT
holds that the cognitive load placed on learners’ WM must be appropriately
managed. According to CLT, three types of cognitive load exist: intrinsic, extraneous,
and germane. More specifically, WM load can be affected by: the inherent complexity
of the instructional material (i.e. intrinsic load), the way in which the to-be-learned
material is presented to the learner (i.e. extraneous load), and the amount of effort
directly devoted to schema acquisition and automation (i.e. germane load; Kirschner,
2002; van Merrie
¨nboer and Ayres, 2005). CLT generally holds that, for training to be
effective, intrinsic load must be managed, extraneous load must be reduced,
and germane load must be increased, so long as the limited capacity of WM is
Intrinsic cognitive load. Intrinsic cognitive load is primarily the result of the inherent
complexity of the information being conveyed to the learner. It refers to a “base” load
that cannot be reduced until new schemas are constructed to guide further learning of
related concepts or skills (Kirschner et al., 2009a). Intrinsic load increases with the
complexity of the to-be-learned information, and it is largely determined by the number
of interacting elements of information that need to be processed in WM simultaneously
in order for full understanding to occur (i.e. the degree of element interactivity; Paas
et al., 2003; Pollock et al., 2002).
Information low in element interactivity can be grasped relatively easily and
understanding does not depend on previously learned information (Paas et al., 2003).
For example, basic training on the mechanics of applying restraints (e.g. handcuffs)
is relatively low in element interactivity because each step of the process (i.e.
withdrawing the handcuffs, placing them on the wrists of an individual, and locking
them) can be effectively learned in isolation. Information that is low in element
interactivity generates a low level of intrinsic load because, in order to learn the
material, only a limited number of elements must be processed simultaneously in WM
(Sweller and Chandler, 1994). Likewise, consider an e-learning course on cultural
diversity where an officer is learning about the spiritual and cultural characteristics of
the community members they serve. Elements of this module could, in theory, be
learned in isolation from one another. For instance, learning about the population levels
and distribution of different cultural groups within their district, and identifying the
key cultural sensitivities that should be taken into consideration when communicating
with different cultural groups could be learned independently of one another. As such,
this e-learning module would represent a low level of element interactivity.
In contrast, information that is high in element interactivity cannot be fully
understood in isolation. That is, a trainee must be able to process all related elements at
the same time in WM in order for meaningful learning to occur (van Merrie
Ayres, 2005). For instance, consider the previous task of learning how to apply and lock
handcuffs. When training officers on how to implement these procedures during
common police encounters element interactivity increases. A patrol officer intervening
in a domestic dispute, for example, must simultaneously consider a multitude of factors
when dealing with the situation, only one of which is whether handcuffs should be used
(and how this can best be accomplished). Indeed, in order to determine whether
(and how) handcuffs should be used, the officer may first have to consider whether
unlawful behaviour is being exhibited, the level of threat that exists in the situation,
and how much force is necessary to neutralize the abusive partner. In other words, in
naturalistic scenarios, many elements must be considered and processed in a collective
fashion in order for the officer to successfully resolve the situation. As a result, material
of this nature will impose a high intrinsic load. Similarly, consider an e-learning
scenario where officers are learning the basics of conflict resolution with a particular
cultural group. Devising a way to successfully resolve the conflict may involve
multiple aspects of a police officer’s knowledge base, which not only includes the
general laws surrounding the dispute, but also the potential cultural sensitivities that
the police officer may encounter. Since the information the officer learned in the
cultural diversity e-training described earlier must be considered simultaneously with
their conflict resolution skills, element interactivity increases.
As mentioned, since element interactivity is directly related to what is being
learned, intrinsic load cannot be reduced by changing the format of instruction (Paas
et al., 2003). However, intrinsic load can be managed within the training environment
¨nboer and Sluijsmans, 2009). This can be accomplished, for example, by
providing the learner with simpler training tasks (e.g. the appropriate use of handcuffs)
that omit some of the interacting elements (e.g. the circumstances under which
handcuffs can and should be used) at the outset of training, particularly if the trainee
lacks schemas to direct their learning (i.e. a novice; van Merrie
¨nboer et al., 2003).
For effective learning to occur, however, all elements central to the task at hand must
eventually be processed simultaneously so that a full understanding of the task is
realized (Paas et al., 2003; van Merrie
¨nboer and Sluijsmans, 2009).
Extraneous cognitive load. Extraneous cognitive load is comprised of the learning-
irrelevant demands placed on the learner during training, and is mostly a function of
the how training is designed and delivered (Clark et al., 2006). Extraneous load is
harmful to learning because it directs a learner’s limited WM resources to activities
that are unrelated to the essential learning processes of schema acquisition and
automation (Paas et al., 2003). Since extraneous load is a result of the manner in which
instructional material is structured and presented to trainees, it can be directly reduced
by modifying the method of instruction.
Using verbal explanations to describe things that are best explained visually is a
clear example of a training method that increases extraneous load. Take, for example,
an online firearms identification course where officers are learning the characteristics
(e.g. make, manufacturer, calibre, barrel length, working mechanisms, etc.) of various
firearms. In this instance, training would be much more efficient and extraneous load
would be reduced if the appropriate training medium was used – where different
firearms are visually presented to trainees rather than verbally explained (e.g. in
graphic form when training is done in an e-learning environment).
Germane cognitive load. Like extraneous load, germane load is also a result of the
method of instruction, or how training is designed and delivered. In direct contrast to
extraneous load, however, the processing demands placed on WM as a result of
germane load are thought to be directly relevant to schema acquisition and automation
(Moreno, 2004; van Merrie
¨nboer and Ayres, 2005). Germane load is important because
it enhances the learners’ understanding of why and when to apply a certain solution to
a problem, or when to engage in a particular behaviour (Paas and van Gog, 2006).
According to CLT, however, the majority of learners do not spontaneously engage in
these germane processes (Paas and van Gog, 2006; Renkl, 1997). As a result, CLT
proponents argue that a central goal of instruction should be to maximize germane
load by stimulating learners to engage in these processes whenever WM space is
available to do so (Paas et al., 2003; Paas and van Gog, 2006).
Promoting germane processes among learners can be achieved in a number of ways.
For instance, consider a police training program that teaches officers how to
communicate effectively with the mentally ill (either in a classroom or e-learning
environment). Although learning may initially occur more easily if the training
scenarios remain very similar across training exercises (e.g. involving only one type of
mental illness), the communication skills that the officer develops as a result of such
training will be more restricted (i.e. less flexible or adaptive) than those developed
when using more varied training exercises (e.g. involving a variety of mental illnesses).
Under the varied training scenario, a higher cognitive load is undoubtedly imposed
on the learner, but this load is directly related to schema acquisition. The schema
that results from this training (v. the low variability training) will likely be more
flexible and adaptive in nature, allowing the officer to ultimately perform more
effectively when on-the-job.
CLT-based training approaches. It is important to note that the three sources of
cognitive load described above – intrinsic, extraneous, and germane – are believed to
be additive in nature (Paas et al., 2003). That is, the total cognitive load experienced
by a learner in a given training program is equal to intrinsic plus extraneous plus
germane load (Kirschner, 2002). Based on this fact, there are a variety of training
approaches that are compatible with a CLT perspective (see Figure 1). If implemented,
these strategies should increase the effectiveness of police training (i.e. greater learning
retention and enhanced transfer of skills/knowledge).
As shown in Figure 1, traditional instruction (i.e. instruction that does not take into
account the cognitive limitations of trainees) can often result in cognitive overload.
When intrinsic load is high, and nothing has been done to control extraneous load, WM
is likely to be so overwhelmed that learning will be impaired. However, if measures are
taken to reduce extraneous load, and intrinsic load is not so high that WM resources
are depleted, the result will be free WM capacity (CLT Strategy 1). Under such
circumstances, it is then advisable for instructional designers to incorporate strategies
into their training that further facilitate the acquisition and automation of schemas, or
to increase germane load (Kirschner, 2002; Sweller et al., 1998) (CLT Strategy 2).
Alternatively, in instances where intrinsic load is so high that using techniques to
foster germane load are likely to overload WM capacity even after all sources of
extraneous load have been eliminated, intrinsic load can be managed by dividing the
instructional materials into more manageable, or smaller, parts during the early stages
of training (CLT Strategy 3). In sum, rather than developing methods to reduce the
overall cognitive load experienced by trainees, CLT researchers are primarily
concerned with determining the most appropriate methods to optimize these three
types of load in order to maximize training results (van Gog and Paas, 2008).
Importantly, even though these strategies are interrelated, each one can be implemented
independently of the others in training contexts. That is, curriculum can be designed
that adheres to the approach illustrated in Strategy 1, Strategy 2, or Strategy 3 (van
¨nboer and Sweller, 2010). Indeed, when the instructional material is too complex for
instructors to facilitate germane processing (as accomplished in Strategy 2), or when the
intrinsic complexity of the material cannot be managed by separating it into meaningful
subcomponents (as accomplished in Strategy 3), it may be appropriate for instructors to
focus their attention on reducing sources of extraneous load (as accomplished in Strategy 1).
Eliminating all irrelevant sources of cognitive load will ensure that learning, at a
minimum, is not impeded by WM overload as a result of inefficient instruction.
2.3 Novices v. experts: differences in cognitive load
CLT is also concerned with how a learner’s level of expertise can affect the various
types of cognitive load experienced by the trainee in the training environment, and as a
WM capacity Traditional
Source: Adapted from Paas et al. (2003) and van Merriënboer and Sweller (2010)
Methods for improving
training by taking into
account various ways to
manipulate cognitive load
within the bounds of
available WM capacity
result, the manner in which learning optimally occurs. This is clearly an issue that
must be considered in the policing context. For example, as a police officer progresses
through their career, what may have initially been multiple interacting elements
(i.e. representing material with a high intrinsic load) will eventually be conceptualized
as a single element in the cognitive schema (i.e. representing material with a lower
intrinsic load; van Merrie
¨nboer and Sweller, 2005). As a result, providing police recruits
with instructional guidance that may be crucial for learning to occur at the outset of
training (e.g. isolating interacting elements, reducing the level of variability across
training scenarios, etc.) can actually become detrimental to learning once schemas have
Consequently, training for novices and experts should often be designed differently
in order to take into account their varying degrees of schema development (Ayres and
Paas, 2009). If instructional designers fail to take into account the experience level of
their trainees when designing and delivering training materials, the expertise reversal
effect can occur. That is, when instructional methods that work well for novices are
also used to train learners who have already acquired a certain level of mastery in the
area of instruction, learning may not take place and can even be impeded (Kalyuga
et al., 2003; Paas et al., 2003). Similar to adult learning theories then, the expertise
reversal effect of CLT acknowledges that a trainee’s previous experience must
be incorporated into the training environment, and that learning should (eventually) be
student-directed for a deeper learning to occur. With that said, the expertise reversal
effect also demonstrates that these more independent learning strategies should only
be used once related schemas have been adequately developed through an appropriate
amount of instructional guidance.
3. How can CLT be used to improve training?
According to CLT, training in any area can be enhanced by taking into account the
reality of the trainees’ cognitive structures and the various types of cognitive load
encountered during training. Cognitive load theorists have developed and proposed
numerous instructional strategies that have consistently been found to be superior to
traditional training approaches. Not only do these strategies make optimal use of the
trainees’ cognitive resources, but many of them can serve to facilitate a training
environment that is consistent with the adult learning paradigm. In this section, some
of the primary instructional design strategies that have emerged from CLT are
reviewed. These strategies focus on decreasing extraneous load, increasing germane
load, and/or managing intrinsic load.
3.1 Minimizing extraneous load
Recall from Figure 1 that failing to control for extraneous load can result in WM
systems becoming overloaded, which can impair learning. CLT has proposed several
instructional effects that serve to reduce the extraneous load placed on trainees.
Adopting these techniques make it possible for the learner’s attention to be directed to
learning-relevant processes that are directly related to schema acquisition and
automation. Some of the most well-known effects include: worked example, problem
completion, split-attention, redundancy, and modality effects.
Worked example effect. Conventional practice problems provide the learner with a
problem to solve without any additional guidance. Such practice problems focus
limited WM resources on tasks that are not primarily concerned with learning (e.g.
searching for the appropriate solution strategy), which undoubtedly increases the
amount of extraneous load experienced by the learner. Worked examples, on the other
hand, show learners in a straightforward, step-by-step fashion what information
is needed to solve a particular problem (Clark et al., 2006; Schwonke et al., 2009).
The learning improvements that occur as a result of employing worked examples
constitute the worked example effect (Cooper and Sweller, 1987; van Gog et al., 2006;
Hilbert and Renkl, 2009; Rourke and Sweller, 2009; Schwonke et al., 2009). The worked
example effect is likely the most heavily studied CLT effect to date (Sweller, 2006).
Research has indicated that worked examples can be an effective instructional tool
for both individual as well as group-based learning activities (Retnowati et al., 2010).
With that said, research has consistently demonstrated that worked examples are best
suited for novice, rather than more experienced, learners (e.g. Kalyuga et al., 2001;
Reisslein et al., 2006). This is because as experience increases and schemas are formed,
worked examples can become a redundant source of information since the trainees no
longer need to invest cognitive effort in studying information (e.g. steps in solving a
problem) that they have already acquired (Clark et al., 2006; Kalyuga et al., 2001).
As a consequence, traditional problem solving is often superior to learning with
worked examples when training experienced learners (Kalyuga et al., 2003).
At first glance, the worked example effect may seem counter-intuitive to the self-
directed, independent learning principles that are central to the adult learning
framework. However, as Knowles (1990) explains, even with adult learners it may be
appropriate to begin with pedagogical instruction (i.e. instructor-driven learning) in
situations where the learners are new to the learning environment (i.e. new recruits).
Likewise, the use of worked examples is also consistent with Bloom’s Taxonomy in the
sense that worked examples are designed to ensure an adequate level of knowledge
and comprehension before more complex tasks are focused on (e.g. application,
analysis, or synthesis). In fact, Werth (2011) found that some police trainees opposed
the self-directed nature of problem-based learning due to the fact that they felt ill
equipped to accomplish the assigned tasks. Beginning the training program by
offering trainees some initial guidance in the form of worked examples ensures that
they will have the essential resources they need to carry out higher-order tasks. Not
only will worked examples reduce the cognitive load placed on the new recruits’ WMs,
but it may also make them more comfortable with the more independent, andragogical
learning exercises they encounter later in the course of training.
Problem completion effect. In order to ensure that learners process worked
examples effectively, trainers may wish to adopt completion examples as an
instructional strategy (Paas and van Merrie
¨nboer, 1994; Renkl and Atkinson, 2003).
A completion example combines the benefits of a worked example with that of a
conventional practice problem (i.e. portions of the problem are provided in a step-by-
step solution format, and other parts require completion by the learner; Clark et al.,
2006). Research has shown that completion examples improve learning outcomes
when compared to conventional problem solving (Clark et al., 2006; Paas, 1992;
¨nboer and de Croock, 1992), a finding that has been referred to as the
problem completion effect.
In line with worked examples, completion examples serve to reduce extraneous load
by directing the learner’s cognitive resources toward aspects of the task that facilitate
learning (i.e. the solution steps explicitly provided). Furthermore, deeper processing
(and better learning) of the instructional materials is encouraged by requiring the
learner to independently realize the completion aspects of the problem (Clark et al.,
2006). In essence, the use of completion examples can help transition recruits from
instructor-driven learning events to self-directed learning events, or from the lower
levels of learning in Bloom’s Taxonomy to the higher levels of learning.
When should completion examples take precedence over fully worked-out
examples? Generally speaking, the amount of guidance provided to trainees should
be a function of both the complexity of the material being learned as well as the
trainee’s level of expertise (Cooper, 1998). For instance, novices should be provided with
more complete examples (i.e. more guidance) than experienced learners, especially
when the material is complex (i.e. characterized by a high level of element interactivity)
(Clark et al., 2006).
Split-attention effect. Split-attention refers to the extraneous cognitive load that
learner’s experience when the material they are studying (e.g. via worked examples) is
designed in such a way that the training requires them to mentally relate multiple
sources of information that are separated in either space (e.g. text and a diagram) or
time (e.g. a diagram and accompanying audio), even though both sources of
information are required for full understanding to be achieved (Clark et al., 2006; Clark
and Mayer, 2008). The extraneous load caused by split-attention can be avoided by
ensuring that the instructional materials are integrated with, rather than separated
from, one another (Kalyuga et al., 1999). The improvement in learning that arises as a
result of this integration is referred to as the split-attention effect (Chandler and
Sweller, 1991, 1992; Sweller and Chandler, 1994; Tindall-Ford et al., 1997).
Redundancy effect. In addition to ensuring that multiple sources of information are
integrated in space (e.g. text and diagram) and time (e.g. audio and visual) when
presented to learners, it is equally important to ensure that both sources of information
are indeed essential for one to come to a complete understanding of the instructional
materials (Cooper, 1998; Pawley et al., 2005). If the integrated materials are redundant
(i.e. both sources provide learners with the same knowledge), then one source can,
and probably should, be removed (Kalyuga et al., 1999; Pawley et al., 2005). When
learning is enhanced as a result of eliminating redundant sources of information from
training materials, the redundancy effect has occurred (Chandler and Sweller, 1991;
Leahy et al., 2003).
The redundancy effect occurs because learners will attempt to process multiple
sources of information in their WM even if they provide the same information (Clark
et al., 2006; Pawley et al., 2005). This additional processing uses WM resources that
could otherwise be directed to activities that are more relevant to learning (e.g.
germane processes). In line with the other CLT effects mentioned above, adherence to
the redundancy principle is particularly crucial when element interactivity is high, and
the learners are novices, rather than experts (Clark et al., 2006).
Modality effect. In situations where physically integrating two related sources
of non-redundant information is difficult, dual-mode presentation of instructional
materials can be used as an alternative technique to deal with the processing demands
resulting from split-attention (Kalyuga et al., 1999; Mayer, 1997). For instance, if
incorporating relevant textual information into a diagram makes the instructional
material appear too cluttered or complex, the text can instead accompany the diagram
in audio format. The enhancement in learning that is observed as a result of this dual
mode presentation has been referred to as the modality effect (Bru
¨nken et al., 2004;
Ginns, 2005; Moreno and Mayer, 1999; Tindall-Ford et al., 1997).
It has been suggested that the modality effect emerges as a result of an increase
in WM capacity (Kalyuga et al., 1999; Mayer, 1997). Essentially, WM is thought to
consist of two partially unconnected processors for auditory and visual information
(Baddeley, 1992). By using two separate modes of presentation, WM may not be
overloaded because both the auditory and visual channels (rather than just the visual
or auditory channel) are being used (Kalyuga et al., 1999). It is important to note,
however, that the beneficial effects of employing both processing channels seems
to arise mostly in situations where: the auditory information adds something above
and beyond the visual information (Leahy et al., 2003), the knowledge or skills
being learned are high in element interactivity (Tindall-Ford et al., 1997), and less
experienced learners are being targeted (Kalyuga et al., 2000; Seufert et al., 2009).
3.2 Maximizing germane load
Not only should extraneous load be minimized by employing the above instructional
techniques when applicable, but strategies that encourage learners to engage in deeper
processing of instructional materials should also be incorporated into training.
Such strategies accord well with adult learning frameworks, as one of the central goals
of self-directed learning is to promote more active, or involved, learning among
trainees (Brookfield, 1986; Knowles, 1980). Moreover, since these strategies promote
more active processing, they can help to move trainees past the basic knowledge
and comprehension stages of Bloom’s Taxonomy to the more complex stages, such as
application, analysis, and synthesis.
Strategies that directly target schema acquisition and/or automation place
additional load on WM – germane load – however, this load can be considered “good”
load since it can facilitate learning. As shown in Figure 1, if enough WM capacity has
been made available by reducing all sources of extraneous load, increasing germane load
through the use of specific training strategies can maximize learning results. Some of the
most well-known training effects to emerge when germane load is focused on are the:
variability, self-explanation, and imagination effects.
Variability effect. When attempting to apply the knowledge (and skills) attained
during training while on-the-job, police officers frequently must apply this knowledge
(or skills) across a wide range of circumstances that differ from those encountered
during training. This makes it very important for training to assist the officer in
developing flexible or adaptive schemas that allow for the transfer of skills to the work
environment (de Croock et al., 1998; Gick and Holyoak, 1983). CLT researchers have
found that the construction of more flexible, adaptive schemas can be achieved by
providing trainees with diverse, rather than many similar, training examples
throughout the course of training (Clark et al., 2006; Sorden, 2005; van Merrie
et al., 2006). The enhancement in learning that results from exposure to variation in
training has been termed the variability effect (Paas and van Merrie
¨nboer, 1994; Quilici
and Mayer, 1996).
Similar to other CLTeffects, however, the impact of including a range of scenarios in
training has also been shown to vary as a function of the complexity of the to-be-
learned information. That is, if the instructional material is too complex, positive
contributions of variability to learning may not be found (Große and Renkl, 2006).
This is because the learner is provided with the additional task of determining the
important commonalities across the varied examples (Clark et al., 2006). As a result,
incorporating varied solutions into training may overload WM unless other types
of cognitive load are appropriately dealt with in advance (e.g. ensuring all sources of
extraneous load are eliminated).
With respect to adult learning models, not only can example variability encourage
deeper processing of the instructional materials, but varying the manner in which the
scenarios are presented to learners can also serve to target different learning styles. For
example, as previously mentioned, the majority of police officers are likely to be
multimodal learners in the sense that they learn best through the use of a variety of
different techniques (Landry, 2011). Using various delivery models to present diverse
training scenarios (e.g. case studies, role plays, oral presentations, etc.) can help ensure
that each learner is provided with the chance to develop their knowledge and practice
their skills in a way that corresponds to their individual learning preferences.
Self-explanation effect. Even when trainees have been provided with varied learning
scenarios, they may need additional support to ensure they attain more than surface
knowledge of the examples. CLT research has shown that a deeper processing of
worked examples can be encouraged by a process called self-explanation (Atkinson
and Renkl, 2007; Clark et al., 2006). Self-explanation often entails having the learner
explicitly elaborate upon underlying principles of the instructional domain (e.g.
Atkinson et al., 2003) or establish connections between different aspects of the problem
(e.g. Wong et al., 2002). As a whole, having students self-explain while learning
encourages them to “attend to the material in a meaningful way while effectively
monitoring their evolving understanding” (Roy and Chi, 2005, p. 272). Although self-
explaining increases the cognitive load experienced by learners, a number of studies
have found that it results in a deeper, more deliberate, processing of instructional
materials, ultimately leading to greater learning gains (e.g. Atkinson et al., 2003;
Bielaczyc et al., 1995; Chi et al., 1994; Hilbert and Renkl, 2009; Renkl, 1997). Collectively,
such findings provide the basis for what has been labelled the self-explanation effect.
Research suggests that providing learners with specific, or detailed, prompts to
explain certain aspects of the task they are being trained on is more likely to result in
greater learning achievements than merely providing learners with the more general
prompt to “think aloud” (Bielaczyc et al., 1995; Busch et al., 2008; Chi et al., 1994).
This appears to be especially true for novice learners who lack pre-existing schemas
(Renkl et al., 1998).
Imagination effect. Under certain circumstances, mentally rehearsing the steps
required to successfully complete a given task can enhance learning as opposed to
merely studying the same materials. The learning gains resulting from adopting this
strategy represent the imagination effect (Cooper et al., 2001; Ginns et al., 2003;
Leahy and Sweller, 2004; Tindall-Ford and Sweller, 2006). CLT researchers have
suggested that the beneficial effects of imagination stem from the fact that
using mental imagery enables learners to more explicitly monitor their understanding
by allowing them to identify what aspects of the learning materials they
are comfortable with and what aspects require additional rehearsal (Leahy and
Unlike the majority of other CLT effects, which have been found to apply mostly to
novice learners, mental rehearsal appears to be most effective when used by more
experienced learners (Cooper et al., 2001; Ginns et al., 2003; Leahy and Sweller, 2005).
This is because the existence of schemas makes more space available in WM so that
processes that create additional WM load, such as mental rehearsal, can be carried out
effectively (Clark et al., 2006; Cooper et al., 2001). Consequently, experienced learners
are more likely to benefit from the imagination technique than novice learners who do
not have enough processing capacity available to engage in such a demanding task,
and who instead benefit more from repeatedly studying the material presented in
worked-out or completion examples, at least at the outset of the training program
(Leahy and Sweller, 2005).
3.3 Managing intrinsic load
In many training domains, even when all sources of extraneous load have been
appropriately reduced, the complexity of the instructional materials may be so high in
the initial stages of training that it inevitably exceeds the trainee’s available WM
resources (van Merrie
¨nboer et al., 2003, 2006). As shown in Figure 1, trainers can
resolve this issue by implementing techniques to effectively manage this intrinsic load.
Two of the techniques that enable one to accomplish this involve sequencing
and fading, which can lead to the sequencing and fading effects. Collaborative learning
is also another potential strategy, though it is not yet as well established as a
Sequencing effect. As illustrated in Figure 1, one way that intrinsic load can be
managed is by initially separating highly complex information into more manageable
subcomponents for learners, before slowly presenting the learner with the entire task
(Ayres, 2006; van Merrie
¨nboer et al., 2006). The improvement in learning that
frequently occurs when such a strategy is employed is called the sequencing effect
(Clark et al., 2006; Kester et al., 2006; van Merrie
¨nboer et al., 2006). The effect occurs
because sequencing the complex material artificially reduces the element interactivity
experienced by the learner when encountering novel information. Sequencing allows
the learner to develop some initial schematic knowledge in the domain by studying the
subcomponents in a sequential fashion. After these initial stages, WM can better
handle the cognitive demands imposed by the whole task (Pollock et al., 2002; van
¨nboer et al., 2006). When learners have more experience in the task domain, or
when the material under question is low in complexity, sequencing may not facilitate
further learning, as WM load is likely not overwhelmed under such instances (e.g.
Ayres, 2006; Clarke et al., 2005; Pollock et al., 2002).
Fading effect. In order to deal with the fact that learners who are novices at the
outset of a training program eventually acquire some level of experience in the
domain as training progresses (i.e. they begin to develop more detailed schemas), CLT
researchers have recommended that trainers gradually decrease the support and
guidance provided to learners (Renkl and Atkinson, 2003). For instance, although
problem-based learning is a promising student-driven training strategy that should not
be abandoned, new recruits may not be prepared to successfully accomplish these
tasks at the outset of a training program (Werth, 2011; Willis, 2010). Fading can be used
to resolve this issue, where trainees are first provided with fully worked-out examples
(i.e. complete guidance), then completion problems (i.e. partial guidance) and, when
ready, conventional problem-solving tasks where no guidance is provided and the
trainee is required to independently arrive at a solution (Atkinson et al., 2003; Kalyuga
et al., 2003; Schwonke et al., 2009). The superior learning that results from the
implementation of this training strategy is referred to as the fading effect, and this
strategy has generally been found to combat the detrimental effects on learning that
can sometimes occur as a result of the expertise reversal effect (Renkl et al., 2004).
Moreover, research conducted in the work setting has shown that this fading technique
can be an effective instructional strategy with adult learners (Kissane et al., 2008).
Collaborative learning. Although not yet fully recognized as a CLT effect, recent
research has suggested another way to manage the intrinsic load associated with
complex learning tasks: collaborative, or group-based, learning. Specifically, research
has suggested that, when dealing with highly complex learning tasks, collaborative
learning can be more beneficial than individual learning, particularly in terms of
learning transfer results (Kirschner et al., 2009a, b).
CLT suggests that the benefits of group learning arise from the fact that many
WMs, rather than just one, are available to process the instructional information.
The increase in WM capacity is thought to reduce the intrinsic load experienced by any
one learner (Kirschner et al., 2009a, b). This results in enhanced learning, provided that
the group is able to communicate effectively. Similarly, it is believed that group
learning will only be superior to individual learning in instances where the task is
high in intrinsic load and thus may overload the WM of one individual (Kirschner et al.,
2009a b). Although research on collaborative learning remains in its infancy, this
strategy may have value in enhancing learning transfer results within the police
training domain, as police officers are often required to work together on a variety of
highly complex tasks while on-the-job. Moreover, this effect accords well with training
designed to consider adult learning principles where group activities are promoted to
facilitate independent, self-directed learning of the instructional materials (Birzer, 2003;
Vodde, 2012). Similarly, the instructor can make use of trainees’ previous experiences
in group-based learning by mixing experts (with more advanced schemas) with
novices (with less advanced schemas) so that WM load can be further reduced for the
4. Putting the effects into practice: an example using e-learning
Carefully considering instructional design, and how it can either positively or
negatively impact the cognitive resources of trainees, is vital for training in
complex domains, such as policing. All of the instructional techniques discussed
above have the potential to be applied to the law enforcement context, and as such,
improve both the effectiveness and efficiency of police training programs. How exactly
can the above CLT principles be incorporated into police training? This section
provides an overview of how CLT can be applied to police training in the online
environment, as this form of training has become increasingly common across
law enforcement agencies (e.g. CPKN, 2009; European Police College, 2012; Interpol,
2012) and can present some unique challenges to instructional designers (Clark and
Just as policing has shifted towards more modern (or technologically driven) forms
of training, such as e-learning (e.g. CPKN, 2009), so too has CLT research (e.g. Clark and
Mayer, 2008; Kalyuga, 2007; van Merrie
¨nboer and Ayres, 2005). E-learning has
generally been conceptualized as any form of computer-based training (Clark and
Mayer, 2008). The benefits of e-learning are twofold. First, organizations are able to
save travel time and costs associated with traditional classroom-based training (Clark
and Mayer, 2008). Second, e-learning allows for a greater degree of interactivity
between the learner and the training environment (Ayres and Paas, 2007).
As alluded to above, the e-learning environment presents some additional cognitive
load challenges to instructional designers beyond those commonly encountered in a
classroom setting (Clark and Mayer, 2008). Essentially this is because a wide array of
media elements can be easily incorporated into e-training, including text, narration,
music, instructional animations, other graphics, and so on (Clark and Mayer, 2008).
Although integrating all of these types of interactive elements into e-learning is
tempting and might be intuitively appealing, to do so can easily lead to situations
where a learner’s cognitive resources are exceeded. Properly managing cognitive load
by incorporating the instructional design principles outlined above remains an
essential component of designing effective instructional materials within the e-learning
context (Clark and Mayer, 2008; van Merrie
¨nboer and Ayres, 2005).
For instance, instructional animations are often incorporated into e-learning
environments. In situations where animations illustrate how to perform a motor skill
(e.g. patting down a suspect, taking a fingerprint, applying handcuffs, etc.), they may
be more effective than static diagrams (Ayres et al., 2009; Ho
¨ffler and Leutner, 2007).
However, the richness and transient nature of animations can have the negative effect
of overloading the learner’s WM when static diagrams are sufficient to achieve
understanding (Ayres and Paas, 2007; Clark and Mayer, 2008). Similarly, the use of
extraneous graphics that are not relevant to learning (e.g. a photo of a police officer
talking to a civilian next to a lesson on communication skills) can place unnecessary,
extraneous load on the learners WM resources (Clark and Mayer, 2008). As a result, it is
important for instructional designers to consider the detrimental effects that seemingly
harmless media elements can have on learning within the e-learning environment.
One way that CLT can be used to assist curriculum designers and trainers in the
policing context, is that it can prompt these individuals to consider (in a very explicit
way) whether the training that is being designed and delivered accords well with what
we now know about effective training. To facilitate this, we provide in Table I a series
of CLT-based questions that instructional designers/trainers should be asking
themselves when designing/delivering a training program for new recruits (i.e. officers
Goal/question Strategy |
Minimize extraneous load
Have I included a sufficient number of problems
in my course where I demonstrate, in a step-by-
step fashion, how to solve the problem being
Use worked examples to provide new
recruits with adequate guidance
Have I presented the same material in multiple
formats (e.g. text and diagram) when it is
understandable in one format alone?
Remove one of the redundant sources of
information (e.g. text or diagram)
Have I avoided splitting the trainees’ attention
between multiple pieces of non-redundant
Integrate related forms of essential
Is it too difficult to physically integrate related
sources of essential training information, or does
the instructional material appear cluttered when
Use audio in place of text to explain a
related diagram or animation
Maximize germane load
Have I provided trainees with a range of different
training demos in order to improve the transfer of
skills to the work environment?
Vary the surface features across the
training demos presented to the trainee
Have I explicitly prompted trainees to engage in a
deeper processing of the materials and
encouraged them to monitor their learning as it
progresses throughout the training program?
Ask trainees to self-explain important
features of the worked examples to
Manage intrinsic load
If the material is complex, have I adequately dealt
with how to present the material to recruits so it
does not overload their working memory?
Present the examples in parts followed by
the entire example once schemas have
If the material is complex, have I effectively
modified the learning environment to help
learner’s deal with the increased complexity?
Ask trainees to collaborate with one
another in order to distribute the working
Tabl e I.
CLT-inspired checklist for
who are not yet experts in the field of training). While not all of the components in
Table I will apply to training offered to more experienced police officers, it can be easily
extended to the training of new recruits within the traditional classroom setting in
addition to training provided in an e-learning environment.
For example, consider an e-learning scenario where new officers are learning basic
investigative skills, including how to: respond to a crime scene, take statements, collect
evidence, interview suspects, and generally manage the case. When planning this
e-course, instructional designers should first ask themselves whether they have taken
every measure to reduce extraneous, or unnecessary, WM load. For instance, have
they used worked examples when applicable (e.g. by demonstrating the steps involved
in the cognitive interview)? Similarly, have they removed redundant sources of
information (e.g. text explaining how to restrain a suspect at the crime scene is not
included in the training when an instructional animation has already been included
that adequately demonstrates the process)? Finally, have they ensured the trainee’s
attention is not being split between two related sources of necessary information
(e.g. text explaining how to properly complete a search warrant form is incorporated
into the picture of the search warrant or audio is explaining the correct procedure
instead of text)?
Once sources of extraneous load have been dealt with, instructional designers
should then consider incorporating strategies into their course that further enhance the
learning of basic investigative skills (i.e. strategies that increase germane load).
For instance, to improve the ability of trainees to use any new acquired skills while
on-the-job, instructors should ask themselves whether they have varied the surface
features of the worked examples they plan to give trainees (e.g. by demonstrating the
techniques used in the cognitive interview across a wide range of crimes, victim, and
perpetrator profiles)? Similarly, instructors should ensure that they have included
specific questions throughout the training lesson that encourage deep processing of the
instructional materials (e.g. prompting trainees to self-explain how to effectively take
statements from victims and witnesses at the crime scene)?
Moreover, many lessons covered in a course teaching officers basic investigative
skills are likely to be very high in intrinsic complexity (e.g. how to deal with suspects,
victims, and witnesses as the first responder to a crime scene, how to appropriately
collect evidence and ensure it is not contaminated, how to prepare a case for court, etc.).
Under these circumstances, instructional designers should consider whether further
precautions are necessary to ensure the trainee’s cognitive resources are not overloaded
at the outset of the course. For example, have they presented the instructional material
to the trainees in a way that reduces the initial burden placed on WM (e.g. by
demonstrating the steps of the cognitive interview one-by-one and then as a whole
process once trainees become familiar with each step)? Similarly, have they offered
learners additional resources to offset the inherent complexity of the material being
learned (e.g. by facilitating collaboration amongst trainees in an e-chat forum during
and after each lesson to reduce the load placed on any one trainee’s WM)?
Finally, instructors should also ensure that they have taken into account the
trainee’s developing schemas throughout the course. In other words, have they taken
measures to effectively combat the expertise reversal effect? One way to do this is to
continually monitor the trainee’s learning over the course of the training program
(i.e. via short evaluations), fading from worked examples to completion problems
when the evaluations demonstrate that learning is occurring (Kalyuga and Sweller,
2004). Eventually, as learning further progresses, trainees can be given traditional
problem-based learning tasks in order to acquire all the necessary skills. Similarly, in
later stages of the course, instructors may wish to use mental rehearsal instead of
self-explanation techniques to facilitate deeper processing of the instructional
materials. Modifying the strategies used to accommodate the experience level of
trainee’s ensures that learning will continue to evolve throughout the entire duration of
the training course.
5. Aspects of CLT that require further research
Despite the potential value of CLT as a framework to improve police training, there
are still aspects of the theory that require further research. While it is clear from the
research reviewed above that an abundance of empirical support has been found for
the instructional effects stemming from CLT, some criticisms of CLT do exist.
These criticisms tend not to focus on the validity of the effects themselves, but rather
on the explanations provided for why these effects emerge (Beckmann, 2010; de Jong,
2010; Schnotz and Ku
The majority of CLT criticisms fall into two categories: conceptual and
methodological. Conceptually, the major criticism levelled against CLT relates to
the difficulty in distinguishing between the different types of cognitive load that are
thought to be important in training contexts, especially extraneous and germane load.
For example, de Jong (2010) makes the point that, conceptually, extraneous and
germane load can only be determined in a post hoc fashion, after the effect of training is
observed (i.e. load is defined as extraneous if the training resulted in poor learning
outcomes, but germane if the training resulted in good learning outcomes). Questions
have also been raised by critics about whether the three types of cognitive load
discussed in CLT are in fact additive and about the concept of cognitive load itself
(e.g. whether cognitive load and cognitive effort are synonymous as assumed in some
CLT research) (de Jong, 2010).
The most common methodological criticism of CLT is that no standard procedure
currently exists for measuring (and distinguishing between) the various types of
cognitive load experienced by trainees (Beckmann, 2010; de Jong, 2010; Schnotz
¨rschner, 2007). While physiological measures of cognitive load are available
(van Gog et al., 2009), the most commonly used measures consist of self-report
questionnaires (e.g. where learners indicate the amount of mental effort exerted during
training; Paas, 1992). However, as de Jong (2010) highlights, there is no standard format
to these questionnaires. Other methodological issues that have been raised relate to the
lack of attention paid by researchers to important individuals difference variables that
may influence how cognitive load is experienced by learners (e.g. WM capacity)
and the low level of ecological validity associated with some CLT studies, which
raises questions about the generalizability of the results to realistic training
environments (de Jong, 2010).
Similar to police training itself, CLT is continually evolving and ongoing research
will help to address these concerns. While more work clearly needs to be done to better
understand the nature of the CLT effects described above, it also seems clear that
empirical support for these effects exists. This fact highlights the potential for
CLT-based instructional strategies to significantly improve the quality of police
training. Indeed, while there have been concerns raised about CLT-effects potentially
not generalizing beyond the research laboratory (e.g. de Jong, 2010), the CLT-effects we
have focused on do frequently extend to naturalistic settings where authentic training
is delivered, such as the classroom setting (e.g. Carroll, 1994; Harskamp et al., 2007;
Kissane et al., 2008; Ward and Sweller, 1990; Zhu and Simon, 1987). It is difficult
to think of a reason why this would not also be true for training delivered in the
6. Where does police training go from here?
Given the extensive empirical support that exists for the instructional effects of
CLT across a variety of complex cognitive domains, it is clear that this theoretical
framework could be valuable to the area of police training. Although promising, CLT
has yet to be integrated into police training theory and practice as it currently exists.
The ultimate goal of this paper was to provide some concrete illustrations of how CLT
training strategies can complement, and potentially enhance, traditional theoretical
approaches to police training.
The few examples provided above clearly demonstrate that CLT principles can be
applied to some of the current methods of police training in a relatively easy fashion,
but whether CLT-informed instruction leads to significant improvements in learning
and transfer in this domain (as compared to traditional forms of instruction) certainly
remains an empirical question. Our hope is that this paper will open up a discourse
between academic researchers, instructional designers, and police trainers so that the
potential benefits of integrating CLT principles into police training may be realized.
By implementing some of the recommendations offered in this paper, we are hopeful
that future research will demonstrate that CLT can be used to improve both the
efficiency and effectiveness of police training programs.
1. Although we chose e-learning to illustrate how CLT-based strategies can be implemented in
the police training context, it is important to note that the same CLT principles can be
extended to classroom-based training (Clark et al., 2006; Clark and Mayer, 2008).
2. With that said, some scholars have noted that for certain police training courses (e.g.
firearms and defensive training), a pedagogical or instructor-driven approach will continue
to be most appropriate (Birzer, 2003).
3. Given the importance we have placed on multimodal learning styles, we would like to stress
that removing sources of information from training (to reduce redundancy) does not mean
that a multimodal instructional strategy cannot be adopted. Indeed, such an instructional
strategy can still be used by trainers so long as it is neither redundant nor splits the learners
attention across multiple sources of information at any given time. That is, multiple modes of
presentation (e.g. an audio explanation of a diagram) can be used to target different learning
styles, as long as they are not providing completely redundant information and they are both
presented to the trainee at the same time.
4. For a more detailed explanation of criticisms against CLT, please see de Jong (2010).
Anderson, L.W., Krathwohl, D.R., Airasian, P.W., Cruikshank, K.A., Mayer, R.E., Pintrich, P.R.,
Raths, J. and Wittrock, M.C. (2000), A Taxonomy for Learning, Teaching, and Assessing:
A Revision of Bloom’s Taxonomy of Educational Objectives, Longman, White Plains, NY.
Atkinson, R.K. and Renkl, A. (2007), “Interactive example-based learning environments: using
interactive elements to encourage effective processing of worked examples”, Educational
Psychology Review, Vol. 19 No. 3, pp. 375-386.
Atkinson, R.K., Renkl, A. and Merrill, M.M. (2003), “Transitioning from studying examples to
solving problems: effects of self-explanation prompts and fading worked-out steps”,
Journal of Educational Psychology, Vol. 95 No. 4, pp. 774-783.
Ayres, P. (2006), “Impact of reducing intrinsic cognitive load on learning in a mathematical
domain”, Applied Cognitive Psychology, Vol. 20 No. 3, pp. 287-298.
Ayres, P. and Paas, F. (2007), “Making instructional animations more effective: a cognitive load
approach”, Applied Cognitive Psychology, Vol. 21 No. 6, pp. 695-700.
Ayres, P. and Paas, F. (2009), “Interdisciplinary perspectives inspiring a new generation of
cognitive load research”, Educational Psychology Review, Vol. 21 No. 1, pp. 1-9.
Ayres, P., Marcus, N., Chan, C. and Qian, N. (2009), “Learning hand manipulative tasks: when
instructional animations are superior to equivalent static representations”, Computers in
Human Behavior, Vol. 25 No. 2, pp. 348-353.
Baddeley, A. (1992), “Working memory”, Science, Vol. 255 No. 5044, pp. 556-559.
Bahadori, M., Sadeghifar, J., Tofighi, S., Mamikhani, J.A. and Nejati, M. (2011), “Learning styles of
the health services management students: a study of first-year students from the medical
science universities of Iran”, Australian Journal of Basic and Applied Sciences, Vol. 5 No. 9,
Barrows, H.S. and Tamblyn, R.M. (1980), Problem-Based Learning: An Approach to Medical
Education, Springer Publishing Company, New York, NY.
Beckmann, J.F. (2010), “Taming a beast of burden – on some issues with the conceptualisation of
cognitive load”, Learning and Instruction, Vol. 20 No. 3, pp. 250-264.
Bennell, C., Jones, N.J. and Corey, S. (2007), “Does use-of-force simulation training in Canadian
police agencies incorporate principles of effective training?”, Psychology, Public Policy and
Law, Vol. 13 No. 1, pp. 35-58.
Bielaczyc, K., Pirolli, P.L. and Brown, A.L. (1995), “Training in self-explanation and
self-regulation strategies: investigating the effects of knowledge acquisition activities on
problem solving”, Cognition and Instruction, Vol. 13 No. 2, pp. 221-252.
Birzer, M.L. (2003), “The theory of andragogy applied to police training”, Policing: An
International Journal of Police Strategies & Management, Vol. 26 No. 1, pp. 29-42.
Birzer, M.L. and Tannehill, R. (2001), “A more effective training approach for contemporary
policing”, Police Quarterly, Vol. 4 No. 2, pp. 233-252.
Bloom, B.S., Engelhart, M.D., Furst, E.J., Hill, W.H. and Krathwohl, D.R. (1956), Taxonomy of
Educational Objectives: Book 1 Cognitive Domain, Longman, White Plains, NY.
Bower, G.H. (1972), “Mental imagery and associative learning”, in Gregg, L. (Ed.), Cognition in
Learning and Memory, Wiley, New York, NY, pp. 51-88.
Breckler, J., Joun, D. and Ngo, H. (2009), “Learning styles of physiology students interested in the
health professions”, Advances in Physiology Education, Vol. 33 No. 1, pp. 30-36.
Brookfield, S.D. (1986), Understanding and Facilitating Adult Learning: A Comprehensive
Analysis of Principles and Effective Practice, Jossey-Bass, San Francisco, CA.
¨nken, R., Plass, J.L. and Leutner, D. (2004), “Assessment of cognitive load in multimedia
learning with dual-task methodology: auditory load and modality effects”, Instructional
Science, Vol. 32 No. 2, pp. 115-132.
Busch, C., Renkl, A. and Schworm, S. (2008), “Towards a generic self-explanation training
intervention for example-based learning”, in Kanselaar, G., Jonker, V., Kirschner, P.A. and
Prins, F.J. (Eds), Proceedings of the 8th International Conference of the Learning Sciences
2008, ICLS, Utrecht.
Canadian Police Knowledge Network (2008), “Police sector council funds national e-learning
initiative”, available at: www.cpkn.ca/node/100 (accessed 1 September 2008).
Canadian Police Knowledge Network (2009), “PSC invests in three year national e-learning
program”, available at: www.cpkn.ca/node/121 (accessed 16 October 2009).
Carroll, W.M. (1994), “Using worked examples as an instructional support in the algebra
classroom”, Journal of Educational Psychology, Vol. 86 No. 3, pp. 360-367.
Chandler, P. and Sweller, J. (1991), “Cognitive load theory and the format of instruction”,
Cognition and Instruction, Vol. 8 No. 4, pp. 293-332.
Chandler, P. and Sweller, J. (1992), “The split-attention effect as a factor in the design of
instruction”, British Journal of Educational Psychology, Vol. 62 No. 2, pp. 233-246.
Chi, M.T.H., de Leeuw, N., Chiu, M. and LaVancher, C. (1994), “Eliciting self-explanations improves
understanding”, Cognitive Science: A Multidisciplinary Journal,Vol.18No.3,pp.439-477.
Clark, R., Nguyen, F. and Sweller, J. (2006), Efficiency in Learning: Evidence-based Guidelines to
Manage Cognitive Load, Pfeiffer, San Francisco, CA.
Clark, R.C. and Mayer, R.E. (2008), E-learning and the Science of Instruction: Proven Guidelines
for Consumers and Designers of Multimedia Learning, 2nd ed., Pfeiffer, San Francisco, CA.
Clarke, T., Ayres, P. and Sweller, J. (2005), “The impact of sequencing and prior knowledge in
learning mathematics through spreadsheet application”, Educational Technology Research
and Development, Vol. 53 No. 3, pp. 15-24.
Cleveland, G. (2006), “Using problem-based learning in police training”, The Police Chief, Vol. 73
No. 11, available at: www.policechiefmagazine.org/magazine/index.cfm?fuseaction ¼
display_arch&article_id ¼1051&issue_id ¼112006 (accessed 2 February 2012).
Cooper, G. and Sweller, J. (1987), “Effects of schema acquisition and rule automation on
mathematical problem-solving transfer”, Journal of Educational Psychology, Vol. 79 No. 4,
Cooper, G., Tindall- Ford, S., Chandler, P. and Sweller, J. (2001), “Learning by imagining”, Journal
of Experimental Psychology: Applied, Vol. 7 No. 1, pp. 68-82.
Cooper, G.A. (1998), “Research into cognitive load theory and instructional design at UNSW”,
available at: http://dwb4.unl.edu/Diss/Cooper/UNSW.htm (accessed 1 September 2008).
de Croock, M.B.M., van Merrie
¨nboer, J.J.G. and Paas, F.G.W.C. (1998), “High versus low contextual
interference in simulation-based training of troubleshooting skills: effects on transfer
performance and invested mental effort”, Computers in Human Behavior, Vol. 14 No. 2,
European Police College (2012), “E-learning”, available at: www.cepol.europa.eu/index.
php?id ¼e-learning (accessed 3 February 2012).
Gick, M.L. and Holyoak, K.J. (1983), “Schema induction and analogical transfer”, Cognitive
Psychology, Vol. 15 No. 1, pp. 1-38.
Ginns, P. (2005), “Meta-analysis of the modality effect”, Learning and Instruction, Vol. 15 No. 4,
Ginns, P., Chandler, P. and Sweller, J. (2003), “When imagining information is effective”,
Contemporary Educational Psychology, Vol. 28 No. 2, pp. 229-251.
Glasgow, C. and Lepatski, C. (2012), “The evolution of police training: the investigative skill
education program”, in Haberfeld, M.R., Clarke, C.A. and Sheehan, D.L. (Eds), Police
Organization and Training: Innovations in Research and Practice, Springer, New York, NY,
Große, C.S. and Renkl, A. (2006), “Effects of multiple solution methods in mathematics learning”,
Learning and Instruction, Vol. 16 No. 2, pp. 122-138.
Groeger, J.A. (1997), Memory and Remembering: Everyday Memory in Context, Longmans,
Harskamp, E., Mayer, R.E., Suhre, C. and Jansma, J. (2007), “Does the modality principle for
multimedia learning apply to science classrooms?”, Learning and Instruction, Vol. 17 No. 5,
Hilbert, T. and Renkl, A. (2009), “Learning how to use a computer-based concept-mapping tool:
self-explaining examples helps”, Computers in Human Behavior, Vol. 25 No. 2, pp. 267-274.
¨ffler, T.N. and Leutner, D. (2007), “Instructional animation versus static pictures: a meta-
analysis”, Learning and Instruction, Vol. 17 No. 6, pp. 722-738.
Interpol (2012), “E-learning catalogue”, available at: www.interpol.int/INTERPOL-expertise/
e-Learning-Catalogue (accessed 7 February 2012).
Johnson, M. (2009), “Evaluation of learning style for first year medical students”, International
Journal for the Scholarship of Teaching and Learning, Vol. 3 No. 1.
de Jong, T. (2010), “Cognitive load theory, educational research, and instructional design: some
food for thought”, Instructional Science, Vol. 38 No. 2, pp. 105-134.
Justice Institute of British Columbia (2010), “Exercise design”, available at: www.jibc.ca/
management-division/applied-training/exercise-design (accessed 20 December 2010).
Kalyuga, S. (2007), “Enhancing instructional efficiency of interactive e-learning environments:
a cognitive load perspective”, Educational Psychology Review, Vol. 19 No. 3, pp. 387-399.
Kalyuga, S. and Sweller, J. (2004), “Measuring knowledge to optimize cognitive load factors
during instruction”, Journal of Educational Psychology, Vol. 96 No. 3, pp. 558-568.
Kalyuga, S., Chandler, P. and Sweller, J. (1999), “Managing split-attention and redundancy in
multimedia instruction”, Applied Cognitive Psychology, Vol. 13 No. 4, pp. 351-371.
Kalyuga, S., Chandler, P. and Sweller, J. (2000), “Incorporating learner experience into the design
of multimedia instruction”, Journal of Educational Psychology, Vol. 92 No. 1, pp. 126-136.
Kalyuga, S., Ayres, P., Chandler, P. and Sweller, J. (2003), “The expertise reversal effect”,
Educational Psychologist, Vol. 38 No. 1, pp. 23-31.
Kalyuga, S., Chandler, P., Tuovinen, J. and Sweller, J. (2001), “When problem solving is superior to
studying worked examples”, Journal of Educational Psychology, Vol. 93 No. 3, pp. 579-588.
Kester, L., Lehnen, C., Van Gerven, P.W.M. and Kirschner, P.A. (2006), “Just-in-time, schematic
supportive information presentation during cognitive skill acquisition”, Computers in
Human Behavior, Vol. 22 No. 1, pp. 93-112.
Kirschner, F., Paas, F. and Kirschner, P.A. (2009a), “A cognitive load approach to collaborative
learning: united brains for complex tasks”, Educational Psychology Review, Vol. 21 No. 1,
Kirschner, F., Paas, F. and Kirschner, P.A. (2009b), “Individual and group-based learning from
complex cognitive tasks: effects on retention and transfer efficiency”, Computers in Human
Behavior, Vol. 25 No. 2, pp. 306-314.
Kirschner, P.A. (2002), “Cognitive load theory: implications of cognitive load theory on the design
of learning”, Learning and Instruction, Vol. 12 No. 1, pp. 1-10.
Kissane, M., Kalyuga, S., Chandler, P. and Sweller, J. (2008), “The consequences of fading
instructional guidance on delayed performance: the case of financial services training”,
Educational Psychology, Vol. 28 No. 7, pp. 809-822.
Knowles, M.S. (1980), The Modern Practice of Adult Education: From Pedagogy to Andragogy,
Follett Publishing Company, Chicago, IL.
Knowles, M.S. (1990), The Adult Learner: A Neglected Species, Gulf Publishing Company,
Krathwohl, D.R., Bloom, B.S. and Masia, B.B. (1964), Taxonomy of Educational Objectives: Book 2
Affective Domain, Longman, White Plains, NY.
Landry, J.M. (2011), “Learning styles of law enforcement officers: does police work affect how
officers learn?”, unpublished doctoral dissertation, Capella University, Minneapolis, MN.
Leahy, W. and Sweller, J. (2004), “Cognitive load and the imagination effect”, Applied Cognitive
Psychology, Vol. 18 No. 7, pp. 857-875.
Leahy, W. and Sweller, J. (2005), “Interactions among the imagination, expertise reversal, and
element interactivity effects”, Journal of Experimental Psychology: Applied, Vol. 11 No. 4,
Leahy, W. and Sweller, J. (2008), “The imagination effect increases with an increased intrinsic
cognitive load”, Applied Cognitive Psychology, Vol. 22 No. 2, pp. 273-283.
Leahy, W., Chandler, P. and Sweller, J. (2003), “When auditory presentations should and should
not be a component of multimedia instruction”, Appl ied Cognitive Psychology, Vol. 17 No. 4,
McCoy, M.R. (2006), “Teaching style and the application of adult learning principles by police
instructors”, Policing: An International Journal of Police Strategies & Management, Vol. 29
No. 1, pp. 77-91.
Mayer, R.E. (1997), “Multimedia learning: are we asking the right questions?”, Educational
Psychologist, Vol. 32 No. 1, pp. 1-19.
¨nboer, J.J.G. and Ayres, P. (2005), “Research on cognitive load theory and its design
implications for e-learning”, Educational Technology Research and Development, Vol. 53
No. 3, pp. 5-13.
¨nboer, J.J.G. and de Croock, M.B.M. (1992), “Strategies for computer-based
programming instruction: program completion vs program generation”, Journal of
Educational Computing Research, Vol. 8 No. 3, pp. 365-394.
¨nboer, J.J.G. and Sluijsmans, D.M.A. (2009), “Toward a synthesis of cognitive load
theory, four-component instructional design, and self-directed learning”, Educational
Psychology Review, Vol. 21 No. 1, pp. 55-66.
¨nboer, J.J.G. and Sweller, J. (2005), “Cognitive load theory and complex learning:
recent developments and future directions”, Educational Psychology Review, Vol. 17 No. 2,
¨nboer, J.J.G. and Sweller, J. (2010), “Cognitive load theory in health professional
education: design principles and strategies”, Medical Education, Vol. 44 No. 1, pp. 85-93.
¨nboer, J.J.G., Kester, L. and Paas, F. (2006), “Teaching complex rather than simple
tasks: balancing intrinsic and germane load to enhance transfer of learning”, Applied
Cognitive Psychology, Vol. 20 No. 3, pp. 343-352.
¨nboer, J.J.G., Kirschner, P.A. and Kester, L. (2003), “Taking the load off a learners’
mind: instructional design for complex learning”, Educational Psychologist, Vol. 38 No. 1,
Miller, G.A. (1956), “The magical number seven, plus or minus two: some limits on our capacity
for processing information”, Psychological Review, Vol. 63 No. 2, pp. 81-97.
Moreno, R. (2004), “Decreasing cognitive load for novice students: effects of explanatory versus
corrective feedback in discovery-based multimedia”, Instructional Science, Vol. 32 Nos 1-2,
Moreno, R. and Mayer, R. (1999), “Cognitive principles of multimedia learning: the role of
modality and contiguity”, Journal of Educational Psychology, Vol. 91 No. 2, pp. 358-368.
Murphy, R.J., Gray, S.A., Straja, S.R. and Bogert, M.C. (2004), “Student learning preferences and
teaching implications”, Journal of Dental Education, Vol. 68 No. 8, pp. 859-866.
Paas, F. and van Gog, T. (2006), “Optimising worked example instruction: different ways to
increase germane cognitive load”, Learning and Instruction, Vol. 16 No. 2, pp. 87-91.
Paas, F., Renkl, A. and Sweller, J. (2003), “Cognitive load theory and instructional design: recent
developments”, Educational Psychologist, Vol. 38 No. 1, pp. 1-4.
Paas, F.G.W.C. (1992), “Training strategies for attaining transfer of problem-solving skills in
statistics: a cognitive load approach”, Journal of Educational Psychology, Vol. 84 No. 4,
Paas, F.G.W.C. and van Merrie
¨nboer, J.J.G. (1994), “Variability of worked examples and transfer of
geometrical problem-solving skills: a cognitive-load approach”, Journal of Educational
Psychology, Vol. 86 No. 1, pp. 122-133.
Palmiotto, M.J., Birzer, M.L. and Unnithan, N.P. (2000), “Training in community policing: a
suggested curriculum”, Policing: An International Journal of Police Strategies &
Management, Vol. 23 No. 1, pp. 8-21.
Pawley, D., Ayres, P., Cooper, M. and Sweller, J. (2005), “Translating words into equations: a
cognitive load theory approach”, Educational Psychology, Vol. 25 No. 1, pp. 75-97.
Pitts, S., Glensor, R.W. and Peak, K.J. (2007), “The police training officer (PTO) program: a
contemporary approach to postacademy recruit training”, The Police Chief, Vol. 74 No. 8,
available at: www.policechiefmagazine.org/magazine/index.cfm?fuseaction ¼display_
arch&article_id ¼1247&issue_id ¼82007 (accessed 2 February 2012).
Police Sector Council (PSC) (2010), “The state of e-learning in Canadian policing”, available at:
Policing_Final.pdf (accessed 5 February 2012).
Pollock, E., Chandler, P. and Sweller, J. (2002), “Assimilating complex information”, Learning and
Instruction, Vol. 12 No. 1, pp. 61-86.
Quilici, J.L. and Mayer, R.E. (1996), “Role of examples in how students learn to categorize
statistics word problems”, Journal of Educational Psychology, Vol. 88 No. 1, pp. 144-161.
Reisslein, J., Atkinson, R.K., Seeling, P. and Reisslein, M. (2006), “Encountering the expertise
reversal effect with a computer-based environment on electrical circuit analysis”, Learning
and Instruction, Vol. 16 No. 2, pp. 92-103.
Renkl, A. (1997), “Learning from worked-out examples: a study on individual differences”,
Cognitive Science, Vol. 21 No. 1, pp. 1-29.
Renkl, A. and Atkinson, R.K. (2003), “Structuring the transition from example study to problem
solving in cognitive skill acquisition: a cognitive load perspective”, Educational
Psychologist, Vol. 38 No. 1, pp. 15-22.
Renkl, A., Atkinson, R.K. and Grosse, C.S. (2004), “How fading worked solution steps works: a
cognitive load perspective”, Instructional Science, Vol. 32 Nos 1-2, pp. 59-82.
Renkl, A., Stark, R., Gruber, H. and Mandl, H. (1998), “Learning from worked-out examples: the
effects of example variability and elicited self-explanations”, Contemporary Educational
Psychology, Vol. 23 No. 1, pp. 90-108.
Retnowati, E., Ayres, P. and Sweller, J. (2010), “Worked example effects in individual and group
work settings”, Educational Psychology, Vol. 30 No. 3, pp. 349-367.
Ross, C. (2009), “Technologies enhance cadet training”, Gazette, Vol. 71 No. 4, p. 4, available at:
www.rcmp-grc.gc.ca/gazette/vol70n4/cadet-eng.htm (accessed 27 August 2009).
Rourke, A. and Sweller, J. (2009), “The worked-example effect using ill-defined problems: learning to
recognise designers’ styles”, Learning and Instruction, Vol. 19 No. 2, pp. 185-199.
Roy, M. and Chi, M.T.H. (2005), “The self-explanation principle in multimedia learning”, in Mayer, R.E.
(Ed.), The Cambridge Handbook of Multimedia Learning, Cambridge University Press,
New York, NY, pp. 271-286.
Schnotz, W. and Ku
¨rschner, C. (2007), “A reconsideration of cognitive load theory”, Educational
Psychology Review, Vol. 19 No. 4, pp. 469-508.
Schwonke, R., Renkl, A., Krieg, C., Wittwer, J., Aleven, V. and Salden, R. (2009), “The worked-
example effect: not an artifact of lousy control conditions”, Computers in Human Behavior,
Vol. 25 No. 2, pp. 258-266.
Seufert, T., Schutze, M. and Bru
¨nken, R. (2009), “Memory characteristics and modality in
multimedia learning: an aptitude-treatment-interaction study”, Learning and Instruction,
Vol. 19 No. 1, pp. 28-42.
Simpson, E.J. (1972), The Classification of Educational Objectives in the Psychomotor Domain,
Gryphon House, Washington, DC.
Sorden, S.D. (2005), “A cognitive approach to instructional design for multimedia learning”,
Informing Science, Vol. 8, pp. 263-279, available at: www.informingscience.us/icarus/
Sprenger, M. (2003), Differentiation Through Learning Styles and Memory, Corwin Press,
Thousand Oaks, CA.
Sweller, J. (1988), “Cognitive load during problem solving: effects on learning”, Cognitive Science,
Vol. 12 No. 2, pp. 257-285.
Sweller, J. (1994), “Cognitive load theory, learning difficulty and instructional design”, Learning
and Instruction, Vol. 4 No. 4, pp. 295-312.
Sweller, J. (2006), “The worked example effect and human cognition”, Learning and Instruction,
Vol. 16 No. 2, pp. 165-169.
Sweller, J. and Chandler, P. (1994), “Why some material is difficult to learn”, Cognition and
Instruction, Vol. 12 No. 3, pp. 184-233.
Sweller, J., van Merrie
¨nboer, J.J.G. and Paas, F.G.W.C. (1998), “Cognitive architecture and
instructional design”, Educational Psychology Review, Vol. 10 No. 3, pp. 251-296.
Tennyson, R.D. and Elmore, R. (1997), “Learning theory foundations for instructional design”,
in Tennyson, R.D., Schott, F., Seel, N.M. and Dijkstra, S. (Eds), Instructional Design:
International Perspectives, Erlbaum, Mahwah, NJ, pp. 55-78.
Tindall-Ford, S. and Sweller, J. (2006), “Altering the modality of instruction to facilitate
imagination: interactions between the modality and imagination effects”, Instructional
Science, Vol. 34 No. 4, pp. 343-365.
Tindall-Ford, S., Chandler, P. and Sweller, J. (1997), “When two sensory modes are better than
one”, Journal of Experimental Psychology: Applied, Vol. 3 No. 4, pp. 257-287.
Van Gog, T., Kester, L., Nievelstein, F., Giesbers, B. and Paas, F. (2009), “Uncovering cognitive
processes: different techniques that can contribute to cognitive load research and
instruction”, Computers in Human Behavior, Vol. 25 No. 2, pp. 325-331.
van Gog, T. and Paas, F. (2008), “Instructional efficiency: revisiting the original construct in
educational research”, Educational Psychology, Vol. 43 No. 1, pp. 16-26.
van Gog, T., Paas, F. and van Merrie
¨nboer, J.J.G. (2006), “Effects of process-oriented worked examples
on troubleshooting transfer performance”, Learning and Instruction, Vol. 16 No. 2, pp. 154-164.
Vodde, R.F. (2009), Adragogical Instruction for Effective Police Training, Cambria Press,
Vodde, R.F. (2012), “Changing paradigms in police training: transitioning from a traditional to an
andragogical model”, in Haberfeld, M.R., Clarke, C.A. and Sheehan, D.L. (Eds), Police
Organization and Training: Innovations in Research and Practice, Springer, New York, NY,
Ward, M. and Sweller, J. (1990), “Structuring effective worked examples”, Cognition and
Instruction, Vol. 7 No. 1, pp. 1-39.
Werth, E.P. (2011), “Scenario training in police academies: developing students’ higher-level
thinking skills”, Police Practice and Research, Vol. 12 No. 4, pp. 325-340.
Willis, R.C. (2010), “Improving training through Bloom’s taxonomy: why police instructors
should avoid a cookie-cutter approach to training new officers”, available
Blooms-Taxonomy/ (accessed 2 February 2012).
Wong, R.M.F., Lawson, M.J. and Keeves, J. (2002), “The effects of self-explanation training on
students’ problem solving in high-school mathematics”, Learning and Instruction, Vol. 12
No. 2, pp. 233-262.
Zhu, X. and Simon, H.A. (1987), “Learning mathematics from examples and by doing”, Cognition
and Instruction, Vol. 4 No. 3, pp. 137-166.
DeLeeuw, K.E. and Mayer, R.E. (2008), “A comparison of three measures of cognitive load:
evidence for separable measures of intrinsic, extraneous, and germane load”, Journal of
Educational Psychology, Vol. 100 No. 1, pp. 223-234.
Gerjets, P., Scheiter, K. and Cierniak, C. (2009), “The scientific value of cognitive load theory: a
research agenda based on the structuralist view of theories”, Educational Psychology
Review, Vol. 21 No. 1, pp. 43-54.
Kalyuga, S., Chandler, P. and Sweller, J. (1998), “Levels of expertise and instructional design”,
Human Factors, Vol. 40 No. 1, pp. 1-17.
Paas, F., Renkl, A. and Sweller, J. (2004), “Cognitive load theory: instructional implications of the
interaction between information structures and cognitive architecture”, Instructional
Science, Vol. 32 Nos 1-2, pp. 1-8.
Peterson, L. and Peterson, M. (1959), “Short-term retention of individual verbal items”, Journal of
Experimental Psychology, Vol. 58 No. 3, pp. 193-198.
About the author
Rebecca Mugford is currently a second year doctoral student in Psychology at Carleton
University in Ottawa, Ontario, Canada. Her research focuses on a variety of policing issues,
including: the evaluation of police recruitment strategies, the validity of psychologically-based
investigative techniques (e.g. case linkage analysis), and the use of psychological theories to
inform police practice. Rebecca Mugford is the corresponding author and can be contacted at:
Shevaun Corey is currently completing her last year as a doctoral student in Psychology at
Carleton University in Ottawa, Ontario, Canada. Her research focuses mainly on determining the
effectiveness of training in criminal justice settings (e.g. police, military and corrections) and the
application of learning theories (e.g. CLT) to training in these settings. In addition, her research
has focused on a variety of policing issues, including: police training, police decision-making and
the validity of psychologically-based investigative techniques (e.g. offender and geographic
profiling). Shevaun Corey can be contacted at: firstname.lastname@example.org
Craig Bennell is an Associate Professor of Psychology at Carleton University in Ottawa,
Ontario, Canada, where he also runs the Police Research Lab. In his current research, he
examines two issues: the reliability, validity, and usefulness of psychologically-based
investigative techniques, such as criminal profiling and case linkage analysis; and factors that
influence the quality of police decision making in use-of-force encounters, including suspect,
officer, and environmental factors. Craig Bennell can be contacted at: email@example.com
To purchase reprints of this article please e-mail: firstname.lastname@example.org
Or visit our web site for further details: www.emeraldinsight.com/reprints