The implications of generative learning strategies for integrating cognitive load and self-

regulation theory into educational innovations

Paper presented at 24th International Congress of School Effectiveness and Improvement

Limassol, Cyprus

Gary Morrison, Professor

Instructional Design and Technology

Old Dominion University

John Nunnery, Executive Director

The Center for Educational Partnerships

and Associate Professor of Educational Research

Old Dominion University

January 2011

The typical textbook used in American classrooms, whether first grade or in college, have

not significantly changed in format or design since the invention of the Gutenberg press over 500

years ago. Moreover, traditional textbooks are not interactive or adaptive to students’ needs,

despite research which suggests that these features may be beneficial for learning (Lee & Park,

2008; Mayer et al., 2003; Park, 1996). In this paper we presented a research and development

paradigm (AdapText Interactive) that responds to the need for improved textbook design and

support for student comprehension. It expands previous research by creating larger portions of

experimental text than has typically been done before, in a domain area (Earth science) that has

been infrequently used in this type of research. As interest moves from printed books to

electronic books, there is an opportunity to design a new generation of interactive textbooks and

to move far beyond the first generation of ‘soft’ versions of the printed books. The first project

we are undertaking project will take the first steps in the design of an innovative teaching

technology that is based on research in instructional technology and educational psychology, to

develop and test interactive i-textbook chapters that can be used as a model for the development

of second generation i-textbooks.

The goals of the project arise from considering several areas of need in STEM education

and textbook design. It is imperative that K-12 students have STEM education experiences

which allow them to form a deep understanding of scientific principles and content. Such

experiences promote the level of science literacy necessary for informed and responsible

citizenship. They also allow students to begin pursuit of STEM-related career training. The focus

of the project is middle school Earth science. In particular, the pedagogical resource to be

developed will support students as they construct and reason with mental models of complex

geologic structures and processes. Mental models are defined as simplified representations of

connected system components or causal mechanisms, that allow the learner to make predictions,

simulate outcomes and solve problems (Greca 2000, Mayer, 2008; Mayer & Chandlger, 2001;

Mayer & Moreno, 2003). The development of this level of understanding is critical for the

individual and their role in society. For example, as a society we need students to pursue science-

related careers in order to address and further understand these and other issues. A the same time,

in order to participate in public discourse about issues such as global warming and the strategies

that might be taken to counter it, the connections between marine and atmospheric conditions,

and how global warming may impact the resources that sustain life, students need to understand

Earth’s geological systems.

We believe that significant opportunities for progress towards meeting these needs exist at the

middle and high school levels. From a developmental perspective, middle and high school is a

critical period for building a knowledge base and deepening students’ appreciation for the

scientific worldview. It is also a time when students’ cognitive development is such that for

some, formal operations are achieved (Brown & Carniff, 2007). Hypothetical reasoning and

abstract thinking are possible, and students can create and utilize simulations to help themselves

understand relationships among system components. Thus, middle and high school students are

able to develop relatively sophisticated causal models of complex phenomena. From a curricular

perspective, students are expected to grasp complex ideas at a more sophisticated level than at

elementary school. Students are, for example, expected to understand that “chemical pollution

and sedimentation are great threats to the chemical and biological well-being of estuaries and

oceans” (Virginia Earth Science Standard of Learning 11b,e). Mastering this idea requires the

student to build a mental representation of factors which affect water quality and marine life and

to be able to simulate causes and consequences of changes in the chemical and biological

makeup of waterways.

From a pedagogical perspective, students are expected to be able to learn effectively from

informational text. While students in middle school science can and do engage with hands-on

materials, textbooks remain the predominant form of instructional resource to the extent that

Heinich (1984) refers to a symbiotic relationship between the teacher and the textbook that

requires the teacher to translate the textbook into instruction. For students with effective reading

comprehension skills and self-regulated learning strategies, informational text provides a way of

deepening and consolidating curricular concepts. There are many ways, however, in which

students can be deficient in their use of comprehension strategies and often times they fail to

adequately comprehend informational text (Winograd & Johnston, 1982). In a science lesson,

time is often devoted to supporting and promoting students’ comprehension skills. This presents

a significant challenge for teachers of diverse groups of learners.

The goals of the first AdapText Interactive: i-EARTH project are to:

1. Re-design selected chapters of Earth science textbooks in order to reduce extraneous

cognitive load

2. Assess learning outcomes associated with electronic versus paper Earth science textbook

formats

3. Demonstrate the learning outcomes associated with interactive animation and simulation

components in electronic Earth science textbook materials

4. Incorporate and assess the effect of adaptive support for students’ use of learning

strategies and comprehension monitoring processes when using i-textbook materials

5. Identify and account for key learner variables that influence the effectiveness of i-

textbook features such as animations, simulations, strategy prompts and feedback.

Research and Development Design

Hypothesis. The hypothesis guiding the project is that middle school students’ learning of

Earth science topics can be enhanced with the use of adaptive, interactive textbook materials

designed in accordance with cognitive load, multimedia learning and self-regulated learning

theories. We plan to explore this hypothesis through a series of pilot studies, factorial

experiments in ecologically valid settings, and focus group opportunities that ultimately will

result in the development of an innovative technological resource – the interactive, adaptive,

supportive middle school science i-textbook. In addition, prototypes of the materials will be

tested in a controlled lab setting where physiological measures (e.g., eye tracking and

electroencephalogram (EEG) data are collected to help us understand how students process text

and pictures. These features are represented in the title of the project: AdapText Interactive

signifies the adaptive and interactive nature of the text. i-EARTH stands for Interactive

Electronic Adaptive Resources That Heighten science mastery.

Theoretical background. We will carefully, and with empirical data to support decision making

throughout, merge cognitive load theory with principles of multimedia learning, self-regulated

and generative learning theories with recent technological innovations. Through the use of a

cross platform development package, we will create a platform which allows assessment of the

effect of integrated graphics, animations, and simulations. However, optimizing learning in this

environment also depends on consideration of the learner’s unique psychological attributes

(Greene, Costa, Robertson, Pan, & Deekens, 2010 ). Specifically, learner processing of

multimedia components of electronically-enhanced learning materials varies according to prior

knowledge, generative learning strategy use, self-regulated learning skills, and particularly

spatial ability (Azevedo, Cromley & Seibert, 2004; Butcher, 2006; Grabowski, 2004; Wittrock

1974a, 1974b, 1974c, 1989, 1992; Mayer & Sims, 1994). In seeking to develop a flexible and

adaptive instructional resource that facilitates the development of mental models we need to

understand how learner variables interact with interactive and multimedia features of the

textbook and how these influence extraneous and germane cognitive load. Before outlining the

planned studies we present the theoretical background that guides the development of the

interactive textbook materials, and describe the learner variables known to influence the usage

and effectiveness of multimedia features.

Cognitive Load Theory. Prior to the past 20 years, researchers interpreted the magic

number seven plus or minus 2 (Miller, 1956) as a measure of how many items a participant could

hold in working memory at one time. It was not until Sweller’s (1988) subsequent research that

researchers began to understand the effect of the design of the instructional materials on mental

model (or schema) development as a result of limited working memory. For example,

memorizing the six Noble gases and their chemical symbols is a relatively easy task. Similarly,

memorizing all the metals and their symbols is a relatively easy task although it would take more

time to learn than the six Noble gases. In contrast, learning to solve a typical time-distance

problem is much more difficult. The basis for the difficulty lies in the content. When memorizing

parts (or all) of the periodic table of elements, each element is learned independently. That is,

argon is not dependent upon learning helium. The difference between these two learning tasks is

described in terms of element interactivity (Schnotz & Kurschner, 2007; Sweller & Chandler,

1994). The memorization of parts of the periodic table is a task of low element interactivity,

which is to say that learning one element is not dependent upon the other. Learning to solve a

time-distance problem is characterized as having high element interactivity. A typical time-

distance problem can have 11 more elements that must be kept in working memory

simultaneously to solve the problem. Thus, this type of task can overwhelm the learner’s

working memory capacity and prevent the development of an appropriate mental model that can

be used to solve future problems.

Cognitive load is defined as the load placed on the cognitive system by a task. This load

is defined by two dimensions. The first is the load from the task which is labeled mental load and

the second is the learner’s cognitive capacity or resources, mental effort, brought to bear on the

task (Sweller, van Merriënboer, & Paas, 1998). There are three types of cognitive load. First is

intrinsic cognitive load that is determined by the level of interactivity between elements of the

task. Second is extraneous cognitive load which is the load imposed or created by the design of

the materials. For the studies, our interests focus on the management of two causes of

extraneous load. The first is the split-attention effect that occurs in most printed materials that

incorporate pictures (Bobis, Sweller, & Cooper, 1993; Chandler & Sweller, 1991). The learner

typically first reads a description of one part of the illustration in the narrative and then must

locate that part of the illustration, maybe on another page, then reads the next phrase or

description, then switches to the illustration. Switching between the illustration and narrative

requires the learner to use working memory resources for administrative purposes of keeping

locations in the text and illustration and maintaining or rehearsing information during location

switching. The second cause of extraneous load is due to redundancy of information between the

text and illustration (Mayer & Moreno, 2002; Reder & Anderson, 1980, 1982). The learner could

achieve understanding from either source, but must now reconcile the differences between the

illustration and narration to determine if they are equivalent.

The third type of cognitive load is germane load. Germane load refers to the working

memory resources available after accounting for intrinsic and extraneous load (see Figure 1) and

is used to engage in generative processing. The three types of cognitive load are additive. The

example on the left in Figure 1 has overwhelmed working memory and the learner will not

develop an appropriate mental model due to the lack of germane cognitive load. The example on

the right of Figure 1 illustrates a reduction in extraneous cognitive load which frees working

memory resources to develop a mental model, that is, increased germane cognitive load.

Research suggests that judicious application of message design, cognitive load theory, and

multimedia learning principles can reduce extraneous cognitive load. Effective scaffolding and

support of cognitive and metacognitive processes can maximize the benefits associated with

germane cognitive load (Anglin & Morrison, 2001; Pociask & Morrison, 2008; Sweller, 1999).

Optimal and efficient learning from technologically-enhanced text results from effective

management of limited working memory resources (Sweller, van Merriënboer, & Paas, 1998).

Research suggests that judicious application of message design, cognitive load theory, and

multimedia learning principles can reduce cognitive load. Effective scaffolding and support of

cognitive and metacognitive processes can maximize the benefits associated with germane

cognitive load (Anglin & Morrison, 2001; Pociask & Morrison, 2008; Sweller, 1999). Therefore,

the first step in our project will be to re-design textbook sections in accordance with cognitive

load theory and to assess the impact of these changes on cognitive load and student performance.

In particular, we will reduce the split attention and redundancy effects.

Figure 1. Additive effect of cognitive load

Learner Variables, Generative Strategies and the Application of Multimedia Learning

Theory to i-textbook design. Over the past decade, Mayer and colleagues (e.g. Mayer et al.,

2003) have conducted a number of experiments using short multimedia presentations about

scientific and technical systems. These experiments have led to the articulation of multimedia

learning principles, including the coherence and interactivity principles. The coherence principle

clearly supports research on cognitive load. Studies have shown that individuals learn better

when extraneous details are removed from informational material (Mayer, 2001, Butcher, 2006).

The interactivity principle states that individuals learn better if they are able to proactively

respond to and exert control over both the pace and the sequence of instruction (Mayer &

Moreno, 2003). However, the results of these types of experiments present a challenge to

instructional designers because conclusions cannot be equivocally drawn. Learner variables such

as prior knowledge and self-regulated learning skills interact with features such as animation,

complexity and interactivity to predict learning outcomes. Such individual differences have not

typically been systematically addressed in studies that have been seeking to manipulate

interactivity or animations.

Because individual difference variables are influential, and because middle school

students require support and prompting to use generative learning strategies and execute self-

Extraneous

Intrinsic

Germane

regulated learning skills (Pressley, 2000), we will assess students’ prior knowledge, generative

and self-regulated learning strategies, and incorporate metacognitive support into the i-textbook

pages as multimedia features are added. In spite of a wealth of research on learning from

multimedia presentations, relatively few details have emerged about how students use generative

learning strategies during extended, interactive multimedia learning episodes. According to

Jonassen (1988), generative learning strategies “are those that require learners consciously and

deliberately to relate new information to existing knowledge” (p.114). Cognitive learning

strategies – techniques that support effective encoding – are included under this general heading

and for our purposes include elaborative rehearsal and organizational strategies (Weinstein &

Mayer, 1986). Successful strategies in multimedia learning environments and in traditional

textbooks have involved prompting to use strategies such as elaborative interrogation (asking

“why” questions; Smith, Holliday & Austin, 2010); the use of self-explanations during a multi-

step or multi-component sequence (Chi, et al., 1989; Renckl, 2002); the use of worked examples

with gradual fading of the completeness of the steps provided (Renckl, Atkinson & Groβe,

2000); directed re-reading of portions of text; help tools to reveal information (Renckl, 2002);

summary writing; paraphrasing; and feedback regarding the degree to which a strategy was used

successfully (Azevedo, 1995). When strategy prompts are integrated into the i-textbook platform

and behaviors are recorded, we have the potential for students to demonstrate not only superior

learning outcomes at posttest but also to demonstrate the processes that lead to successively more

sophisticated levels of understanding and learning through the quality of their responses to and

the need for activating these prompts.

Our approach will be to build in prompts that will facilitate strategy use. Of particular

interest is the identification of when and how often these prompts should be incorporated. Some

insights may be gained from Azevedo, Greene and colleagues, (Azevedo, Moos, Greene,

Winters, & Cromley, 2008; Greene et al., 2010), who documented differences in the strategic

processes used by students with high or low domain knowledge. Specifically, even for students

with effective self-regulated learning skills, low prior knowledge hampered learning outcomes

(Greene et al., 2010). Students with both low prior knowledge and poor self-regulated learning

skills may require a great deal of support. Furthermore, this support may need to be fixed rather

than under the learner’s control, especially for less skilled learners.

Self-regulated learning (SRL) involves the metacognitive, motivational and behavioral

control of learning (Zimmerman, 2000) and consists of three phases that are cyclical in nature.

Here, we are most concerned with the metacognitive guidance of learning, which entails

students’ ability to monitor and reflect upon their knowledge and the quality of their learning

processes. In the Forethought stage of SRL learning goals are set, plans are formulated, and

strategies are selected. In the Performance or Volitional Control stages, the learning task is

embarked on through self-instruction, and comprehension and progress are monitored

throughout. In the Self-Reflection performance is judged and feedback is reviewed in order to

improve future learning episodes. Thus, prompting and scaffolding that is integral to the learning

materials could conceivably support both the Forethought and Performance stages of SRL. It

would benefit the learner who both lacks the metacognitive self-regulation to begin the SRL

process, as well as supporting the learner who was aware of the need to be strategic but who

lacked the specific knowledge to execute a particular strategy at an optimal point in time.

Contemporary research findings have highlighted the role played by SRL skills in accounting for

variance in learning outcomes for multimedia environments (Azevedo, Guthrie & Seibert, 2004).

Specifically, learning in multimedia and hypermedia settings allows the learner such great

control over navigation that metacognitive skills must be used very effectively. These findings

will provide a foundation for designing a new type of interaction in the interactive textbook. We

will focus our prompts in the interactive textbook during the Performance phase, in order to

maximize the opportunity for students to engage in generative processing. In addition to

assessing self-regulated learning skill as an independent variable, we will examine the possibility

of interaction effects for different levels of SRL skill when learning from interactive i-textbook

pages. We will also examine whether the expert-reversal effect can be applied from the

perspective of students with high versus low SRL skills (Young, 1996). The learning outcomes

of students with very good SRL skills may actually be hampered by a high degree of rigid

metacognitive support. Likewise, students with very poor SRL skills may be hindered by the

condition in which they have full control over their navigation and processing of the material

(McManus, 2000), or they may demonstrate the ability to select the best strategy for their skill

level (Ross, Morrison, & O’Dell, 1988).

Materials and measures

In addition to the rubric used to assess the textbook and i-textbook pages, individual

measures of cognitive load, cognitive abilities, self-regulated learning, domain knowledge,

domain attitudes, time on task, and process will be collected.

Cognitive Load. The most common measure of cognitive load is a subjective rating scale.

Prior research has found that participants are capable of providing a numeric indicator of how

they perceive their mental burden (Gopher & Braune, 1984; Paas, 1992). Moreover, deLeeuw &

Mayer, (2008) presented evidence that questions about task difficulty versus effort could

distinguish between extrinsic and intrinsic cognitive load. Although we will not only use self-

report data, we will seek to establish whether or not it is possible to assess cognitive load in

middle and high school students using questions that parallel those asked by deLeeuw and

Mayer.

The most commonly used approach to measuring cognitive load is subjective measures

(Paas, Tuovinen, Tabbers, & Van Gerven, 2003; van Gog, Kester, Nievelstein, Giesbers, & Paas,

2009) consisting of multiple times or a single item. In contrast are physiological measures that

can offer a continuous assessment of cognition and cognitive load. However, the use of

physiological measures to assess cognitive load and cognition has the potential to a) verify

subjective measures of cognitive load and b) provide important continuous indicators of

cognition that can be recorded and eventually adapted as the learner interacts with the material

(van Gog, Pass, & Sweller, 2010). Some measures such as heart rate and hormone levels do not

appear to be as sensitive as other measures to changes in mental effort (Atonenko, Paas, Grabner,

& van Gog, 2010). Other physiological measures such as eye tracking and electroencephalogram

(EEG) provide continuous measures of cognition and cognitive load that are more sensitive to

changes. For example, May, Kennedy, Williams, Dunlap, and Brannan (1990) reported four

studies where eye tracking was successfully used to measure cognitive load. Similarly, Paas

(2003) and van Gog et al. (2009) summarize additional studies that have successfully used eye

tracking to measure cognitive load. The laboratory studies will use a heads-free eye tracking

system and a 24-channel wireless EEG acquisition system to continuously track a learner’s

visual attention while interacting with the instructional material. The eye tracker will be used to

measure number and duration of fixations, along with visual search and navigation behaviors.

Similar eye tracking measures have been used successfully to track cognitive processes and

processing demand (Amadieu, van Gog, Paas, Ticot, & Marine, 2009; Rayner, 1998), cueing and

complexity of animations (Boucheix and Lowe, 2010; de Koning, Tabbers, Rikers, & Paas,

2010), and multimedia learning (Jamet, Gavota, & Quaireau, 2008; Hyona, 2010) in prior

studies.

More recently, EEG has been used to measure of cognitive load (Atonenko, et al. 2010).

The use of EEGs provides researchers with a continuous rating of cognitive load that can provide

a more accurate assessment of instructional interventions than a single subjective measure

(Atonenko, et al. 2010). Of particular interest are the P3 measures of attention (Debener, Makeig,

Delorme, & Engle, 2005). For example, various P3 measures from an EEG can provide insights

into involuntary, stimulus directed attention mechanisms as well as events that cause updating of

stimuli in working memory. Eye tracking will be correlated with simultaneous tracking of

electrical activity in the brain using an EEG machine to measure cognition. Specifically,

predominant signal peaks called event-related potential (ERP) in the P300 region will be linked

to measure the presence, absence, and amount of cognition associated with each of the areas in

the learning environments (Hillyard, 2009; Makeig, 2009; van Gog, Pass, & Sweller, 2010). By

adding EEG and eye-tracking measures with the more traditional self-reports of cognitive load,

we believe that we can obtain reliable measures of changes in mental effort. Using these data, we

can then determine when we need to include prompts for either generative strategies or self-

regulated learning prompts to help the learner develop a relevant mental model before the

learner’s working memory is overwhelmed. For classroom-based studies, physiological measures

are not feasible as they are designed for collection in a controlled lab environment. However, the

use of physiological measures to assess cognitive load and cognition has the potential to a) verify

subjective measures of cognitive load and b) provide important indicators of cognition. Eye

tracking and EEG data, for example might reveal areas of the page that students reach before

returning to a previous section, or otherwise dwell upon. These data could be triangulated with

think aloud data and may reveal an area of the text that is difficult. Subsequent iterations of the i-

textbook could target this section for prompting and strategy use, such as re-reading,

paraphrasing, or elaborative interrogation.

Cognitive abilities. Participants will be individually assessed on their verbal and

nonverbal working memory, to provide an estimate of their working memory capacity. A

modified version of Daneman and Carpenter’s listening span task (Daneman and Carpenter,

1980) will be used for verbal working memory. A modified version of the Wide Range

Assessment of Memory and Learning subtest Finger Windows (Sheslow & Adams, 2004) will be

used to assess nonverbal working memory. Students will be assessed on their spatial ability using

a composite version of Shepard and Metzler’s mental rotation task (Shepard & Metzler, 1971).

Self-regulated learning. Self-regulated learning and generative strategy use will be

measured using self-report, think aloud, and behavioral trace data. Students will take the

Motivated Strategies for Learning Questionnaire (MSLQ; Pintrich, Smith, Garcia & McKeachie,

1991), which is a self-report measure of self-regulated learning strategies that has been

effectively modified for use in junior high and middle school as well as college level populations

(Liu, 2003; Wolters, 2004). It is domain specific in the sense that individuals report on their use

of generative strategies and motivational preferences for a particular course. In selected studies,

students will also complete a think-aloud protocol in which they will be prompted to talk out

loud about what they are doing as they are doing it. Think aloud protocols have been

successfully used to capture self-regulated learning data in middle school and high school

students (Azevedo et al., 2008; McManus, 2000). Finally, trace data will be gathered from the

iPad1 software. The usage of prompts, highlighting and note-taking, and navigational behaviors

of students will be recorded. This will provide valuable feedback regarding actual usage of the

interactive features of the App, and will enable triangulation with learning outcomes.

Domain knowledge and attitudes. A pre-test will be given to students ahead of each

experimental study. The pre-test will consist of a prompt to write a short essay, followed by

multiple choice questions. Posttests will include comprehension questions that draw from

different levels of Bloom’s taxonomy, and will include text-explicit, text-implicit, and inference-

based questions (Morrison, Ross, Gopalakrishnan, & Casey, 1995). Questions requiring

simulation of a process will also be included in order to reveal the sophistication of students’

mental models for the relationships between components of specific geologic systems and

processes. A short survey will also be administered to students to assess their attitudes towards

learning Earth science content.

Time on task and process measures. Total time on task will be recorded from each iPad.

This is for formative evaluation reasons as well as pedagogical considerations. In addition, this

measure is of interest because research has mixed results regarding the effect of adding

interactivity (Tabbers & deKoeijer, 2010) and prompting (Atkinson & Renkl, 2007) on time on

task. For the paper version of the textbook pages, classroom behavioral observations will be

conducted to assess time on task for a random subsample of students in each class. Process

measures of usage (e.g. page turning, highlighting, clicking on graphics, and note-taking.) will be

recorded from the iPad.

Data Analysis. Qualitative and quantitative data analytic techniques will be used as

appropriate for each study. For the process data studies, think aloud data will be coded according

1 The term iPad is used as a generic term for any current or future pad type device that can be used in this study.

to a rubric that will establish processes and sub-processes in each phase of the self-regulated

learning cycle. Students’ pre-test essays will also be coded according to a rubric to establish the

sophistication of their mental model and comprehension of a particular phenomenon. Self-report

survey data will be processed in order to generate total scores for particular aspects of self-

regulated learning (e.g. cognitive versus metacognitive strategy use). Behavioral traces will be

coded according to strategy use. Behavioral measures taken during the learning task will be

recorded and in some cases (e.g. time on task) may be used as a dependent variable. For

experimental studies, multivariate analysis of variance and covariance will be used (MANOVA,

MANCOVA) to determine the effects of the independent variables on outcome variables such as

comprehension. Additional measures and assessments will be incorporated as part of the ongoing

formative and summative evaluation (see Evaluation section).

Procedure: Sequence of studies (Figure 2).

Year 1. The outcome of Year 1 will be the re-design of sections of middle school Earth

science textbook chapters. Potential topics might include earthquakes and volcanoes, erosion and

deposition, solar system, ocean water and ocean life, running water and ground water, and

moisture, clouds and precipitation. Specific topics will be determined in consultation with

teachers and partners to identify units with appropriate difficulty and that fit the classrooms’

schedules, that is, selecting topics that have not been addressed in earlier assignments. In order to

achieve this goal, we will undertake three studies. In study 1a we will gather samples of 5-7

existing Earth science textbooks, randomly select pages and illustrations using a protocol and,

using a rubric code features which increase extraneous cognitive load and decrease

comprehension because of split-attention and redundancy effects. Graphics will be classified

according to purpose and function (Carney, & Levin, 2002; Levin, Anglin, & Carney, 1987). The

primary hypothesis in study 1a is that existing middle school science textbooks contain features

which contribute to split-attention and redundancy effects. A second hypothesis is that a high

proportion of the images and graphics used in these textbooks are decorative, rather than

representational, organizational, interpretational, or transformational in function. Levin, Anglin,

& Carney (1987) reported effect sizes of .50 and higher for four of the five picture categories all

categories but decorative pictures. Thus, a secondary question is whether textbooks are using

pictures in a manner that supports learning (Carney & Levin, 2002).

In study 1b, N=120 middle school students in an Earth science course will review the

redesigned pages in a pilot test of the revised version. This version will incorporate Integrated

Graphics and no redundancy. Independent variables will include prior knowledge and self-

reported self-regulated learning. Dependent variables will include self-assessments of different

types of cognitive load (DeLeeuw & Mayer, 2008), time on task (Paas, Tuovinen, Tabbers, &

Van Gerven, 2003), and students’ understanding of key principles via a posttest of

comprehension and problem solving. Because this study is a pilot study, data will be collected

with a view to making further revisions of the materials.

Study 1c will be a 2x2 factorial experiment for study 1c. With 120 middle school

students, we will compare the effect of the original format textbook pages versus the Integrated

Graphic textbook pages, and the effect of reading these pages on paper versus reading them

electronically via an iPad2. Students in the electronic reading conditions will be provided with

the iPad ahead of time in order to reduce novelty effects. Students will engage in the reading task

during a regular class period, complete a short distractor task, and then a posttest. Independent

2 We will use the term iPad to refer to both the iPad and other similar pad type devices

measures will mirror those in study 1b. Additional dependent variables will include time on task

and an assessment of cognitive load. Hypotheses are that students in both paper and electronic

Integrated Graphics conditions will experience lower assessments of cognitive load, and superior

performance on the comprehension posttest. We do not expect to observe performance

differences for the manipulation of electronic versus paper medium. One research question of

interest is if there are any differences in time-on-task with the two different media (Shepperd,

Grace, & Koch, 2008). With a 2 X 2 balanced factorial design and 30 students per cell, a power

of 0.82 is obtained for detecting a moderately large effect size.

Year 2. Year 2 begins with a study in which process data are gathered. It continues with

studies which examine the effects of bringing interactive and adaptive features to the electronic

learning materials, as well as the effect of including prompts for generative strategy use, and the

effect of providing feedback following strategy use. Four smaller subsamples of students will

participate in Study 2a, the goal of which is to gather data describing the processes used as

students read the text. Eye tracking and EEG data will be collected for two groups of students as

they read either the electronic original format text (n=20) or the electronic Integrated Graphics

text (n=20). Think aloud data will be collected for two more groups of students as they read

either the original format (n=20) or Integrated Graphics text (n=20). We hypothesize that eye

tracking and EEG data will reveal evidence of lower cognitive load in the Integrated Graphics

text format conditions. We hypothesize that analyses of think aloud data will reveal fewer

instances of comprehension breakdown in the Integrated Graphics text condition. The data will

also be analyzed to determine specific points in the text that are difficult to understand. These

results will be used to identify when to add supports (prompts) in later studies.

After closely reviewing the results of study 1c and 2a, in Study 2b we will pilot test

refined versions of the electronic pages. Student and teacher surveys and focus groups with 10-

15 science teachers will also be conducted. The objective is to increase the richness of

information about strengths and weaknesses in students’ generative learning strategies and

science reasoning/critical thinking skills. We will use the resulting information to select a small

number of critical generative learning strategies that will be incorporated as prompts in the Year

2 version of the materials. Prior research suggests that both pictures (Anglin, Vaez, &

Cunningham, 2004; Levie & Lentz, 1982) and generative strategies (Johnsey, Morrison, & Ross,

1992; Rigney, 1978) need appropriate prompts and training in use for effectiveness. In addition,

consultation with science teachers and science experts will facilitate the design of interactive

elements into the textbook graphics. Interactive features will include learner-controlled labeling

within the graphic.

Data from Study 2a and 2b will be used to inform the nature and position of the prompts

in the final version of the learning materials for Study 2c. Study 2c will be a 2x2x2 factorial

design involving (n=120) students in classroom settings. Independent variables will mirror study

1b with regard to learner differences. For the i-textbook pages, independent variables will

include static versus interactive graphics, prompted strategy use versus no prompted strategy use,

and feedback versus no feedback following strategy use. Feedback is expected to be beneficial.

For example, Azevedo (1995) found that immediate feedback raised performance by an average

effect size of .80. Static graphics will be of the Integrated Graphic type, with labels included as

part of the graphic. Interactive graphics will have labels which can be viewed by touching a hot

area within the image. In the prompted strategy use conditions, students will be prompted by the

application to engage in a strategy prior to moving to the next section of the text. In the feedback

conditions, students will be prompted to engage in the strategy, review the product of an ‘expert’

using the strategy, and will then be prompted to compare their own work to that of the expert.

Students will complete the textbook reading task during class time, will be provided with a short

distractor task, and then will take a posttest. The posttest will parallel those in the previous

studies but students’ notes and other physical traces of strategy use will be collected and

examined. Additional measures of time on task and cognitive load will be gathered. Our

hypotheses are that there will be main effects for graphic type, prompting and feedback.

Interaction effects may be present, such that students who experience the interactive graphics

with prompted support for strategy use and feedback, outperform students in other conditions.

One research question of interest in this study is whether there is evidence of cutting and pasting

for the generative strategy (Shepperd et al., 2008).

Year 3. Study 3a will explore the impact of incorporating animations and simulations

into the textbook pages (Mayer, 2001). Animations can help the learner develop conceptual

models of how the phenomena interacts (Mayer & Moreno, 2002). Similarly, simulations allow

the learner to explore complex and dynamic relationships and to formulate and test hypotheses

about the outcomes if parameters of a simulation were to be changed (Lee, Plass & Homer,

2006). Recent research, however, by Mayer, Hegarty, Mayer, and Campbell (2005) suggests that

static pictures are more effective than animations. Animations may be more effective for learners

with limited spatial ability (Bartholome & Bromme, 2009; Huk, 2006) or to animate processes

that cannot be seen such as the movement of air or evaporation of water. This strategy can be

seen as an extension of the congruence principle (Tversky, Morrison & Betrancourt, 2002),

which states that effective animations and graphics must parallel the structural elements and

causal behavior of the individual’s mental model during simulations. Thus, if the quality of the

learner’s mental model will be enhanced by the animation or simulation then they should be

included (Reed, 2010). However, if the learner’s lack of prior knowledge is such that they would

attend to irrelevant information then animations may not be the most useful multimedia strategy

(Lowe, 1999). A significant advantage of the static visuals over animations is the lower cost for

development. In this experiment, graphic type (interactive graphics vs animated graphics vs

animation with simulation) will be crossed with explanation modality (text vs narration),

resulting in a 3 x 2 factorial design. As before, we will assess the quality of prior knowledge

using a single essay prompt given prior to the experimental session. These data will allow

analyses to consider low and high domain knowledge as an independent variable. On the

condition that prompting and feedback are beneficial to learning, all conditions will include

prompting and feedback. Dependent variables will parallel those in study 2c. The hypotheses

concern the improvement in learning outcome for animation versus static graphics, and the

further improvement for animation plus simulation. We expect that this effect may depend on the

nature of the posttest, and that students in the simulation condition may outperform other

students on measures of problem solving and transfer only (Lee, Plass & Homer, 2006). We also

suspect that learner variables such as prior knowledge may interact with manipulated i-textbook

features.

In study 3b we pilot test materials for study 4a. A single class of students (n=25) will use

the interactive, adaptive materials in order to ensure that they are sufficiently easy to use, and to

inform the need for any additional prompts. A panel of teachers will provide feedback during the

development of the redesigned materials.

Year 4. In study 4a we manipulate the variables of learner control and adaptive support,

for three levels of self-regulated learning skill. Using self-report data, students will be classified

as having high, medium or low self-regulated learning skills. As before, prior knowledge will be

assessed via a pre-test essay prompt, and will be used to determine if low and high prior

knowledge interacts with the other independent variables. Self-regulated learning skill will be

crossed with one of three types of text. The first will provide strategy-related prompts and

feedback at the end of each section. The second will provide prompting and feedback should the

student’s performance on end-of-section comprehension questions fall below a criterion level.

The third type gives the learner control over whether or not to activate the prompting and

feedback prior to the beginning of each section. Graphic features of the textbook will match

those found to be most effective in study 3a and 3b. Dependent variables will mirror those in

study 2c. We hypothesize that significant interactions will occur. Learners with low SRL skills

will perform the most poorly in the learner control condition and will exhibit the best

performance in the continuous support condition. Learners with high SRL skills may actually

perform better in the learner control condition than in the continuous support condition. This

finding has been documented as a parallel expert reversal effect with self-regulated learning as

the independent variable (Kalyuga, Chandler, & Sweller, 2003; Young, 1996). Dependent

variables will parallel those in studies 1c, 2b and 2c. The results of this experiment will be

compared to research investigating the ability of students to control the pace and sequence of

their learning (the interactivity principle), research on intelligent tutoring systems, and research

on the impact of high and low domain knowledge and self-regulated learning skills on navigation

through and learning from technology-enhanced text.

By the fourth year we will have gathered empirical data that will allow us to draw

conclusions about the types of electronic textbook pages which best promote learning of Earth

science content and the role of learner variables in predicting learning outcomes. We will

conduct a final round of focus groups and classroom observations to gather information about

how i-textbook chapters could best be used in the course of a regular instructional period. We

will assess students’ attitudes towards the final versions of the materials. Our final materials will

include redesigned and empirically tested i-textbook chapter materials suitable for up to one

marking period of an Earth Science course. During Year 4 we will produce the guidelines for

publishers and authors. We will continue to disseminate the results of the project. Importantly,

we will develop and give webinars to convey information about i-textbook redesign, the project’s

findings, and information for teachers regarding how the materials can be used in the classroom.

We will also begin to develop teacher professional development seminars and supporting

materials in order to facilitate teachers’ use of the i-textbooks.

Figure 2. Timeline for the research

Conclusion

Textbooks in the United States have generally been “designed to sell,” not designed to

support student learning. Even existing texts are rife with features that increase extraneous

cognitive load, and make access to higher level content difficult for all students, but particularly

those with special needs. Currently, e-texts do not capitalize on the computing capacity behind

them to integrate well-established principles of learning theory, or to flexibly adapt to learner

needs, whether the learner has executive function of self-regulation difficulties, or is an “expert.”

Through this planned research and development effort, we hope to create the next generation of

interactive expository materials that will increase access to advanced science content for diverse

student populations, and provide a framework for large-scale production of adaptive materials

for major publishers of educational materials.

References

Advisory Committee on Student Turn the page: Making college textbooks more affordable. Washington, DC: Advisory Committee on Student Financial Assistance. Retrieved December 14, 2010 from http://www2.ed.gov/about/bdscomm/list/acsfa/turnthepage.pdf

Amadieu, F., van Gog, T., Paas, F., Tricot, A., & Marine, C. (2009). Effects of prior knowledge and concept-map structure on disorientation, cognitive load, and learning. Learning and Instruction, 19, 376-386.

Anglin, G. J., Vaez, H., & Cunningham, K. L. (2004). Visual representations and learning: The role of static and animated graphics. In D. H. Jonassen (Ed.), Handbook of research for educational communications and technology (2 ed., pp. 865-916): Mahwah, NJ, Lawrence Erlbaum Associates, Publishers.

Anglin, G. J., & Morrison, G. R. (2001). Cognitive load theory: implications for instructional design research and practice. Paper presented at the Annual meeting of the Association of Educational Communications and Technology, Atlanta, GA.

Atkinson, R.K., & Renkl, A. (2007).Interactive example-based learning environments: Using interactive elements to encourage effective processing of worked examples. Educational Psychology Review 19, 375-386.

Atonenko, P., Paas, F., Grabner, R., & Van Gog, T. (2010). Using electroencephalography to measure cognitive load. Educational Psychological Review, 22(425-438). doi: 10.1007/s10648-010-9130-y

Azevedo, R., Cromley, J., & Seibert, D. & Tron, M. (2003). The role of co-regulated learning during students’ understanding of complex systems with hypermedia. Paper presented at the Annual Meeting of the American Educational Research Association, Chicago, IL.

Azevedo, R., Guthrie, J. & Seibert, D. (2004). The role of self-regulated learning in fostering students’ conceptual understanding of complex systems with hypermedia. Journal of Educational Computing Research 30, 87-111.

Azevedo, R. & Hadwin, A. (2005). Scaffolding self-regulated learning and metacognition: Implications for the design of computer-based scaffolds. Instructional Science 33, 367-379.

Azevedo, R., Moos, D.C., Greene, J.A., Winters, F.I., & Cromley, J.G. (2008). Why is externally-facilitated regulated learning more effective than self-regulated learning with hypermedia? Educational Technology Research and Development 56, 45-72.

Bartholomé, T., & Bromme, R. (2009). Coherence formation when learning from text and pictures: What kind of support for whom? Journal of Educational Psychology 101, 282-293.

Best, R.M., Rowe, M., Ozuru, Y., & McNamara, D. S. (2005). Deep-level comprehension of science texts: The role of the reader and the text. Topics in Language Disorders, 25(1), 65.

Bobis, J., Sweller, J., & Cooper, M. (1993). Cognitive load effects in a primary-school geometry task. Learning & Instruction, 3(1), 1-21.

Boucheix, J.M. & Lowe, R.K. An eye tracking comparison of external pointing cues and internal continuous cues in learning with complex animations. Learning and Instruction, 20, 123-135

Boyd, F., & Thompson, M. (2008). Multimodality and literacy learning: Using multiple texts to enhance content-area reading. In K. A. Hinchman & H. Sheridan-Thomas (Eds), Best practices in adolescent literacy instruction (pp. 151-163). NY: Guilford.

Brown, D.F., & Canniff, M. (2007). Designing curricular experiences that promote young adolescents’ cognitive growth. Middle School Journal 39, 16-23.

Butcher, K.R. (2006). Learning from text with diagrams: Promoting mental model development and inference generation. Journal of Educational Psychology 98, 182-197.

Carney, R. N., & Levin, J. R. (2002). Pictorial illustrations still improve students' learning from text. Educational Psychology Review, 14(1), 5-26.

Chandler, P., & Sweller, J. (1991). Cognitive load theory and the format of instruction. Cognition & Instruction, 8(4), 293-332.

Chandler, P., & Sweller, J. (1992). The split-attention effect as a factor in the design of instruction. British Journal of Educational Psychology 62, 233–246.

Chi, M.T.H., Bassok, M., Lewis, M.W., Reimann, P., & Glaser, R. (1989). Self-explanations: how students study and use examples in learning to solve problems. Cognitive Science 15, 145-182.

Clay, M. M. (1991). Becoming literate: The construction of inner control. Portsmouth, N.H: Heinemann. Cobb, P., Confrey, J., deSessa, A., Lehrer, R., & Schauble, L. (2003). Design experiments in educational research. Educational Researcher, 32, 9 -13. Daneman, M. & Carpenter, P. (1980). Individual differences in working memory and

reading. Journal of Verbal Learning and Verbal Behavior 19, 450-466. Debener, S., Makeig, S., Delorme, A., & Engle, A. K. (2005). What is novel in the novelty

oddball paradigm? Functional significance of the novelty P3 event-related potential as revealed by independent component analysis. Cognitive Brain Research, 22, 309-321. doi: 10.1016/j.cogbrainres.2004.09.006

De Koning, B. B., Tabbers, H.K., Rikers, R. M.J. P., & Paas, F. (2010). Attention guidance in learning from a complex animation: Seeing is understanding? Learning and Instruction, 20, 111-122

DeLeeuw, K.E., & 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 100, 223-234.

Deouell, L.Y. & Knight, R.T. (XXXX). Executive function and higher-order cognition: EEG studies. In L.R. Squire (Ed.), Encyclopedia of Neuroscience (pp. 105-109). Amsterdam: Elsevier

Fountas, I. C. & Pinnell, G.S. (2005). Leveled books, K–8: Matching texts to readers for effective teaching. Portsmouth, NH: Heinemann. Grabowski, B. J. (2004). Generative learning contributions to the design of instruction and

learning. In D. J. Jonassen (Ed.), Handbook of educational communications and technology (2nd ed., pp. 719-743). Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers.

Greca, I. (2000). Mental models, conceptual models, and modelling. International Journal of Science Education 22, 1-11.

Greene, J.A., Costa, L-J., Robertson, J., Pan, Y. & Deekens, (2010). Exploring relations among college students’ prior knowledge, implicit theories of intelligence, and self-regulated learning in a hypermedia environment. Computers and Education 55, 1027-1043.

Hall, T., Strangman, N., & Meyer, A. (2003). Differentiated instruction and implications for UDL implementation. Wakefield, MA: National Center on Accessing the General Curriculum. Retrieved from http://aim.cast.org/learn/historyarchive/backgroundpapers/differentiated.

Hillyard, S.A. Event-related potentials (ERPs) and cognitive processing. . In L.R. Squire (Ed.), Encyclopedia of Neuroscience (pp. 13-18). Amsterdam: Elsevier Hoffler, T.N., & Lentner, D. (2007). Instructional animation versus static pictures: A meta-analysis. Learning and Instruction 17, 722-738.

Hyona, J. (2010). The use of eye movements in the study of multimedia learning. Learning and Instruction, 20, 172-176.

Jamet, E., Gavota, M., & Quaireau, C. (3008). Attention guiding in multimedia learning. Learning and Instruction, 18, 135-145.

Johnsey, A., Morrison, G. R., & Ross, S. M. (1992). Promoting generative learning in computer-based instrution through the use of elaboration strategies in training. Contemporary Educational Psychology, 17, 125-135.

Jonassen, D.H. (1988). Instructional Designs for Microcomputer Courseware. Heinich, R. M. (1984). The proper study of instructional technology. Educational

Communications and Technology Journal, 32, 67-87. Huk, T. (2006). Who benefits from learning with 3D models? The case of spatial ability.

Journal of Computer Assisted Learning 22, 392–404 Kalyuga, S., Chandler, P., & Sweller, J. (2003). Expertise reversal effect. Educational

Psychologist, 38(1), 23-31. Lee, J., & Park, O. (2008). Adaptive instructional systems. In J. M. Spector, M. D. Merrill, J.

v. Merriënboer & M. P. Driscoll (Eds.), Handbook of research on educational communications and technology (3rd ed., pp. 470-484). New York: Taylor Francis.

Lee, H., Plass, J. & Homer, B. (2006). Optimizing cognitive load for learning from computer-based science simulations. Journal of Educational Psychology 98, 902-913.

Levie, W. H., & Lentz, R. (1982). Effects of text illustrations: A review of research. Educational Communications and Technology Journal, 30(4), 195-232.

Levin, J. R., Anglin, G. J., & Carney, R. N. (1987). On empirically validating functions of pictures in prose. In D. M. Willows & H. A. Houghton (Eds.), The Psychology of illustration, Volume 1: Basic research (pp. 51–80). New York: Springer-Verlag.

Lowe, R. (1999). Extracting information from an animation during complex visual processing. European Journal of the Psychology of Education 14, 225-244.

Makeig, S. (2009). Electrophysiology: EEG and ERP analysis. In L.R. Squire (Ed.), Encyclopedia of Neuroscience (pp. 879-882). Amsterdam: Elsevier

May, J. G., Kennedy, R. S., Williams, M. C., Dunlap, W. P., & Brannan, J. R. (1990). Eye movement indices of mental load. Acta Psychologica, 75, 75-89.

Mayer, R.E. (2001). Multimedia Learning. New York: Cambridge University Press

Mayer, R. E. (2008). Applying the Science of Learning: Evidence-Based Principles for the Design of Multimedia Instruction. American Psychologist, 63(8), 760-769.

Mayer, R. E., & Chandler, P. (2001). When Learning Is Just a Click Away: Does Simple User Interaction Foster Deeper Understanding of Multimedia Messages? Journal of Educational Psychology, 93(2), 390-397.

Mayer, R.E., Dow, G.T., & Mayer, S. (2003). Multimedia learning in an interactive self-explaining environment: What works in the design of agent-based microworlds? Journal of Educational Psychology 95, 806-813.

Mayer, R. E., Hegarty, M., Mayer, S., & Campbell, J. (2005). When static media promote active learning: Annotated illustrations versus narrated animations in multimedia instruction. Journal of Experimental Psychology, 11(4), 256-265. doi: 10.1037/1076-898X.11.4.256

Mayer, R.E., & Moreno, R. (2002). Aids to computer-based multimedia learning. Learning and Instruction 12, 107-119.

Mayer, R.E., & Moreno, R. (2003). Nine ways to reduce cognitive load in multimedia learning. Educational Psychologist 38, 43-52.

Mayer, R.E. & Sims, V.K. (1994). For whom is a picture worth a thousand words? Extensions of a dual-coding theory of multimedia learning. Journal of Educational Psychology 86, 389-401.

Miller, G. A. (1956). The Magical Number Seven, Plus or Minus two: Some limits on our capacity for processing information. Psychological Review, 63, 81-97.

McManus, T. (2000). Individualizing instruction in a web-based hypermedia learning environment. Nonlinearity, advance organizers, and self-regulated learners. Journal of Interactive Learning Research 11, 219-251.

Moje, E. & Speyer, J. (2008). The reality of challenge texts in high school science and social studies: How teachers can mediate comprehension. In K. A. Hinchman & H. Sheridan-

Thomas, (Eds), Best practices in adolescent literacy instruction (pp151-163). (pp.185-211). NY: Guilford.

Morrison, G. R., Ross, S. M., Gopalakrishnan, M., & Casey, J. (1995). The effects of feedback and incentives on achievement in computer-based instruction. Contemporary Educational Psychology 20, 32-50.

Paas, F., Tuovinen, J.E., Tabbers, H.K., & Van Gerven, P.W.M. (2003). Cognitive load measurement as a means to advance cognitive load theory. Educational Psychologist 38, 63-71.

Park, O., & Lee, J. (1996). Adaptive instructional systems. In D. Jonassen (Ed.), Handbook of research on educational communications and technology (2nd ed., pp. 651-684). Maway, NJ: Lawrence Erlbaum Associates, Publishers.

Pintrich, P.R., Smith, D.A.F., Garcia, T. & McKeachie, W.J. (1991). A manual for the use of the Motivated Strategies for Learning Questionnaire. (Tech. Rep. No. 91-B-004). Ann Arbor: University of Michigan.

Pociask, F. D., & Morrison, G. R. (2008). Controlling split attention and redundancy in physical therapy instruction. Educational Technology Research and Development, 64(4), 379-399.

Pressley, M. (2000). What should comprehension instruction be the instruction of? In M. L. Kamil, P.B. Mosenthal, P.D. Pearson, & R. Barr (Eds.), Handbook of Reading Research,(Vol. 3, pp.545-561). Mahwah: NJ: Lawrence Erlbaum.

Rayner, K. (1998). Eye movements in reading and information processing: 20 years of research. Psychological Bulletin, 124, 372-422

Reed, S.K. (2010). Cognitive architectures for multimedia learning. Educational Psychologist 41, 87-98.

Reder, L., & Anderson, J. R. (1980). A comparison of texts and their summaries: Memorial consequences. Journal of verbal learning and verbal behavior, 19, 121-134.

Reder, L., & Anderson, J. R. (1982). Effects of spacing an embellishment on memory for main points of a text. Memory & Cognition, 10, 97-102. Renkl, A., Atkinson, R.K., & Groβe, C.S. (2004). How fading worked solution steps works – a cognitive load perspective. Instructional Science 32, 59-82.

Renckl, A. (2002). Learning from worked-out examples: Instructional explanations supplement self-explanations. Learning and Instruction 12, 149-176. Renkl, A., Atkinson, R.K., & Große, C.S. (2004). How fading worked solution steps works - a cognitive load perspective. Instructional Science 32, 59-62.

Rigney, J. (1978). Learning strategies: A theoretical perspective, pp. 164-205. In H. F. O’Neil (ed.), Learning Strategies. New York: Academic Press.

Ross, S. M., Morrison, G. R., & O'Dell, J. (1988). Obtaining more out of less text in CBI: Effects of varied text density levels as a function of learner characteristics and control strategy. Educational Communications and Technology Journal, 36, 131-142.

Shepard, R. & Metzler, J. (1971). Mental rotation of three dimensional objects. Science 171, 701-703.

Shepperd, J. A., Grace, J. L., & Koch, E. J. (2008). Evaluating the eectronic textbook: Is it time to dispense with paper text? Teaching of Psychology, 35, 2-5.

Sheslow, D., & Adams, W. (2004). The Wide Range Assesment of Memory and Learning. Dallas, TX; Pro-Ed, Inc.

Sweller, J. (1988). Cognitive load during problem solving: Effects on learning. Cognitive Science, 12(2), 257-285.

Sweller, J. (1999). Instructional design in technical areas. Australian Education Review, No. 43 (pp. 168). Australia; Victoria.

Sweller, J. & Chandler, P. (1994). Why some material is difficult to learn. Cognition and Instruction 12, 185-223.

Sweller, J., van Merriënboer, J. J. G., & Pass, F. (1998). Cognitive architecture and instructional design. Educational Psychology Review, 10, 251-296.

Schnotz, W., & Kurschner, C. (2007). A reconsideration of cognitive load theory. Educational Psychology Review 19, 469-508.

Tarmizi, R. A., & Sweller, J. (1988). Guidance during mathematical problem solving. Journal of Educational Psychology, 80(4), 424-436.

Tabbers, H.K., & deKoeijer, B. (2010). Learner control in animated multimedia instructions. Instructional Science: An International Journal of the Learning Sciences 38, 441-453. Tran, M. (2009). Arnold Schwarzenegger to scrap school textbooks in favour of ebooks:

Governor of California seeks to cut budget deficit by replacing ‘outdated’ textbooks with electronic reading devices. Retrieved December 14, 2010 from http://www.guardian.co.uk/world/2009/jun/09/arnold-schwarzenegger-school-textbooks-ebooks

Tversky, B., Bauer-Morrison, J., & Betrancourt, M. (2002). Animation: Can it facilitate? International Journal of Human-Computer Studies 57, 247-262.

van Gog, T., Kester, L., Nievelstein, F., Giesbers, B., & Paas, F. (2009). Uncovering

cognitive processes: Different techniques that can contribute of cognitive load research and instruction. Computers in Human Behavior, 25, 325-331. doi: 10.1016/j.chb.2008.12.021

van Gog, T., Paas, F. & Sweller, J. (2010). Cognitive load theory: Advanced in research on worked examples, animations, and cognitive load measurement. Educational Psychological Review, 22, 375-378

Weinstein, C.E., & Mayer, R.E. (1986). The teaching of learning strategies. In M. Wittrock, (Ed.) Handbook of Research on Teaching. New York: MacMillan Publishing. (p.315-327).

Winograd, P. & Johnston, P. (1982). Comprehension monitoring and the error detection paradigm. Journal of Reading Behavior 14, 61-70.

Wittrock, M. C. (1974). A generative model of mathematics education. Journal of Research in Mathematics Education, 5(4), 181-196.

Wittrock, M. C. (1974). Learning as a generative process. Educational Psychologist, 19(2), 87-95.

Wittrock, M. C. (1974). Learning as a generative process. Educational Psychologist, 11, 87-95.

Wittrock, M. C. (1989). Generative processes of comprehension. Educational Psychologist, 24, 345–376.

Wittrock, M. C. (1992). Generative teaching: An enhancement strategy for the learning of economics in cooperative groups. American Educational Research Journal, 29, 861-876.

Young, J.D. (1996). The effect of self-regulated learning strategies on performance in learner controlled computer-based instruction. Educational Technology Research and Development 44, 17-27.

Zimmerman, B. J. (2000). Attaining self-regulation: A social cognitive perspective. In M.Boekaerts & P. R. Pintrich (Eds.). Handbook of self-regulation (pp. 13–39). New York: Academic Press.