This article was downloaded by: [KRISS - Korea Research Institute of], [Seung Nam Min]

On: 28 April 2014, At: 06:19

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Ergonomics
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/terg20

Legibility difference between e-books and paper books

by using an eye tracker
a b b c
Jung-Yong Kim , Seung-Nam Min , Murali Subramaniyam & Young-Jin Cho
a
Industrial and Management Engineering, Hanyang University, Seoul, Korea
b
Center for Medical Metrology, Division of Convergence Technology, Korea Research
Institute of Standards and Science, Daejeon, Korea
c
Department of Process Innovation, C&BIS Corporation, Ansan, Korea
Published online: 24 Apr 2014.

To cite this article: Jung-Yong Kim, Seung-Nam Min, Murali Subramaniyam & Young-Jin Cho (2014): Legibility difference
between e-books and paper books by using an eye tracker, Ergonomics, DOI: 10.1080/00140139.2014.909951

To link to this article: http://dx.doi.org/10.1080/00140139.2014.909951

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Ergonomics, 2014
http://dx.doi.org/10.1080/00140139.2014.909951

Legibility difference between e-books and paper books by using an eye tracker
Jung-Yong Kima, Seung-Nam Minb*, Murali Subramaniyamb and Young-Jin Choc
a
Industrial and Management Engineering, Hanyang University, Seoul, Korea; bCenter for Medical Metrology, Division of Convergence
Technology, Korea Research Institute of Standards and Science, Daejeon, Korea; cDepartment of Process Innovation,
C&BIS Corporation, Ansan, Korea
(Received 20 June 2013; accepted 25 March 2014)

The aim of the study was to evaluate the difference in legibility between e-books and paper books by using an eye tracker.
Downloaded by [KRISS - Korea Research Institute of], [Seung Nam Min] at 06:19 28 April 2014

Eight male and eight female subjects free of eye disease participated in the experiment. The experiment was conducted using
a 2 £ 3 within-subject design. The book type (e-book, paper book) and font size (8 pt, 10 pt, 12 pt) were independent
variables, and fixation duration time, saccade length, blink rate and subjective discomfort were dependent variables. In the
results, all dependent variables showed that reading paper books provided a better experience than reading e-books did.
These results indicate that the legibility of e-books needs further improvement, considering fixation duration time, saccade
movement, eye fatigue, device and so on.
Practitioner Summary: This study evaluated the legibility difference between e-books and paper books from the viewpoint of
readability, eye fatigue and subjective discomfort by using an eye tracker. The results showed that paper books provided a better
experience than e-books. This indicates that the readability of e-books needs further improvement in relation to paper books.
Keywords: legibility; e-book; paper book; eye tracker

1. Introduction
Since the late twentieth century, e-book technology has been rapidly developing on the basis of digital innovation, and the
e-books produced by this technology are being supplied through various multimedia devices (Golovchinsky 2008).
Legibility is a very important issue for readers for a good reading experience and present-day e-readers have mainly been
developed with superior legibility (Nordstrom 2010). However, many researchers are interested in the subject of whether e-
books could replace paper books. For e-books and paper books, comparison analyses have been conducted from various
viewpoints. Wilson, Landoni, and Gibb (2003) analysed the effects that the design mode of e-books has on three factors: the
level of satisfaction (quality, feasibility, etc.), usability (task success, time) and memory (recognition, memory), using over
100 subjects including university students, lecturers and researchers. They found that paper books have greater task success
(usability) and a higher level of satisfaction than e-books. Daniel and Woody (2013) examined students’ usage and
performance that resulted from employing a randomly assigned format (textbook, printed text pages, printed manuscript in
Microsoft Word, electronic PDF file or electronic textbook) through a Likert scale (subjective questionnaire). They claimed
that additional reading time was involved when using electronic texts without the benefit of additional learning. Nielsen
(2010) conducted a readability study on four reading conditions (printed book, PC, iPad and Kindle) and claimed that paper
books provided better and quicker readability than other conditions. And he also found that reading the printed book was
more relaxing than using other conditions. However, studies by both Daniel and Woody (2013) and Nielsen (2010) did not
investigate the relationship between legibility and comprehension using an eye tracker.
Just and Carpenter (1980) and Menz and Groner (1982) claimed that the typical variables for evaluating legibility were
fixation, saccade, pupil size and blink rate. Rayner and Pollatsek (1989) and Biedert, Buscher, and Dengel (2010) suggested
that fixation duration time and saccade length were indicators of legibility, demonstrating that as fixation duration time
increased and saccade length decreased, it got tougher to read. Leyland et al. (2013) and Hooper and Hannafin (1986)
claimed that if the range that eyes recognised increased, fixation duration time decreased, and saccade length increased,
resulting in faster reading and long sentences. Kang, Jung, and Lee (2010) argued that if one line was long, then saccade
length increased so that eyes could not move to another one easily. Iqbal, Zheng, and Bailey (2004) and Recarte et al. (2008)
claimed that dilatation of the pupil was a reliable measure for mental workload. However, Al-Omar, Al-Wabil, and Fawzi
(2013) and Näätänen (1992) argued that pupils were highly influenced by ambient lighting or emotional changes. Gao et al.
(2013) and de Waard and Studiecentrum (1996) also used blink rate as an indicator of mental workload, and Gao et al.
(2013) and Fogarty and Stern (1989) suggested that lowering blink rate meant increasing visual demand. In addition, Stern,

*Corresponding author. Email: msnijn12@hanmail.net

q 2014 Taylor & Francis

 2 J.-Y. Kim et al.

Boyer, and Schroeder (1994) argued that blink rate could be used as an indicator of fatigue. The eye tracker has been used as
a tool to quantitatively evaluate reading ability and efficiency. However, there have been only a limited number of recent
studies regarding e-books.
Siegenthaler, Wurtz, and Groner (2010) compared the legibility of five e-book reading devices and a paper book by
using an eye tracker. They claimed that there were no significant differences between reading devices in total reading time
and reading speed. They also concluded that in some circumstances, with the function of changing font size in the e-book
devices, readers can enjoy better legibility than in paper books. However, their study was not about comparing the legibility
with different font sizes.
Therefore, this study sought to evaluate the relative legibility of e-books and paper books, under conditions where the
circumstances of paper books and devices were similarly controlled.
Downloaded by [KRISS - Korea Research Institute of], [Seung Nam Min] at 06:19 28 April 2014

2. Methods
2.1 Participants
We conducted a trial in this study using eight males and eight females in their 20s –30s who had had no eye disease for six
months and had good health status. Corrected visual acuity of the subjects was controlled by more than 1.0.

2.2 Apparatus
We used the cable-type Dikablis Eye Tracking System eye tracker (Ergoneers), an eye gaze display to measure gaze
movement (Figure 1a). To measure the coordinates of gaze movements, sampling occurred at a frequency of 25 Hz. We
used Dikablis Recorder, Dikablis Analysis and D-Lab Analysis software made by the same company for measurements,
extraction and analysis of data. We selected the M model, made by the company that recorded the greatest supply rate in
2011 (Figure 1b), as the e-book device used for the trial. In order to enable content to be downloaded to the e-book device,
we formatted files into PDF format, controlling for the number of letters on each page, and made paper books containing the
same font size and gaps for the trial (Figure 1c).

2.3 Experimental design

In order to study differences between e-books and paper books in terms of legibility and subjective discomfort according to
font size, we used a 2 £ 3 within-subject design. We selected book type (e-book, paper book) and font size (8 pt, 10 pt, 12 pt) as
independent variables, and fixation duration time, saccade length, blink rate and subjective discomfort as dependent variables.
We prepared text materials from one book written in Korean, and made six (A– F) of them for each variable to constitute the e-
book and the paper book. For the trial orders, we used a Latin square design to remove the learning effects of subjects.

2.4 Procedure
The whole process of the trial is shown in Figure 2. We explained the purpose and procedures of this trial to the subjects
before beginning, and we directed subjects to write down their personal information and visual acuity. Before the trial, we
instructed them on how they would use the e-books, paper books and eye trackers for adjustment, and conducted adjustment
practice to remove learning effects.
Once subjects became accustomed to using the devices, we got them to read e-books and paper books in accordance
with trial design orders (Figure 3). Under each condition, subjects were directed to read three selected pages. We guided
them towards reading carefully and as if they were reading in daily life, by informing them in advance that we would ask
them simple questions regarding the content to verify whether they understood it. When a trial of one condition was
finished, we asked about their subjective discomfort.

Figure 1. Experimental environments: (a) eye tracker, (b) e-book, (c) paper book.
 Ergonomics 3

Figure 2. Flow of experiment.

Downloaded by [KRISS - Korea Research Institute of], [Seung Nam Min] at 06:19 28 April 2014

Figure 3. Participant reading an e-book equipped with eye tracker.

All subjects read books at a distance of 0.5 m, as the distance between books and eyes could affect variables. Intensity of
illumination was set at 300 lx, a standard value when reading.

2.5 Data analysis

We used coordinates of gaze movements measured by the eye tracker to evaluate the legibility of e-books and paper books,
and based on these, we calculated fixation duration time, saccade length and blink rate. We employed a calculation formula
to correct for the difference in coordinates of gaze movement measured by the eye tracker of real books (Figure 4). The
coordinates of gaze measured by the eye tracker by using formula (1) were converted to position values.

110ðX k 2 X min Þ
Rk ¼ ; ð1Þ
X max 2 X min

where
Rk: the x-axis coordinates of paper book and e-book (mm);
Xmax: the maximum x-axis value among measured coordinates of gaze;
Xmin: the minimum x-axis value among measured coordinates of gaze;
Xk: the x-axis coordinates needed to be obtained among measured coordinates of gaze.
Fixation duration time of gaze and saccade length were calculated by using the definition of fixation and saccade
suggested in typical literature. Fixation of gaze was defined as when gaze stays steady for at least 100 ms (Ito et al. 2011;
Kim et al. 2006) and when movement smaller than a letter occurs in reading (Lee 2004), and saccade length was defined as
the distance between saccades (Wickens et al. 2012).
The calculation formula for fixation duration time is as follows:

ST i
T fixation ¼ ; ð2Þ
N fixation
 4 J.-Y. Kim et al.
Downloaded by [KRISS - Korea Research Institute of], [Seung Nam Min] at 06:19 28 April 2014

Figure 4. Transformation of coordinates between the distance of real book and measured eye movement.

where
T fixation : average fixation time;
Ti: ith fixation time;
Nfixation: number of fixation occurrences.
Saccade length was calculated as follows:

SLi
L saccade ¼ ; ð3Þ
N saccade

where
L saccade : average saccade length;
Li: each saccade length;
Nsaccade: number of saccade occurrences.
Blink rate was defined as the proportion of total reading time that consists of blinking.
Blink rate ¼ time eyes are closed/(Total reading time – time turning over pages).
To measure subjective discomfort, we asked ‘How uncomfortable was it to read?’ whenever each trial condition was
over, and collected data ranging from 0 (most comfortable) to 10 (most uncomfortable).
Finally, we conducted analysis of variance (ANOVA) under a null hypothesis that fixation duration time, saccade length,
blink rate and subjective discomfort are all the same regardless of book type (e-book, paper book) and font size (8 pt, 10 pt, 12 pt).

3. Results
The summary of ANOVA results is shown in Table 1. The main effects of fixation duration time, saccade length, blink rate
and subjective discomfort were found to be statistically significant, whereas interaction effects were not statistically
significant.
The differences in legibility and discomfort between paper books and e-books are shown in Figure 5. The results show
that reading e-books is more difficult and uncomfortable than reading paper books, judging by fixation duration time,
saccade length, blink rate and subjective discomfort.
Font size also affects legibility and discomfort (Figure 6). At font size 8, fixation duration time and saccade length are
shortest, whereas subjective discomfort is highest.

4. Discussion
We conducted trials in order to evaluate legibility between e-books and paper books. We have summarised the detailed
results in Table 2.
 Ergonomics 5

Table 1. Summary of analysis of variance (ANOVA).

Dependent variable Book type Font size Book type £ font size
Fixation duration time 14.142*** 6.807*** 0.880
Saccade length 7.777** 14.696*** 0.431
Blink rate 3.435* 0.438 0.435
Subjective discomfort 4.061* 9.483*** 0.753
*p , 0.1; **p , 0.05; ***p , 0.01.
Downloaded by [KRISS - Korea Research Institute of], [Seung Nam Min] at 06:19 28 April 2014

Figure 5. Legibility and subjective discomfort according to book type: (a) fixation duration time, (b) saccade length, (c) blink rate,
(d) subjective discomfort.

Figure 6. Legibility and subjective discomfort according to font size: (a) fixation duration time, (b) saccade length, (c) blink rate,
(d) subjective discomfort.
 6 J.-Y. Kim et al.

First, fixation duration time and saccade length were analysed to determine legibility. As Table 2 shows, the fixation
duration time for e-books was longer than that for paper books, and saccade length was shorter. Based on a study claiming
that as books get harder to read, fixation duration time increases and saccade length decreases (Biedert, Buscher, and Dengel
2010; Rayner and Pollatsek 1989; Leyland et al. 2013; Hooper and Hannafin 1986), these results suggest that e-books have
weaker legibility than paper books. Shorter fixation duration time implies faster recognition, which indicates that readers
can recognise text in paper books faster than in e-books, and longer saccade length suggests that the distance from
recognising a larger range and reading more letters to the next fixation is long, which can be considered to mean that readers
can more easily recognise paper books than e-books.
Second, blink rate was analysed in order to assess eye fatigue. The blink rate in reading e-books was higher than in
reading paper books. This indicates that while reading e-books one is more easily fatigued than while reading paper books,
as expected. That is, as blink rate increases, fatigue increases (Fogarty and Stern 1989; Stern, Boyer, and Schroeder 1994;
Downloaded by [KRISS - Korea Research Institute of], [Seung Nam Min] at 06:19 28 April 2014

de Waard and Studiecentrum 1996; Gao et al. 2013). Also, considering the opinions collected through the survey, subjects
felt less fatigue from reading paper books with reflected natural light than from reading e-books with artificial light
(Näätänen 1992; Al-Omar, Al-Wabil, and Fawzi 2013). Therefore, higher blink rate indicates higher fatigue, which means
readers feel less fatigue when reading paper books than when reading e-books.
Third, subjective discomfort was analysed from the subjective aspect of users. Reading e-books produced a higher level
of discomfort than reading paper books did (Nielsen 2010; Daniel and Woody 2013). Siegenthaler, Wurtz, and Groner
(2010) reported a notable lack of usability in the current e-book reading devices and highlighted that the lack of usability
influenced the subjective legibility rating. This means that users feel more uncomfortable when reading e-books than when
reading paper books. While they felt more uncomfortable when reading letters of 8 pt than when reading 10 pt or 12 pt, there
was no significant difference between reading 10 pt and reading 12 pt, confirming that the subjective level of discomfort is
the same for both fonts. Considering the aspect of font size, the shorter fixation duration time and saccade length when
reading 8 pt rather than 10 pt or 12 pt seems to indicate that, under circumstances where limited words can be recognised
due to small font size (Siegenthaler, Wurtz, and Groner 2010), users exhibit brief, repeated looks (short fixation duration
time) and short movements (short saccade length) (Figure 7).
For book type and font size, legibility, fatigue and subjective aspects, there are significant results; however, there
appeared to be no interaction between book type and font size. Therefore, book type and font size are judged to have
independent characteristics that do not affect dependent variables, which indicates that one can consider book type and font
size separately when designing books.
Typical literature highlights desirable e-book functions (such as table of contents) that have hyperlink capability, guide
words, simple and detailed search tools, clearly represented outsource links and selectable font size and type (Wilson,
Landoni, and Gibb 2003). However, this study analysed pure legibility, fatigue and subjective evaluation based on the same
font sizes as those of paper books, not based on functions that only e-books have.
Future studies should conduct trials on legibility by selecting various letter types and font sizes for both e-books and
paper books; additionally, further studies related to kerning (spacing between letters) or the width of letters are also needed,
and studies on levels of reader satisfaction and fatigue according to the lighting of e-books could provide information on

Table 2. Summary of legibility.

Book type Font size
Fixation duration time P,E 8 pt , 10 pt, 12pt
Saccade length P.E 8 pt , 10 pt, 12 pt
Blink rate P,E Not significant
Subjective discomfort P,E 8 pt . 10 pt, 12pt
Note: P: paper book, E: e-book.

Figure 7. Difference in fixation duration time and saccade length between font sizes when reading Korean text.
 Ergonomics 7

proper lighting. Regarding fixational eye movement during reading, a deeper level of investigation is needed that would
divide categories into micro-saccade, drift, tremor, etc., since the effect that e-books and paper books have on the eyes and
the differing properties of devices could affect results.

5. Conclusion
This study aimed to conduct ergonomics evaluation for designing user-centred, high-legibility e-books through quantitative
comparison between the new reading medium, the e-book, and the traditional paper book. It was revealed that e-books have
lower legibility than paper books from the aspects of fixation duration time, saccade length and blink rate, and they produce
higher subjective discomfort. It could be inferred that, functionally, e-books cannot yet totally replace paper books. More
tests and improvements are required prior to replacing traditional paper books with this new reading medium. For more
active distribution of e-books, further studies on lighting conditions and UX (user experience) should be conducted to
Downloaded by [KRISS - Korea Research Institute of], [Seung Nam Min] at 06:19 28 April 2014

improve the legibility of e-books.

References
Al-Omar, D., A. Al-Wabil, and M. Fawzi. 2013. “Using Pupil Size Variation during Visual Emotional Stimulation in Measuring Affective
States of Non Communicative Individuals.” In Universal Access in Human-Computer Interaction. User and Context Diversity, edited
by C. Stephanidis and M. Antona, 253– 258. Berlin: Springer. doi:10.1007/978-3-642-39191-0_28.
Biedert, R., G. Buscher, and A. Dengel. 2010. “The eyeBook – Using Eye Tracking to Enhance the Reading Experience.” Infrmatik-
Spektrum 33 (3): 272–281. doi:10.1007/s00287-009-0381-2.
Daniel, D. B., and W. D. Woody. 2013. “E-Textbooks at What Cost? Performance and Use of Electronic v. Print Texts.” Computers &
Education 62: 18 – 23. doi:10.1016/j.compedu.2012.10.016.
De Waard, D., and V. Studiecentrum. 1996. “The Measurement of Drivers’ Mental Workload.” PhD diss., Groningen University, Traffic
Research Center, Haren.
Fogarty, C., and J. A. Stern. 1989. “Eye Movements and Blinks: Their Relationship to Higher Cognitive Processes.” International Journal
of Psychophysiology 8 (1): 35 – 42. doi:10.1016/0167-8760(89)90017-2.
Gao, Q., Y. Wang, F. Song, Z. Li, and X. Dong. 2013. “Mental Workload Measurement for Emergency Operating Procedures in Digital
Nuclear Power Plants.” Ergonomics 56 (7): 1070– 1085. doi:10.1080/00140139.2013.790483.
Golovchinsky, G. 2008. “Reading in the Office.” Paper presented at the Proceeding of the 2008 ACM Workshop on Research Advances in
Large Digital Book Repositories, 21 – 24. doi:10.1145/1458412.1458420.
Hooper, S., and M. J. Hannafin. 1986. “Variables Affecting the Legibility of Computer Generated Text.” Journal of Instructional
Development 9 (4): 22 – 28. doi:10.1007/BF02908315.
Iqbal, S. T., X. S. Zheng, and B. P. Bailey. 2004. “Task-Evoked Pupillary Response to Mental Workload in Human-Computer
Interaction.” Paper presented at the CHI’04 Extended Abstracts on Human Factors in Computing Systems, 1477– 1480. doi:10.1145/
985921.986094.
Ito, J., P. Maldonado, W. Singer, and S. Grün. 2011. “Saccade-Related Modulations of Neuronal Excitability Support Synchrony of
Visually Elicited Spikes.” Cerebral Cortex 21 (11): 2482– 2497. doi:10.1093/cercor/bhr020.
Just, M. A., and P. A. Carpenter. 1980. “A Theory of Reading: From Eye Fixations to Comprehension.” Psychological Review 87 (4):
329– 354. doi:10.1037/0033-295X.87.4.329.
Kang, B., K. Jung, and J. Lee. 2010. “Visual Patterns and Performance Evaluation in the Scrolling Design of a Mobile Information
Device.” Journal of Korean Society of Design Science 23 (3): 261– 271.
Kim, G., J. Kim, H. Park, and J. Lee. 2006. “A Study of the Attention to the Internet Ads through Visual Perception Process.” KOBAKO
(Study of Advertisement) 72: 31 –58.
Lee, C. 2004. Gaze Movement to Reading Korean Letter. Seoul: Seoul National University Press.
Leyland, L. A., J. A. Kirkby, B. J. Juhasz, A. Pollatsek, and S. P. Liversedge. 2013. “The Influence of Word Shading and Word Length on Eye
Movements during Reading.” The Quarterly Journal of Experimental Psychology 66 (3): 471–486. doi:10.1080/17470218.2011.599401.
Menz, C., and R. Groner. 1982. “The Analysis of Some Componential Skills of Reading Acquisition.” In Cognition and Eye Movements,
edited by R. Groner and P. Fraisse, 169–178. Amsterdam: Elsevier North Holland.
Näätänen, R. 1992. Attention and Brain Function. Hillsdale, NJ: Psychology Press.
Nielsen, J. 2010. “iPad and Kindle Reading Speeds.” http://www.nngroup.com/articles/ipad-and-kindle-reading-speeds/
Nordstrom, R. 2010. “Legibility Investigations of the BiNem Display for e-Reader Applications.” Masters diss., Uppsala University,
Department of Information Technology, Sweden.
Rayner, K., and A. Pollatsek. 1989. The Psychology of Reading. Abingdon: Routledge.
Recarte, M. A., E. Pérez, A. Conchillo, and L. M. Nunes. 2008. “Mental Workload and Visual Impairment: Differences between Pupil,
Blink, and Subjective Rating.” The Spanish Journal of Psychology 11 (2): 374–385.
Siegenthaler, E., P. Wurtz, and R. Groner. 2010. “Improving the Usability of e-Book Readers.” Journal of Usability Studies 6 (1): 25– 38.
Stern, J. A., D. Boyer, and D. Schroeder. 1994. “Blink Rate: A Possible Measure of Fatigue.” Human Factors: The Journal of the Human
Factors and Ergonomics Society 36 (2): 285–297.
Wickens, C. D., J. G. Holland, R. Parasuram, and S. Banbury. 2012. Engineering Psychology and Human Performance. Upper Saddle
River, NJ: Prentice Hall.
Wilson, R., M. Landoni, and F. Gibb. 2003. “The WEB Book Experiments in Electronic Textbook Design.” Journal of Documentation
59 (4): 454– 477. doi:10.1108/00220410310485721.