Cognitive Load and Learner Expertise: Split-Attention and Redundancy Effects in Reading

English as a Second Language with Vocabulary Definitions

Alexander Seeshing Yeung

University of Western Sydney, Macarthur

Paper presented at the annual conference of the Australian Association

for Research in Education

in Brisbane, Australia, 30 November to 4 December 1997

Abstract

Two experiments were conducted to examine the effects of cognitive load

management using vocabulary definitions in reading passages for readers

of English as a second language (ESL) with different levels of

expertise. Experiment 1 found that vocabulary definitions integrated

within a passage (integrated format) improved 5th-grade ESL learners'

comprehension (high-level processing) whereas a separate vocabulary

list (separated format) improved vocabulary learning (low-level

processing). Experiment 2 found a reverse format x level interaction

effect with secondary school ESL learners. The efficiency of

instruction depends partly on its ability to manage cognitive load

associated with the learning task. An identical presentation format

may facilitate learning by reducing cognitive load but may interfere

with learning either through split-attention or redundancy effects

depending on learner expertise.

Working memory limitations may have an important impact on learning in

complex areas (e.g., Just & Carpenter, 1992; Paas & Van Merrienboer,

1994; Sweller & Chandler, 1994). Cognitive load has been identified as

a major factor to be considered in instructional design in areas such

as science (e.g., Chandler & Sweller, 1991), geometry (e.g., Paas &

Van Merrienboer, 1994; Mousavi, Low, & Sweller, 1995), language (Just &

Carpenter, 1992), as well as in technical instructions (e.g., Sweller,

Chandler, Tierney, & Cooper, 1990; Chandler & Sweller, 1996), and

statistics (Paas, 1992). The effectiveness of instructional design is

often dependent upon its ability to reduce unnecessary cognitive load.

The purpose of this study is to examine the cognitive load effects when

students are given vocabulary definitions for reading comprehension and

vocabulary learning.

The effects of using vocabulary meanings in a reading text may be

predicted using cognitive load theory (Sweller, 1993, 1994). In modern

reading material for young readers, a separate glossary of word

meanings is often found. This separate glossary may enable the learner

to acquire word meanings; but if these meanings must be used for

understanding the text, the process becomes more complex. When an

unfamiliar word is encountered, students need to leave the text, turn

to the vocabulary list, temporarily store the meaning of the word, and

then revert to the text and try to incorporate it into the passage.

This process of attending to two distinct sources of information may

impose a high cognitive load, referred to as the split-attention

effect. The effect occurs when learners must split their attention

between and mentally integrate multiple sources of information. It has

been shown to be a problem with some instructional designs (e.g.,

Chandler & Sweller, 1991, 1992; Sweller & Chandler, 1991, 1994;

Sweller, Chandler, Tierney, & Cooper, 1990; Tarmizi & Sweller, 1988;

Mousavi, Low, & Sweller, 1995).

In order to ameliorate the cognitive load due to split attention, some

material which requires mental integration of separate sources of

information may be modified such that they become physically integrated

(e.g., Chandler & Sweller, 1991, 1992; Sweller & Chandler, 1994;

Sweller, Chandler, Tierney, & Cooper, 1990; Ward & Sweller, 1990).

However, although this technique has proved to be useful in maths and

science instruction (e.g., Tarmizi & Sweller, 1988; Sweller &

Chandler, 1994), it has never been tested in reading instruction. The

technique involves physically integrating, in this case, a text and

word definitions so that working memory load due to mental integration

is greatly reduced.

Instead of using a separate vocabulary list, word definitions can be

provided within a passage. Although this technique is not commonly

used in English texts, it used to be found in some classic Chinese

literature. The effectiveness of this integrated format of vocabulary

meaning presentation has never been seriously assessed, it may be

expected to avoid the possible split attention of a separated format,

such as a separate glossary. In the integrated format, the reader has

direct access to the meanings of unfamiliar words. Immediacy of

processing may reduce storage demands in information processing

(Carpenter & Just, 1983; Just & Carpenter, 1980); thus when vocabulary

meanings are readily available, the cognitive load involved in

temporary storage should be considerably lowered. Reinking and Rickman

(1990) have also demonstrated that physically placing word meanings

close to target words may improve comprehension.

Historically there have been debates on whether reading is a bottom-up

process that emphasises prerequisites at lower-level processes such as

word recognition for successful comprehension; a top-down process that

proposes that higher levels of processing, such as semantic and

pragmatic computations, drive and direct lower-level processing such as

word recognition; and an interactive-compensatory process that

considers the interaction of processes at both higher and lower levels

(Goodman, 1976; Gough, 1972; LaBerge & Samuels, 1974; Stanovich, West,

& Feeman, 1981). Particularly from the perspective of the

interactive-compensatory model (Stanovich, 1980; Stanovich, West, &

Feeman, 1981), processing at either a higher or lower level reduces the

resources available to the other level (see also Glanzer, Fischer, &

Dorfman, 1984; Just and Carpenter, 1992; Kintsch, 1988; van Dijk &

Kintsch, 1983). Accordingly, unless cognitive load reduction is

substantial and to such an extent that sufficient mental resources are

available for thorough processing of information at both higher and

lower levels, while an integrated format should benefit comprehension,

but at the expense of vocabulary learning. In contrast, the separated

format may result in better vocabulary learning because it is ideally

formatted for that purpose, but at the expense of comprehension

performance due to split attention.

However, facilitative effects of the integrated technique in

comprehension should occur only if the vocabulary meanings are

essential for comprehension. A student who does not need assistance

may only be compelled to process some redundant information. The

redundancy effect is another major obstacle to schema acquisition

(e.g., Bobis, Sweller, & Cooper, 1993, 1994; Chandler & Sweller, 1991;

Sweller & Chandler, 1994). Unlike the split-attention effect that

occurs when multiple sources of information are essential for

comprehension, the redundancy effect occurs when the learner is

required to process non-essential information. It is this processing of

unnecessary information that imposes an undue cognitive load.

If the occurrence of the redundancy or split-attention effects is

dependent on whether a particular source of information is necessary,

then the students' levels of expertise may influence the occurrence of

either effect. Whereas a particular source of information may be

essential for less experienced learners and therefore need to be

integrated with other information in order to reduce split-attention,

for more experienced learners, the same source of information may be

redundant.

When vocabulary definitions are given in an integrated format to a

young or low-ability reader, more mental resources may be available for

comprehension, due to the reduction of split attention. A test of word

meanings, however, may find superiority of the separated format because

word meanings separated from the text can be learnt without devoting

cognitive resources to the text. At least to some children, vocabulary

meanings may be more easily learnt from a separate list (Nicholson,

1991). Thus if students are learning word meanings only, the text may

be a source of redundancy, but cannot be ignored easily in the

integrated format.

Higher-ability readers, in contrast, may find the word meanings

redundant and when required to process it, comprehension may be

weakened because of an increase in cognitive load. Because the

redundant information is easier to ignore when word meanings are

presented in a separate list, comprehension scores should be higher for

a separated format whereas reverse results should be obtained on word

meaning scores. Students may be able to comprehend words in context

but may have difficulty emitting a meaning when requested. This task

may be easier if word meanings are attended to when integrated with the

text. Essentially, the integrated format should be better for word

meanings because learners are less likely to ignore them. In sum,

vocabulary meanings presented in an integrated format may either

increase or decrease cognitive load depending on the expertise of the

readers and the type of task.

Experiment 1

Experiment 1 examined the effects of vocabulary definitions presented

in either an integrated or a separated format using primary school

children. The purpose was to examine the effects at both the lexical

and semantic levels of comprehension. There were three conditions: (a)

vocabulary definitions integrated within the passage, (b) vocabulary

definitions listed at the end of the passage, and (c) a control

condition without any vocabulary definitions provided. It was

hypothesized that the integrated condition would improve performance in

comprehension whereas the separated condition would improve performance

in vocabulary meaning recall, compared to the control; and that a

separate list of vocabulary meanings would enhance vocabulary meaning

recall but interfere with comprehension because of a split-attention

effect, whereas the integrated definitions would facilitate

comprehension but interfere with vocabulary meaning recall because the

presence of a passage may distract attention away from precise

vocabulary meanings. Apart from a comparison between the two

experimental conditions and the control group, the focus of this

experiment was, therefore, the interaction effect between the

integrated and separated conditions at two levels of comprehension

processing, i.e., vocabulary learning and passage comprehension.

Method

Participants

The participants were 134 Primary 5 students of a primary school in

Hong Kong, ages ranging from 10 to 11 years, learning English as a

second language. They were randomly assigned to three groups.

Materials and Procedure

A passage taken from a standardised comprehension test of the

Australian Council for Educational Research (ACER, 1986) for Years 5

and 6 (227 words) was used (see Appendix). Questions were added to the

original comprehension test paper so that there were ten

multiple-choice questions. Familiar words were used in the questions

(all had frequencies greater than 50 per million and two ESL teachers

agreed that the questions would not be too difficult for these

students). The vocabulary test for each passage was a list of 12

target vocabulary items in a random order for different students.

Participants were randomly assigned to three groups in each of three

classes. Vocabulary definitions were inserted above target words in the

text for the learning phase in the integrated condition and a list of

meanings for the target words were provided at the end of the text for

the separated condition. Intact classes were used to present materials

and test students.

Integrated condition. During the learning phase, the meaning of each

vocabulary item with a frequency of less than 50 per million (Thorndike

& Lorge, 1944) was given immediately above the vocabulary item in the

passage. For example, immediately above the target word "twisted", its

meaning "turned with force" was printed.

Separated condition. Vocabulary meanings were placed in the form of a

list immediately after the passage. For example, the target word

"twisted" in the passage was printed in a separate glossary with its

meaning "turned with force" printed next to it.

Control. The passage for the control condition was the original

passage without any definitions given.

Testing phase. During the testing phase, the original passage was used

without any word meanings given. A comprehension test with 10

multiple-choice questions and a vocabulary test with 12 vocabulary

items were used to test comprehension and vocabulary meaning recall

respectively. The test items of both tests were arranged in four

different, random orders to minimise the possibility of students

copying from each other. All procedures in the testing phase were the

same across conditions.

Results and Discussion

The mean correct responses of students (in percentages) and standard

deviations in the integrated and separated conditions are shown in

Table 1. The data were analysed with a 2 (level: vocabulary and

comprehension) x 3 (format: integrated, separated, and control)

multivariate analysis of variance (MANOVA) with repeated measures in

the level dimension, followed by a simple contrast for the interaction

effect between the integrated and the separated conditions. Oneway

analyses of variance (ANOVA) were then used to examine the main effects

of format at the vocabulary and comprehension levels separately. All

statistical analyses were conducted at the .01 level of significance

throughout this paper. Results of the MANOVA indicated that the main

effect of format was significant, F (2, 131) = 8.82, MSE = 554.73 but

the main effect of level was nonsignificant, F (1, 131) = 0.43, MSE =

251.28. The treatment x level interaction was statistically

significant, F (2, 131) = 25.55, MSE = 251.28. The focus of this study

on the format x level interaction effect was tested using simple

contrasts. A significant format x level interaction effect was found

between the integrated and separated formats (t = 6.97). Although not

of theoretical interest and is presented only for completeness, the

format x level interaction effect was also significant between the

integrated format and the control (t = 4.86).

Oneway ANOVA at the comprehension level found significant format

effects, F (2, 131) = 25.47, MSE = 305.17, and post-hoc Scheffe tests

showed that students in the integrated condition performed better than

those in the separated or control conditions. Oneway ANOVA at the

vocabulary level found significant format effects, F (2, 131) = 7.07,

MSE = 500.84, and post-hoc Scheffe tests showed that students in the

separated condition performed better than those in the other

conditions.

Table 1

Means and standard deviations of comprehension and vocabulary test

scores (in percentages) of 134 fifth-grade ESL students in three

conditions in Experiment 1

______________________________________________________________

Integrated Separated Control

______________________________________________________________

Performance

n 44 38 52

Comprehension M 45.91 26.05 21.35

SD (17.16) (18.39) (17.04)

Vocabulary M 28.22 42.98 25.96

SD (21.21) (19.13) (25.33)

_______________________________________________________________

As expected, integrated vocabulary definitions assisted the readers in

comprehension of the passage but interfered with the learning of

vocabulary meanings which were learnt best as a separate list.

Although one might speculate that both conditions providing vocabulary

definitions should outperform the control without such provision, the

integrated format of vocabulary definitions did not substantially

improve vocabulary scores and the separated format did not

substantially improve comprehension. These results can be interpreted

by assuming that during text comprehending, the cognitive load involved

in the search for meanings and storage of such meanings is reduced if

vocabulary meanings are readily accessible when integrated with the

text. Under these circumstances, more mental resources can be reserved

for processing at the semantic level. Nevertheless, the separated

definitions enabled better vocabulary meaning recall because for this

purpose, the text becomes redundant. In contrast, during the

processing of the passage at the semantic level, the readers had to

turn to the vocabulary meaning whenever they encountered a new word,

and attention was split. In fact, it was obvious that students in the

separated condition had to turn the pages back and forth to obtain the

meaning of an unfamiliar word from the separate list and then return to

the passage for the processing of meanings in the text. The

split-attention effect resulted in lowered performance in passage

comprehension; but by ignoring the passage when studying the glossary,

performance on the vocabulary test was improved.

Experiment 2

The purpose of this experiment was to examine the effects of vocabulary

definitions using the same integrated and separated techniques of

Experiment 1 with secondary school students. For vocabulary definitions

to be effective, the passage must impose a high cognitive load to

warrant the use of these definitions. Given a reading passage of

moderate or low level of difficulty or if the expertise and experience

of the reader is such that definitions are not essential for

comprehension, the readers may prefer to ignore such definitions. As a

consequence, in contrast to Experiment 1, it was hypothesised that for

students at the secondary school level, the separated definitions would

not result in higher performance in vocabulary meaning recall because

higher-ability readers might not find the supply of word definitions

helpful, and they could use their own strategies and the syntactic and

semantic clues for comprehension instead of making use of the

vocabulary meanings provided. Ignoring a vocabulary list may result in

lower vocabulary scores, even for high-ability readers who, while

understanding vocabulary in context, may derive some benefit from

vocabulary items when required to emit vocabulary meanings. In the

integrated format of presentation, the vocabulary meanings may be

difficult to ignore until the reader has read them and found that they

are not helpful in understanding the text. Processing of unnecessary

information can result in a redundancy effect which hampers learning

(Bobis, Sweller, & Cooper, 1993; Sweller & Chandler, 1994). Therefore,

contrary to the results obtained with 5th-graders in Experiment 1,

integrated definitions used with 8th-graders can be expected to yield

lower performance in passage comprehension when compared with separated

definitions. Unlike the situation of younger readers in Experiment 1,

who required the vocabulary meanings for comprehension, the separated

definitions can be ignored by 8th-graders, depressing scores somewhat

on a vocabulary test, but eliminating a split-attention effect for

comprehension. The focus of this experiment is therefore the format x

level interaction effect between the integrated and separated formats.

Method

Participants

The participants were 126 Year 8 ESL students in a school in Hong Kong.

They were randomly assigned to three groups.

Materials and Procedure

A passage taken from a standardised comprehension test of the ACER

(1986) for Years 7 and 8 was used (300 words). Ten questions were

asked in the multiple-choice comprehension test, and the vocabulary

test was a list of 12 target vocabulary items in a random order. The

procedure was similar to that in Experiment 1.

If cognitive load is a function of an interaction between ability and

the material, then results using 8th-graders should exhibit a reverse

pattern to 5th-graders in that the integrated definitions may induce a

redundancy effect that could hamper comprehension whereas a separate

list would eliminate this redundancy effect for passage comprehension

but ignoring the definitions would result in lower vocabulary scores.

Accordingly, a reverse format x level interaction was hypothesised.

Results and Discussion

Table 2 shows the means and standard deviations of comprehension and

vocabulary performance in percentages. Similar to Experiment 1, for

performance scores, the data were analysed with a 2 (level: vocabulary

and comprehension) x 3 (format: integrated, separated, and control)

multivariate analysis of variance (MANOVA) with repeated measures in

the level dimension, followed by a simple contrast for the interaction

effect between the integrated and the separated conditions. Oneway

analyses of variance (ANOVA) were then used to examine the main effects

of format at the vocabulary and comprehension levels separately.

Results of the MANOVA indicated that the main effect of format was

significant, F (2, 123) = 15.16, MSE = 261.92 but the main effect of

level was nonsignificant, F (1, 123) = 3.25, MSE = 193.79. The

treatment x level interaction was statistically significant, F (2, 123)

= 23.56, MSE = 193.79. Again, the focus of this study on the format x

level interaction effect was tested using simple contrasts. A

significant format x level interaction effect was found between the

integrated and separated formats (t = -2.97); and also between the

integrated format and the control (t = -6.84) although this interaction

effect is not of theoretical interest and is not a focus of the present

study.

Table 2

Means and standard deviations of comprehension and vocabulary test

scores (in percentages) of 126 eighth-grade ESL students in three

conditions in Experiment 2

______________________________________________________________

Integrated Separated Control

______________________________________________________________

n 42 42 42

Comprehension M 22.86 25.71 24.29

SD (14.53) (15.32) (16.10)

Vocabulary M 40.08 30.16 12.10

SD (19.93) (14.83) ( 6.69)

_______________________________________________________________

Oneway ANOVA at the comprehension level found that the format effects

were nonsignificant, F (2, 123) = 0.36, MSE = 235.08. However, oneway

ANOVA at the vocabulary level found significant format effects, F (2,

123) = 38.30, MSE = 220.63, and post-hoc Scheffe tests showed that

students in both experimental conditions performed better than those in

the control group, and the integrated condition outperformed the

separated condition.

As expected, the format x level interaction effect for performance

scores was significant and in a reverse direction as in Experiment 1

with 5th-graders. There is support for the differential cognitive load

effects on different cognitive tasks for learners with different

expertise. Perhaps, the list of vocabulary meanings was ignored and

treated as if it were a passage in the control. In contrast, the

integrated definitions provided vocabulary meanings directly above the

target words making them more difficult to ignore. To the 8th-graders

using the integrated format, the information contained in the

vocabulary definitions may have been largely redundant for the purpose

of comprehension. This redundancy resulted in reduced scores (even

lower than the control group, although not statistically significant)

because the partly redundant vocabulary definitions were unnecessarily

processed, but increased scores on the vocabulary test because the

definitions could hardly be ignored in this format.

In summary, the occurrence of the redundancy or split-attention effects

may be dependent on whether a particular source of information is

necessary for intelligibility. Therefore, levels of expertise or

experience might influence whether one or the other effect might occur.

The reverse directions of format x level interactions in Experiments 1

and 2 provided evidence that it is not the presentation format of

instructional material per se that causes cognitive load differences.

General Discussion

The present investigation found that instructional formats used by

different learners could yield quite different effects. The two

experiments showed that an identical format of instruction yielded

reverse effects when used by younger or older ESL learners. Using less

expert readers in Experiment 1, by eliminating split-attention, the

integrated format reduced the cognitive load related to search for

meaning and thus improved comprehension. Nevertheless, while

comprehension was improved by an integrated format, vocabulary learning

was impeded. Vocabulary learning was learnt best when the vocabulary

meanings were separated from the text. Under these circumstances, the

text acted as a redundant source of information imposing an extraneous

cognitive load (Sweller, 1993). In contrast, the more experienced

readers in Experiment 2 did not require the vocabulary meanings to

comprehend the text. When presented in an integrated format, the

meanings were redundant but hard to ignore. They were easier to ignore

in a separated format. The redundancy effect increased cognitive load

and decreased comprehension when the material was processed in an

integrated format. While attending to vocabulary meanings in the

integrated format reduced comprehension for more experienced readers,

increased knowledge of vocabulary might be expected and was obtained

when compared to the separated format that permitted students to ignore

vocabulary.

The interaction effects found in the two experiments cannot be

explained solely in terms of the nature of the material, or the

presentation format, or the learners' abilities. The reverse directions

of the interactions provide evidence of effects which have to take into

account both the material and the readers' expertise. Because the

focus of the present study is on the interaction effects that were

hypothesised to be in opposite directions for 5th and 8th grade ESL

learners, whether the main effects are significant is not a main

concern. Performance may improve only when cognitive load is reduced

substantially and to such an extent that mental resources are readily

available for information processing. Hence, given a complex cognitive

task such as comprehension in a second language, the extent of

cognitive load reduction through instructional techniques may not

result in dramatic improvement. Furthermore, because comprehension

requires complex processing at the lexical, syntactic, and semantic

levels almost simultaneously, reduction of cognitive load by providing

vocabulary definitions may not be sufficient for improving performance

to a substantial extent. The particularly low performance of the 8th

graders in comprehension (M = 24.29%) partly reveals the difficulty of

the task.

The greatest limitation of this study is probably related to direct

measurements of cognitive load. Whereas cognitive load theory is used

to explain the results, direct measures of cognitive load was not

available. However the pattern of performance scores and the

interaction effects found in the present experiments can be well

interpreted in terms of the split-attention and redundancy effects.

Even though this does not preclude alternative interpretations of the

results, the findings of this investigation have important implications

for instructional design. An instructional method or a specific format

of presentation may result in facilitation when used with some learners

but retardation with different learners. Specifically, for less

experienced readers, the integrated format, compared to the separated

format, reduced split-attention effects for comprehension but induced

redundancy effects for vocabulary learning. In contrast, for more

experienced readers, who did not need vocabulary definitions for

comprehension, the integrated format induced redundancy effects and

hampered performance in comprehension. Thus, when using vocabulary

definitions to assist reading activities, it is essential to consider

the expertise and previous knowledge of the readers in order to use

either an integrated or separated format of presentation for a specific

purpose. A detailed analysis of both the material used and the

students who will be using it is required before an ideal instructional

format is devised. This analysis will benefit students of differential

expertise and experiences in reading activities as well as in other

learning activities. There is at least some evidence that cognitive

load theory can assist in this analysis.

References

Australian Council for Educational Research (1986). Progressive

achievement tests in reading (2nd ed.). Victoria: Brown prior Anderson.

Bjorklund, D. F., & Harnishfeger, K. K. (1990). The resources construct

in cognitive development: Diverse sources of evidence and a theory of

inefficient inhibition. Special Issue: Limited resource models of

cognitive development. Developmental Review, 10, 48-71.

Bobis, J., Sweller, J., & Cooper, M. (1993). Cognitive effects in a

primary-school geometry task. Learning & Instruction, 3, 1-21.

Bobis, J., Sweller, J., & Cooper, M. (1994). Demands imposed on

primary school students by geometric models. Contemporary Educational

Psychology, 19, 108-117.

Carpenter, P. A., & Just, M. A. (1983). What your eyes do while your

mind is reading. In K. Rayer (Ed.), Eye movements in reading:

Perceptual and language processes (pp. 275-307). San Diego, CA:

Academic Press.

Chandler, P., & Sweller, J. (1991). Cognitive load theory and the

format of instruction. Cognition and Instruction, 8, 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.

Chandler, P., & Sweller, J. (1996). Cognitive load while learning to

use a computer program. Applied Cognitive Psychology, 10, 151-170.

Glanzer, M., Fischer, B., & Dorfman, D. (1984). Short-term storage in

reading. Journal of Verbal Learning and Verbal Behavior, 23, 467-486.

Goodman, K. S. (1976). Reading: A psycholinguistic guessing game. In

H. Singer & R. Ruddell (Eds.), Theoretical models and processes of

reading. Newark, DE: International Reading Association.

Gough, P. B. (1972). One second of reading. In J. F. Kavanagh & I. G.

Mattingly (Eds.), Language by ear and eye. Cambridge: MIT Press.

Just, M. A., & Carpenter, P. A. (1980). A theory of reading: From eye

fixation to comprehension. Psychological Review, 87, 329-354.

Just, M. A., & Carpenter, P. A. (1992). A capacity theory of

comprehension: Individual differences in working memory. Psychological

Review, 99, 122-149.

Kintsch, W. (1988). The use of knowledge in discourse processing: A

construction-integration model. Psychological Review, 95, 163-182.

LaBerge, D., & Samuels, S. J. (1974). Toward a theory of automatic

information processing in reading. Cognitive Psychology, 6, 293-323.

Mousavi, S., Low, R., & Sweller, J. (1995). Reducing cognitive load by

mixing auditory and visual presentation modes. Journal of Educational

Psychology, 87, 319-334.

Nicholson, T. (1991). Do children read words better in context or in

list? A classic study revisited. Journal of Educational Psychology,

83, 444-450.

Paas, F. G. W. C. (1992). Training strategies for attaining transfer of

problem-solving skill in statistics: A cognitive-load approach.

Journal of Educational Psychology, 84, 429-434.

Paas, F. G. W. C., & Van Merrienboer, J. J. G. (1994). Variability of

worked examples and transfer of geometrical problem-solving skills: A

cognitive-load approach. Journal of Educational Psychology, 86,

122-133.

Reinking, D., & Rickman, S. S. (1990). The effects of

computer-mediated texts on the vocabulary learning and comprehension of

intermediate-grade readers. Journal of Reading Behavior, 22, 395-411.

Stanovich, K. E. (1980). Towards an interactive-compensatory model of

individual differences in the development of reading fluency. Reading

Research Quarterly, 16, 32-71.

Stanovich, K. E., West, R. F., & Feeman, D. J. (1981). A longitudinal

study of sentence context effect in second-grade children: Tests of an

interactive-compensatory model. Journal of Experimental Child

Psychology, 32, 185-199.

Sweller, J. (1993). Some cognitive processes and their consequences

for the organisation and presentation of information. Australian

Journal of Psychology, 45, 1-8.

Sweller, J. (1994). Cognitive load theory, learning difficulty and

instructional design. Learning and Instruction, 4, 295-312.

Sweller, J., & Chandler, P. (1991). Evidence of cognitive load theory.

Cognition & Instruction, 8, 351-362.

Sweller, J., & Chandler, P. (1994). Why some material is difficult to

learn. Cognition and Instruction, 12, 185-233.

Sweller, J., Chandler, P., Tierney, P., & Cooper, M. (1990). Cognitive

load and selective attention as factors in the structuring of technical

material. Journal of Experimental Psychology: General, 119, 176-192.

Tarmizi, R., & Sweller, J. (1988). Guidance during mathematical

problem solving. Journal of Educational Psychology, 80, 424-436.

Thorndike, E. L., & Lorge, I. (1944). The teacher's word book of

30,000 words. New York: Columbia University.

van Dijk, T. A., & Kintsch, W. (1983). Strategies of discourse

comprehension. San Diego, CA: Academic Press.

Ward, M., & Sweller, J. (1990). Structuring effective worked examples.

Cognition and Instruction, 7, 1-39.

12

Cognitive Load