The Image of Images as an Aid to Improve Learning

An Eye-tracking Experiment Studying the Effect of Contrasts

in Computer-based Learning Material

Mona Holmqvist Olander

, Eva Wennås Brante

and Marcus Nyström

Department of Pedagogical, Curricular and Professional Studies, University of Gothenburg, Gothenburg, Sweden

School of education and environment, Kristianstad University, Kristianstad, Sweden

The Faculties of Humanities and Theology, Lund University, Lund, Sweden

Keywords: Text-picture Integration, Eye-tracking, Dyslexia, Variation Theory, Contrasts in Text and Pictures.

Abstract: This study addresses differences in the design of computer-based learning materials—text with or without

pictures—and the aim is to show in what way these differences affect learning outcomes. In total, 46 young

adults participated: 19 with dyslexia and 27 controls. Approximately half of each group received the

condition text only, and half received the text and an integrated picture. The learning material was presented

on computer screens, and the participants’ viewing patterns were registered by eye-tracking. The

respondents answered text-based and picture-based questions, as well as oral questions, during the

experiment. The assumptions about learning are based on variation theory and on the importance of

contrasts in discerning important aspects of the learning material. The results show that whether material of

the same content (surrealism) is presented in text only (without explicit contrasts embedded in the text) or in

text and picture form (which offers a contrast) affects learning outcomes, particularly for the participants

with dyslexia, who showed a fourfold increase from pre-test to post-test (from .10 to .40).

1 INTRODUCTION

There is an assumption that pictures in learning

material support learning, and pictures are frequently

used in computer-based learning material. But what

happens when learners meet both text and pictures at

the same time, as in information graphics?

Holsanova et al. (2009) have shown that how

pictures are integrated with a text is vital for an

average reader. When text and pictures are closely

integrated, it seems that more of the text is read.

Holsanova et al. (2009) measured the number of

integrative saccades—that is, “transitions between

semantically related pieces of verbal and pictorial

information” (Holsanova et al., 2009, p.8)—and

thereby found that the quantity of integrative

saccades was reduced when the text was separated

from the pictures. Furthermore, when pictures and

text are serially arranged (as in a classic comic strip)

instead of radially arranged (with a large picture in

the middle and small pictures and text like satellites

arranged around it), the reader tends to read more

and for a longer time, and integrative saccades were

almost twice as common. One possible explanation

for the longer reading time spent on the serially

arranged pictures and text was, according to

Holsanova et al. (2009), that the pictures and the text

were arranged in the direction in which people in

Western society read, from left to right. Another is

that the text was arranged in a helpful way: it started

with general knowledge and built on it with more

and more facts. It seems obvious that the way the

pictures are exposed, as well as what kind of

pictures they are and what they contain, is of

importance in terms of where viewers look and for

how long. But in what way this contributes to a

deeper understanding of the written text has not yet

been clarified.

The results presented in this paper originate from

a study investigating the way computer-based

learning materials with and without pictures affect

learning. This was tested in two different groups of

subjects, namely, young adults with and without

dyslexia. Based on the findings in the main study,

further analysis was conducted to describe in what

way contrasts in the learning material seem to

influence the learning outcome more than the

309

Holmqvist Olander M., Wennås Brante E. and Nyström M..

The Image of Images as an Aid to Improve Learning - An Eye-tracking Experiment Studying the Effect of Contrasts in Computer-based Learning

Material.

DOI: 10.5220/0004960303090316

In Proceedings of the 6th International Conference on Computer Supported Education (CSEDU-2014), pages 309-316

ISBN: 978-989-758-021-5

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

presence or absence of pictures in the computer-

based learning material does. The aim of the paper is

to study how contrasts either within text only or

between text and picture affect young adults’

abilities to discern and learn from computer-based

learning material presented in an experiment about

art genres. The study is based on variation theory,

and a multi-method approach (Yin, 2006) has been

used to measure both the participants’ viewing by

using eye-tracking and their expressed knowledge

by using multiple-choice questions and oral

questions during the experiment.

2 THEORETICAL FRAMEWORK

In this paper we take the learning theory approach of

variation theory (Holmqvist, Gustavsson and

Wernberg, 2008; Holmqvist, 2011; Marton and

Booth, 1997; Pang and Ling, 2010; Runesson, 1999)

to study which factors are necessary for learning

under the two different conditions text only versus

text plus picture in two different groups of

participants, young adults with and without dyslexia.

Variation theory supports dual coding theory

(DCT; Clark and Paivio, 1991) by assuming that

processing printed words and images requires the

simultaneous discernment of a part-whole

relationship in the object of learning (Svensson,

1976; Gibson and Pick, 2000; Gärdenfors and

Johansson, 2005). It also assumes that variety should

be designed so as not to overload learners’ memory

systems (Holmqvist, 2004). Information via the two

senses of hearing and seeing does not overload the

system as it activates two different working

memories (sounds and images), while dual

information delivered to one sensory memory can

cause an overload (Mayer 2005; 2009). Informed by

these studies, our main study (Holmqvist Olander,

Nyström, Wennås Brante, manuscript) used only

printed words and images—that is, only visual

memory—to examine the effect of the part-whole

relationship on learning through printed words and

images.

2.1 Variation Theory

Variation theory holds that learning takes place

through contrast (to make it possible for the learner

to separate an aspect from the object),

generalization, and fusion. For example, in art

genres, which we used in this study, one feature of

surrealism is that the paintings are not depictions of

reality. Contrasting a surrealistic painting with a pop

art painting achieves a contrast regarding the

specific aspect of whether the work is reality-based.

Through the contrast, it becomes possible to separate

this feature from the painting as such. This aspect

can be generalized to other art genres, such as

impressionism, and it becomes possible for the

learner to decide that Sunrise by Claude Monet does

not belong to the genre of surrealism because the

painting does depict reality, even if the expression of

it is a bit hazy. Such contrasts can be offered to the

learner in various ways; in our case, we chose to

present the contrasts either in text or in a

combination of pictures and text so as to study

whether and how the different types of material

facilitate or complicate learning.

2.2 Variation Theory and Reading

To read is to combine signs with sounds, putting

sounds together into meaningful words and

interpreting them in the context in which they are

found. Drawing on variation theory about how

knowledge is attained (Marton and Tsui, 2004), a

person who is reading needs to simultaneously

experience both the parts—that is, the phonemes

represented by one or more letters—and the

wholeness—that is, the words in that particular text

as well as the entire text itself. One example

(Wennås Brante, 2010) involves the difficulties

involved in differentiating b from d. If a child sees a

dog, it does not matter whether the dog turns around;

it remains the same dog. But the signs c and l are the

parts that make up wholes as they can be used both

independently and in any combination. The

combinations can be b, d, p and q. The critical

feature is to know the importance of the parts’

directions in the combination so as to understand

that the letters differ and not are the same when seen

from another perspective. To make the distinction,

learners must discern the parts simultaneously with

their directions. If they learn only about b, there is a

risk that a d or p will be interpreted as a b the first

times the learners see the new letters. If they learn

about b, d, p, and q at the same time, the variation

needed to discern the importance of the parts’

directions will be offered. Based on this example, it

is easy to understand the complexity of the critical

features that reading comprehension includes.

Through the use of interviews and eye-tracking it is

possible to ask participants what they have actually

been aware of and can express about the text, at the

same time as we study what they could have been

aware of by following their eye gazes. This differs

from what they have been presented, as we do not

CSEDU2014-6thInternationalConferenceonComputerSupportedEducation

310

know which parts of the text they seem to neglect

and which parts they seem to focus on.

2.3 Picture Viewing

Perceived objects are affected by visual factors as

contrasting or semantic factors (Henderson, 2003;

Nyström and Holmqvist, 2008). This means that

people use their knowledge about the world to gaze

intelligently through a real-world scene (Henderson,

2003). They seem to do the same thing when

looking at a picture. Yarbus (1967) showed in a

famous eye-tracking study that a task influences the

visual pattern. When the task was to find out

relationships between people in a picture, one kind

of a scan-path emerged which differed from the

scan-path emerging when the task was to find out

whether, for example, the people in the same picture

have eaten dinner. Viewers tend to know what they

are looking for and steer their fixations in that

direction, according to Yarbus (1967).

Brockmole and Boot (2009) have reported that

task-irrelevant aspects of visual stimuli also

influence the fixation durations. This complicates

the use of pictures as comprehension aids because it

is hard to ensure that every picture tied to a text has

only task-relevant aspects. If it does not and if the

reader encounters problems comprehendingthe parts

simultaneously with the whole in a single gaze, the

pictures become just one more part to be distracted

by. Furthermore, a study by Nummenmaa et al.

(2006) shows that people seem to allocate attention

to emotional pictures rather than to neutral ones.

Even when instructed to attend to neutral pictures,

fixations on the emotional ones form. This means,

according to Nummenmaa et al. (2006), that

orientation towards emotional stimuli is highly

exogenous. One cannot, in other words, completely

steer one’s attention to the pictures that are perhaps

most relevant or that contain the most information;

one is drawn to emotional pictures. Text and picture

integration can thereby be efficient or misleading,

depending on which pictures are presented.

A study (Hannus and Hyönä, 1999) of 10-year-

old children showed that high-ability children paid

more attention to pictures in a text and could make

better use of them than low-ability children did and

could. Comprehension scores were also higher for

high-ability children when they encountered a text

with illustrations, but comprehension scores for low-

ability children did not improve under the same

conditions (Hannus and Hyönä, 1999).

Holsanova et al. (2009) showed that how pictures

are integrated with a text in information graphics is

vital for an average reader. The way pictures are

designed and offered, as well as what they contain,

is important in terms of where a viewer looks and

for how long. However, this does not tell anything

about the viewer’s awareness. How the viewing

pattern contributes (or does not) to a deeper

understanding of the written text has not yet been

clarified.

A person reading a text processes it visually,

phonologically, and semantically at the same time.

When reading a text integrated with pictures, the

person attends to the pictures, as well. What does

this mean for dyslexic readers’ reading

comprehension?

3 PREVIOUS FINDINGS

The results of the main study show that there are

significant differences between the two groups of

participants and interesting differences between the

two different conditions. The results show

significant differences in both comprehension scores

and visual behaviours between respondents with and

without dyslexia in the two different conditions (text

only or text plus picture). Respondents with dyslexia

did not use strategies to gain a global overview of

the information, something that the participants

without dyslexia did more quickly. This trend

supports the hypothesis that the participants with

highest learning outcomes used global-overview

strategies. Participants with dyslexia showed

haphazard viewing patterns, strengthening the

assumption that a global overview as a strategy

promotes learning (Holmqvist Olander, Nyström &

Wennås Brante, manuscript). The condition text and

picture decreased learning for respondents with

dyslexia, with one exception: the example of

surrealism, which this paper examines further.

3.1 Dyslexia and Phonological

Awareness

One pre-requisite to reading is discerning a given

sound among the continuous flow of sounds people

produce when they speak (Shaywitz, Morris and

Shaywitz, 2008) and then mapping the sound to a

letter (Byrne, 2005). This phonological awareness is

one part of the phonological ability needed to read;

thus, a weak phonological ability is considered one

explanatory factor in dyslexia (Goswami and Bryant,

1990; Ramus et al., 2003; Vellutino et al., 2004;

Everatt and Reid, 2009). In the main study,

participants were tested for phonological ability

TheImageofImagesasanAidtoImproveLearning-AnEye-trackingExperimentStudyingtheEffectofContrastsin

Computer-basedLearningMaterial

311

using a screening test called Duvan (Lundberg and

Wolff, 2003). The test is executed in groups and

tests for different sub-skills in phonological ability.

The results from the main study (Holmqvist

Olander, Nyström, Wennås Brante, manuscript)

show a significant difference between the groups,

supporting the assumption that participants with

dyslexia have a lower phonological ability than the

rest of the participants do.

3.2 Viewing Patterns

In the experiment, pictures were inspected for less

time than texts were, yet a difference between

groups was found. The control group performed an

earlier inspection of the picture, thus rendering a

global impression of the material (Holmqvist

Olander, Nyström, Wennås Brante, manuscript).

Viewing analysed only in the condition of text +

picture and the picture was viewed by 95.2% of the

controls and 78.3% of those from the participants

with dyslexia. The participants spent 45.9 seconds

(SD = 2.1) looking at the text and 2.2 seconds (SD =

2.3) viewing the picture. They were looking

elsewhere for 0.8 seconds (SD = 1.0). When

interviewed about viewing patterns, some

respondents with dyslexia had an understanding of

the way they processed material containing both text

and pictures that was found inconsistent with how

they really placed their gazes (Wennås Brante,

2013). They processed the material in a much more

random way then they thought they did, thus missing

opportunities to gain a global understanding of the

material.

3.3 Contrasts

The texts from the experiment in the main study

contained a contrast formulated in words. The

contrast could be explicit (that is, the two values

were specified and mentioned) or implicit (that is,

one value not was specified). The parts of the texts

containing a contrast were the ones most discerned

(Wennås Brante, Holmqvist Olander and Nyström,

2013), even when the contrasts were of less

importance for learning about the content. If a

contrast was implicit—that is, if a counter-value was

not specified—it was hard for the dyslexic group to

discern it (Wennås Brante, Holmqvist Olander &

Nyström, 2013). The micro-analysis of the oral

answers revealed also that contrasts are exceedingly

effective for discernment (Wennås Brante,

Holmqvist Olander and Nyström, 2013).

The experiment also employed different kinds of

pictures that could influence the viewing behaviour.

Three pictures differing in motif and color were

chosen to test how the viewing behavior differed: a

cubistic picture that had a motif that was difficult to

see in brown nuances, a romantic picture with a

clear motif painted in natural colors, and finally a

surrealistic picture with an odd motif and clear

colors. No great difference was found in viewing

patterns of the cubistic and romantic picture, but

participants without dyslexia had significantly

higher dwell times on the three pictures (p = 0.034),

a result that originated from dwell time on the

surrealistic picture (Wennås Brante, manuscript).

However, most participants with dyslexia inspected

the surrealistic picture before they started reading

the text, and this had an impact on their reading

comprehension: the oral answers from the group

with dyslexia contained more inferences regarding

the surrealistic image than the same group’s answers

about the other art genres did (Wennås Brante,

manuscript).

4 IMPLEMENTATION

The analysis is based on data collected during the

main project (Holmqvist Olander, Nyström and

Wennås Brante, manuscript; Wennås Brante,

Holmqvist Olander and Nyström, 2013, Wennås

Brante, 2013; Wennås Brante, manuscript). So far

no analysis or results have been presented on the

task level regarding the outcomes of the questions in

the experiment.

4.1 Participants and Design

In total, 46 respondents were recruited via an

advertisement on two universities’ websites and in a

residential college for adult education through a

teacher, requesting young adult native speakers of

Swedish with and without dyslexia and with normal

or corrected-to-normal vision. In total, 19 dyslexic

and 27 non-dyslexic (control) respondents were

recruited. Their ages ranged from 19 to 41, with a

mean age of 23.8 years (SD = 4.2).

The reading tasks consisted of six different short

expository texts which demand somewhat different

reading strategies than narrative texts do (Hyönä et

al., 2002), such as understanding the topic and the

hierarchically related sets of sub-topics and their

relationships with one another. Because expository

texts are frequently used in schools and education, it

is of interest to determine whether pictures

contribute to the comprehension of expository texts.

CSEDU2014-6thInternationalConferenceonComputerSupportedEducation

312

Respondents were presented first with the name of a

genre, then with information from either of the two

experimental conditions (text only or text plus

picture). The computer-based learning material in

the two conditions was identical, except from text

only in one condition and text with pictures in the

other condition. In total it consisted of six screens;

1.Title of art genre, 2. Text about the art genre

(including picture in one condition and without in

the other), 3. Blank screen (the researcher asks what

information the respondent met on the previous

screen – oral post-test), 4. Text question (verbal

post-test), 5. Confidence rating and 6. Multiple-

choice question illustrated by three paintings

(picture post-test). When the respondents had

studied the information at the second screen and

were ready to answer questions about it, they clicked

the mouse, and an empty screen appeared. While

instructed to keep looking towards the screen area,

respondents were by one researcher asked “What

information about [genre] appeared on the last

screen?” Their verbal answers were recorded

through a webcam. When they had no more

information to provide, they were asked to click the

mouse and proceed to answer three multiple-choice

questions with only one of the alternative answers.

The first question offered a choice of five text

responses to a question about the genre. In the

second question, respondents were asked to estimate

how confident they were about their previous answer

on a scale of one (very uncertain) to seven (very

certain). Finally, they answered the same questions

again as they had during the initial test, with the

positions of the three pictures randomly shuffled, in

order to assess whether learning had taken place

during the experiment (Holmqvist Olander, Nyström

and Wennås Brante, manuscript).

The eye-tracker used in the main study was an

SMI HiSpeed sampling data binocularly at 500 Hz.

This speed is sufficient to measure saccades

accurately. Stimuli were shown on a 19-inch screen

connected to a PC which also collected each

subject’s eye-tracking. To ensure that it would be

possible to detect which work a person was looking

at, text was presented with letters spanning

approximately half a degree of visual angle. Rows

had double line spacing.

5 RESULTS

The results of the main study show significant

differences among the two groups of participants

regarding both viewing patterns and learning

outcomes (Holmqvist Olander, Nyström, Wennås

Brante, manuscript; Wennås Brante, Holmqvist

Olander, Nyström, 2013; Wennås Brante, 2013,

Wennås Brante, manuscript). However, analysing

the computer based learning material at the task

level shows one exception regarding an item in

which both groups exhibited the same kind of

strategy—namely, the task regarding surrealism.

This finding was first measured by the eye-tracking

equipment (Wennås Brante, manuscript). The results

of the analysis indicate that all participants used the

same kind of viewing pattern when viewing the

painting of surrealism. The analysis of the

participants’ answers on the text and picture

questions supports this initial indication and

demonstrates a correlation between the pattern of

eye movements and learning outcome. The result

contradicts previous findings as it highlights the

impact of the design of the computer-based learning

material regarding illustrations. Pictures or not is not

a simple question that can be stated to improve or

worsen the learning outcome. However, the text

material about surrealism did not offer contrasts that

are significantly pointed out, resulting in fewer

discerned contrasts in the task on surrealism (20–

64%) than in, for instance, the task on

impressionism (65–77%). Further, this difference

between the groups is significant (dyslexic 20–33%,

controls 41–64%) for the surrealism task but not for

the impressionism task (dyslexic 65–72%, controls

65–77%) (Wennås Brante, Holmqvist Olander,

Nyström, 2013). This indicates that contrasts that are

not clearly expressed are harder to discern for those

with dyslexia than they are for the rest.

In the condition text plus picture, the surrealistic

picture itself comprises a contrast with reality as

experienced by the participants; moreover, the

picture is viewed to a higher degree than the other

pictures (Wennås Brante, manuscript). This is true

also for the group of dyslexics, and they do not show

any similarities with the controls’ viewing pattern

regarding the other stimuli. The result shows that the

same learning content presented by two different

representations, text only or text and picture

integrated, is perceived differently as a consequence

of the discerned contrasts offered by the computer-

based learning material. This is shown by the

increased learning outcome regarding the condition

text plus picture, as the participants’ discernment of

the surrealistic painting contrasted with their own

views of reality when they observed the aberrant

painting. It is more difficult to draw such a strong

contrast when present with the text-only stimulus

about surrealism. The results of the pre- and

TheImageofImagesasanAidtoImproveLearning-AnEye-trackingExperimentStudyingtheEffectofContrastsin

Computer-basedLearningMaterial

313

post-tests from the picture question appear in table 1.

The text in the learning material about surrealism

describes objects in a way that differs from their

ordinary use in reality (e.g., ‘Instead, a connection

between, for example, a green hot dog and a gigantic

comb wer necessary’). But not until participants see

the paintings can they fully grasp how different the

image is from what is possible in reality. If the

results of the participants’ answers from the text-

based questions and the picture-based questions

about surrealism are broken out from the material

previously presented, a new and different pattern

emerges in the group of dyslexics regarding this

specific genre in relation to the mean of the

remaining five genres.

Table 1: Means of the results from picture-based question

about surrealism in comparison to the means of the other

five tasks in the two different groups of participants.

Mean

points/

question

Dyslexic

Text only

Non-

dyslexic

Text only

Dyslexic

Text plus

picture

Non-

dyslexic

Text plus

picture

Pre-test

Surrealism

.44 .38

.10

.21

Post-test

Surrealism

.78 .77

.40

.50

Pre-test

Other

genres

.24 .37 .26 .36

Post-test

Other

genres

.60 .66 .68 .63

Only three participants in the entire group failed to

correctly answer the picture question related to

surrealism. Of these three, one came from the group

in which the participants did not see the picture (text

only). The other two were in the group with the

condition text and picture, but one of those had not

fixated at all on the picture (the only respondent to

demonstrate that behavior), and the other one had

fixated on the picture only once.

The results in table 1 show that there is a

learning outcome for the group of participants with

dyslexia in the condition text plus picture regarding

the task on surrealism that differs from the ordinary

pattern—a fourfold increase—more than for any

other genre or condition. In the condition text only,

there is no difference in outcomes between the

questions for surrealism and the questions on the

other art genres; the participants scored roughly

twice as high on the post-tests as on the pre-tests.

Regarding the text-based question about

surrealism, the group of participants with dyslexia

showed slightly higher scores (.80) for the task about

surrealism than the mean scores for the other genres

(.68), but this is far from the fourfold increase

apparent for the picture-based question (table 2).

This supports the assumption about contrast’s

impact on learning: there are implicit contrasts in the

text, resulting in lower scores for the text only

condition for both text- and picture-based questions.

On the other hand, there was a high increase for the

condition text plus picture regarding the picture-

based question about surrealism.

Table 2: Means of the results from the text-based question

about surrealism compared to the mean of the other five

tasks in the two different groups of participants.

Mean

points

Dyslexic

Text only

Non

dyslexic

Text only

Dyslexic

Text plus

picture

Non-

dyslexic

Text plus

picture

Surrealism

.89 1.00 .80 1.00

Other

genres

.89 .91 .68 .86

6 CONCLUSIONS

The main finding in this study is the importance of

purposeful designing computer-based learning

materials in relation to content and learner; in

particular, it is expedient to offer contrasts

embedded in text or pictures in order to make it

possible for the viewer to grasp the meaning of the

information. The results show that participants with

dyslexia are particularly helped by explicit

contrasts—for instance, demonstrating what

something is by also showing what it is not—and

their learning outcomes are also affected to a higher

degree than those of individuals without dyslexia.

Illustrations in computer-based learning material

have to be meaningful and purposeful designed, and

not be used as decorations, to influence learning.

Findings about decreased learning, when using

learning material with illustrations for students with

dyslexia, point out the need of further research about

how illustrations carefully can be used to improve

instead of prevent learning. The results from the eye-

tracking data validate the differences in viewing-

pattern in between the groups of participants and

reveals the strategies used when examining

computer-based learning material.

ACKNOWLEDGEMENTS

This project is funded by the Swedish Research

Council (project number 2010-5379), for which we

CSEDU2014-6thInternationalConferenceonComputerSupportedEducation

314

are grateful. The project is also supported by

University of Gothenburg, Lund University, and

Kristianstad University. We thank all the

participants for their contribution to the study, as

well as professor emeritus Ference Marton,

University of Gothenburg, for valuable comments

through the entire project.

REFERENCES

Brockmole, J. R., Boot, W. R., 2009. Should I stay or

should I go? Attentional disengagement from visually

unique and unexpected items at fixation, Journal of

Experimental Psychology: Human Perception and

Performance, 35(3), 808-815.

Byrne, B., 2005. Theories of learning to read. In M.J.

Snowling & C. Hulme (Eds.) The science of reading: A

handbook, 104–119. Blackwell publishing. Malden, Ma.

Clark, J. M., Paivio, A., 1991. Dual coding theory and

education, Educational psychology review, 3(3), 149-

210.

Everatt, J., Reid, G, 2009. Dyslexia: An overview of

recent research. In G. Reid (Ed.) The Routledge

Companion to Dyslexia, (pp. 3-21). Routledge.

London.

Gibson, E. J., Pick, A. D, 2000. An ecological approach to

perceptual learning and development, Oxford

University Press. USA.

Gardenfors, P., Johansson, P. (Eds.), 2005. Cognition,

education, and communication technology, L.

Erlbaum Associates. Mahwah, N.J.

Goswami, U., Bryant, P, 1990. Phonological skills and

learning to read, Journal of Child Psychology and

Psychiatry, 32 (7), 1173-1176.

Hannus, M., Hyona, J., 1999. Utilization of illustrations

during learning of science textbook passages among

low-and high-ability children, Contemporary

Educational Psychology, 24(2), 95-123.

Henderson, J. M., 2003. Human gaze control during real-

world scene perception, Trends in Cognitive Sciences,

7(11), 498-504.

Holmqvist, M., 2004. En främmande värld. Om lärande

och autism. [A Strange World. About Learning and

Autism.] Lund: Studentlitteratur.

Holmqvist, M., Gustavsson, L. & Wernberg, A., 2008

Variation theory – An organizing principle to guide

design research in education. In Kelly, A.E., Lesh, R.,

&. Baek J. (Eds.) Handbook of design research

methods in education, p 111–130. New York:

Routledge.

Holmqvist, M. 2011. Teachers’ learning in a learning

study. Instructional Science, 39(4), 497–511.

Holmqvist Olander, M., Nyström, M., Wennås Brante, E.,

manuscript. Eye-tracking as an indicator of impaired

executive function in people with dyslexia.

Holsanova, J., Holmberg, N., Holmqvist, K., 2009.

Reading information graphics: The role of spatial

contiguity and dual attentional guidance, Applied

Cognitive Psychology, 23(9), 1215-1226.

Hyönä, J., Lorch Jr, R. F., Kaakinen, J. K., 2002.

Individual differences in reading to summarize

expository text: Evidence from eye fixation patterns,

Journal of Educational Psychology, 94(1), 44-55.

Lundberg, I., Wolff, U., 2003. DUVAN. Dyslexiscreening

för ungdomar och vuxna, Psykologiförlaget.

Stockholm.

Marton, F., Booth, S., 1997. Learning and awareness,

Lawrence Erlbaum. Mahwah, N.J.

Marton, F., Tsui, A., 2004. Classroom discourse and the

space of learning, Lawrence Erlbaum Associates.

Mahwah, NJ.

Mayer, R. E., 2005. Cognitive theory of multimedia

learning. In R. E. Mayer (Ed.), The Cambridge

handbook of multimedia learning, pp. 31-48,

Cambridge University Press. Cambridge.

Mayer, R., 2009. Multimedia principle. In R. Mayer (Ed.),

Multimedia learning 2

ed, pp. 223-241, Cambridge

University Press. New York.

Nummenmaa, L., Hyönä, J., Calvo, M. G., 2006. Eye

movement assessment of selective attentional capture

by emotional pictures, Emotion, 6(2), 257-268.

Nyström, M., Holmqvist, K., 2008. Semantic override of

low-level features in image viewing-both initially and

overall, Journal of Eye-Movement Research, 2(2), 2-1.

Pang, M. F., Lo, L. M., 2012. Learning study: helping

teachers to use theory, develop professionally, and

produce new knowledge to be shared. Instructional

Science, 1-18.

Ramus, F., Rosen, S., Dakin, S. C., Day, B. L., Castellote,

J. M., White, S., Frith, U., 2003. Theories of

developmental dyslexia: insights from a multiple case

study of dyslexic adults, Brain, 126(4), 841-865.

Runesson, U., 1999. Variationens pedagogik. Skilda sätt

att behandla ett matematiskt innehhåll. [The pedagogy

of variation: different ways of handling a

mathematical topic], Acta Universitatis

Gotheoburgensis. Göteborg, Sweden.

Shaywitz, S. E., Morris, R., & Shaywitz, B. A. (2008).

The education of dyslexic children from childhood to

young adulthood. Annu. Rev. Psychol., 59, 451-475.

Svensson, L., 1976. Study skill and learning, Universitatis

Gothoburgensis. Göteborg, Sweden.

Vellutino, F., Fletcher, J., Snowling, M., Scanlon, D,

2004. Specific reading disability (dyslexia): What

have we learned in the past four decades?, Journal of

child psychology and psychiatry, 45(1), 2-40.

Wennås Brante, E., 2010. Identifying critical aspects from

learners’ perspective. Paper presented at The 8

International Conference on Education and

Information Systems, Technologies and Applications:

EISTA 2010, Orlando, Florida, USA, 29 June – 2 July,

2010.

Wennås Brante, E. (2013). Att orientera sig i text och bild.

Skillnader mellan förmodad och faktisk läsning för

läsare med dyslexi. Acta Didactica Norge, 7(1), (Art.

12, 20 sider).

Wennås Brante, E. (manuscript). Non-congruent pictures

TheImageofImagesasanAidtoImproveLearning-AnEye-trackingExperimentStudyingtheEffectofContrastsin

Computer-basedLearningMaterial

315

effects’ on reading comprehension: Differences and

similarities in patterns of transition between text and

picture among young adults with and without dyslexia.

Wennås Brante, E., Holmqvist Olander, M., Nyström, M.,

2013. Exploring the impact of contrasting cases in text

and picture processing, Journal of Visual Literacy,

32(2):15–38.

Yarbus, A. L., 1967. Eye movements during perception of

complex objects, Eye movements and vision, 7, 171–

196.

Yin, R. K., 2006. Mixed methods research: Are the

methods genuinely integrated or merely parallel.

Research in the Schools, 13(1), 41–47.

CSEDU2014-6thInternationalConferenceonComputerSupportedEducation

316