ArticlePDF Available

Abstract and Figures

CAPTCHA is one of the most common solutions to check if the user trying to enter a Website is a real person or an automated piece of software. This challenge-response test, implemented in many Internet Websites, emphasizes the gaps between accessibility and security on the Internet, as it poses an obstacle for the learning-impaired in the reading and comprehension of what is presented in the test. Various types of CAPTCHA tests have been developed in order to address accessibility and security issues. The objective of this study is to investigate how the differences between various CAPTCHA tests affect user experience among populations with and without learning disabilities. A questionnaire accompanied by experiencing five different tests was administered to 212 users, 60 of them with learning disabilities. Response rates for each test and levels of success were collected automatically. Findings suggest that users with learning disabilities have more difficulties in solving the tests, especially those with distorted texts, have more negative attitudes towards the CAPTCHA tests, but the response time has no statistical difference from users without learning disabilities. These insights can help to develop and implement solutions suitable for many users and especially for population with learning disabilities.
Content may be subject to copyright.
Interdisciplinary Journal of e-Skills and Lifelong Learning Volume 12, 2016
Cite as: Gafni, R., & Nagar, I. (2016). CAPTCHA: Impact on user experience of users with learning disabilities. Inter-
disciplinary Journal of e-Skills and Life Long Learning, 12, 207-223. Retrieved from
http://www.informingscience.org/Publications/3612
Editor: Janice Whatley
An earlier, shorter version of this paper was presented at the Chais conference 2016, in Raanana, Israel, and included in
Y. Eshet-Alkalai, I. Blau, A. Caspi, N. Geri, Y. Kalman, & V. Silber-Varod (Eds.), Proceedings of the 11th Chais Con-
ference for the Study of Innovation and Learning Technologies 2016: Learning in the Technological Era. Raanana: The
Open University of Israel.
CAPTCHA: Impact on User Experience of Users with
Learning Disabilities
Ruti Gafni and Idan Nagar
The Academic College of Tel-Aviv–Yaffo, Yaffo, Israel
rutigafn@mta.ac.il; Idan.nagar@gmail.com
Abstract
CAPTCHA is one of the most common solutions to check if the user trying to enter a Website is a
real person or an automated piece of software. This challenge-response test, implemented in
many Internet Websites, emphasizes the gaps between accessibility and security on the Internet,
as it poses an obstacle for the learning-impaired in the reading and comprehension of what is pre-
sented in the test. Various types of CAPTCHA tests have been developed in order to address ac-
cessibility and security issues. The objective of this study is to investigate how the differences
between various CAPTCHA tests affect user experience among populations with and without
learning disabilities. A questionnaire accompanied by experiencing five different tests was ad-
ministered to 212 users, 60 of them with learning disabilities. Response rates for each test and
levels of success were collected automatically. Findings suggest that users with learning disabili-
ties have more difficulties in solving the tests, especially those with distorted texts, have more
negative attitudes towards the CAPTCHA tests, but the response time has no statistical difference
from users without learning disabilities. These insights can help to develop and implement solu-
tions suitable for many users and especially for population with learning disabilities.
Keywords: CAPTCHA, cyber security, user experience, learning disabilities, dyslexia
Introduction
Malicious programs try to access Websites for many reasons. One of the main issues of cyber
security deals with the question whether the agent trying to reach a Website is a real person or a
malicious automated program (“bot”). One of the most common solutions to decide whether the
agent trying to access the Website is legal is called CAPTCHA (Completely Automated Public
Turing test to tell Computers and Humans Apart). The first CAPTCHA test was invented by Luis
von Ahn, Manuel Blum, Nicholas Hopper, and John Langford of Carnegie Mellon University in
2000 and is still used today (Singh & Pal, 2014). The “T” in the name stands for Turing Test to
Tell, because CAPTCHA tests are like Turing
Tests. In the original Turing Test, a human judge
was instructed to ask a series of questions to two
players, without seeing them, one of which was a
computer and the other a human. Both players pre-
tended to be the human, and the judge had to distin-
guish between them. CAPTCHA tests are similar to
the Turing Test in that they distinguish humans
from computers, but they differ in that the judge is
now a computer (Von Ahn, Blum, & Langford,
(CC BY-NC 4.0)
This article is licensed to you
under a Creative Commons Attribution-
NonCommercial 4.0 International License. When
you copy and redistribute this paper in full or in
part, you need to provide proper attribution to it
to ensure that others can later locate this work
(and to ensure that others do not accuse you of
plagiarism). You may (and we encourage you to)
adapt, remix, transform, and build upon the mate-
rial for any non-commercial purposes. This li-
cense does not permit you to use this material for
commercial purposes.
CAPTCHA: Impact on User Experience of Users with Learning Disabilities
208
2004). Using the CAPTCHA tests can prevent instances of spam in blogs, protect Website regis-
trations, protect Email addresses from scrapers, prevent attacks, etc., while ensuring that those
who enter the Website are only human people. The test performs a challenge-response authentica-
tion process, presenting a challenge to the user, and the right to access the Website is given if
solved. If the user fails to solve the test then he/she is considered to be a machine, otherwise the
user is considered as an authentic human being user, and the access is allowed. The user must use
cognitive abilities, which are at the present time impossible for machines (Saini & Bala, 2013).
Cyber threats became abundant, and the attempts to reach computers by unauthorized agents are
growing. Therefore, CAPTCHA implementations can be found on more than 3.5 million sites
globally, in all kinds of Websites, like sites for fill-in forms, write comments, buy tickets, etc.,
and human beings solve CAPTCHA tests more than 300 million times a day (Angre, Kapadia, &
Ugale, 2015). Moreover, companies and researchers are looking forward to replacing passwords
with CAPTCHA tests, expanding even more the usage of CAPTCHA (Hande, & Ali, 2015; Red-
dy, Krishna, & Reddy, 2015).
The conflict between approachability and protection in the internet world is a complex issue,
dealing with the necessity to allow a wide range of different users to access the desired Website,
but preventing the access of malicious elements. CAPTCHA tests are an efficient approach to this
goal, but are difficult for users with learning disabilities (LD), who have difficulties in reading,
understanding, and performing the tests. CAPTCHA tests must be, on the one hand, very easy for
the user in order to pass, and, on the other hand, very difficult for the bots to pass.
There are some definitions of Learning Disabilities, which were first defined by Hammill (1990).
Since then, the definition was developed and re-defined. According to Katchergin (2015), re-
searchers and professionals in the field of disabilities tend to base their definitions on those of the
NJCLD (National Joint Committee on Learning Disabilities, 1994). The NJCLD definition claims
that Learning Disabilitiesis a general term that refers to a heterogeneous group of disorders
manifested by significant difficulties in the acquisition and use of listening, speaking, reading,
writing, reasoning, or mathematical skills. These disorders are intrinsic to the individual, pre-
sumed to be due to central nervous system dysfunction. The number of people with learning disa-
bilities in the population is not known exactly, but the numbers used by the professionals varied
from 3% to 30%, with the most frequent estimate being 10% (Katchergin, 2015). Sometimes this
percentage can be higher in practice, and it is even estimated that one-third of internet users suffer
from certain symptoms of learning disabilities (McCarthy & Swierenga, 2010). According to the
study of Foley (2012), the most common learning disability is dyslexia. Dyslexia (from Greek,
dys=difficulty + lexis=words), also known as reading disorder, is a learning disability character-
ized by trouble reading despite one’s normal intelligence. A dyslexic person has difficulty in as-
sociating the graphic symbols and letters with their corresponding sounds and cannot organize
them mentally in a correct sequence (Madeira, Silva, Marcelino, & Ferreira, 2015). Different
people are affected to varying degrees. Problems may include sounding out words, spelling
words, reading quickly, writing words, pronouncing words when reading aloud, and understand-
ing what one reads. Often these difficulties are first noticed at school. The cause of dyslexia is
believed to involve both genetic and environmental factors. It occurs most often in people with
attention deficit hyperactivity disorder (ADHD) and is associated with problems with mathemat-
ics. The underlying mechanism involves problems with the brain’s processing of language. Dys-
lexia is diagnosed by a series of tests of a person’s memory, spelling, ability to see, and reading
skills. It is separate from reading difficulties due to poor teaching, or hearing or vision problems
(Katchergin, 2015).
People with learning disabilities may find it difficult and disturbing to succeed in performing the
CAPTCHA tests, because they consist in combining cognitive and physical effort. They must
Gafni & Nagar
209
read distorted texts, understand them, perform calculations, and even move small pictures with
accuracy.
Many studies have been conducted on CAPTCHA, focusing on the security analysis (Azad,
2013), on the differences between types of tests (Foley, 2012; Singh & Pal, 2014), and examining
the combination between security and user friendliness (Gossweiler, Kamvar, & Baluja, 2009).
Only few focused on the user experience (Fidas, Voyiatzis, & Avour, 2011; Tangmanee & Suja-
rit-apirak, 2013). This research deals with the influence of CAPTCHA tests among users having
learning disabilities, examining user experience, actual performance, and success.
Theoretical Background
User Disabilities, Accessibility, and User Experience
Bevans (2009) study describes the user experience as including emotions, beliefs, preferences,
conceptions, psychological and physical reactions, behaviors, and achievements occurring before,
during, and after usage. According to Onwudebelu, Sanjo, Obi, and Alaba (2010), the use of
CAPTCHA tests is a nuisance. Some users feel threatened by these tests, irritated, and frustrated,
as they are unable to understand the need for it. Others reported that the text displayed is unclear,
and they struggle to solve it. Most reported the tests slow them down and interfere with their ac-
tivity on Websites.
Web accessibility has become an important issue since the dramatic rise in the use of the Internet.
Web accessibility deals with overcoming barriers, which users with disabilities face when trying
to access information on Websites. In some countries (U.S.A., Israel, etc.) laws relating to Inter-
net accessibility were legislated, in order to improve the usability of the Internet for disabled us-
ers. However, a large number of Websites are still not adapted. For example, nearly a third of the
official government Web sites of the 50 states and the District of Columbia tested did not meet
the most fundamental requirements for Web accessibility (White, Goette, & Young, 2005). There
are many recommendations for development of appropriate and friendly Websites that need to be
accessible towards those users, and in this way improve the experience for internet users (McCar-
thy & Swierenga, 2010; Pascual, Ribera, & Granollers, 2015), including guidelines (W3C, 2008)
and clear principles (Matej, 2013). These recommendations help and encourage designers and
web developers to make Websites accessible to all users, including users with disabilities and im-
pairments (Foley, 2012). The Internet may greatly facilitate the provision of accessible infor-
mation to people with learning disabilities. However, problems of navigation and, therefore, re-
trieval represent a barrier for this cohort. A study conducted by Williams and Hennig (2015) ex-
amined how the Web-page design affects the access to content for people with learning disabili-
ties. For example, they examined whether the orientation of the page, vertical or horizontal, and
the positions of the words in the page facilitate faster access. They found that the propensity to
imbibe information serially(word-for-word) rather than to skim or look “globally” has im-
portant Website design implications. According to Sagirani, Nugroho, Santosa, and Kumara
(2015) there are some recommendations and basic concepts in creating a design that can foster
user experience. Planning the product should focus on the content, presentation, functionality, and
interaction. Specifically, good interaction between users with limitations and the application can
provide improvements for children with special educational needs, especially on their cognition,
emotion, motivation, attention, perception, and behavior. Aside from accessibility difficulties,
frustration, and poor user experience (Ghazarian, 2014), user abandonment and decline in the
website's conversion rates (the percentage of visitors who take a desired action) are additional
consequences of CAPTCHA tests suffering from user-unfriendliness (Mujumdar & Polisetti,
2011).
CAPTCHA: Impact on User Experience of Users with Learning Disabilities
210
The activity of CAPTCHA tests needs to be trivial enough to be performed successfully by hu-
man persons, but they often present some difficulties (Singh & Pal, 2014). Each of the different
types has some drawbacks.
Types of CAPTCHA Tests
During the last years, several types of CAPTCHA tests have been defined and developed. Each
type has its pros and cons. Here are the descriptions of the most common types of CAPTCHA
tests:
Text-based CAPTCHA test – is the most used kind of test (Figure 1), called “reCAPTCHA”,
consisting of a sequence of numbers and letters, twisted and shown in a distorted manner. This
mechanism was originally aimed to help digitize printed text that was hard to read for OCR (Op-
tical Character Recognition) and was acquired by Google in 2009, in order to digitize antique
manuscripts (“reCAPTCHA”, n.d.). The user needs to identify and decipher what is shown and
then type the exact sequence into a text box. If the user cannot decipher the text, it is possible to
retry with a different text. There is also an option to hear the letters, which was developed for vi-
sion-impaired people. The user hears a sequence of letters and/or numbers and must type the se-
quence in the text-box, but this is often performed with a noisy background, which does not help
too much.
Figure 1. Text-based CAPTCHA
In 2013, reCAPTCHA began implementing behavioral analysis of the browser’s interactions with
CAPTCHA. This analysis (Figure 2) occurs before displaying the CAPTCHA and presents a
more difficult test in cases there are reasons to think the user is a bot. From 2014 this mechanism,
called No CAPTCHA reCAPTCHA, started to be used in most of Google services (“reCAP-
TCHA”, n.d.).
Figure 2. No CAPTCHA reCAPTCHA
Arithmetic operation based CAPTCHA test contains a very basic arithmetic operation, for
example 1+3 =” (Figure 3), which can be performed by almost every human being. The user
needs to enter the result of the operation into a text box.
Figure 3. Arithmetic operation CAPTCHA
Gafni & Nagar
211
Picture based CAPTCHA test – in this kind of test, a number of pictures are shown to the user
with a simple question. In the example shown in Figure 4, there are eight different pictures, of
which four show piggy banks”. The user is asked to click on all of the piggy bank images. The
user must identify the pictures and select those which represent the correct answer, and there is no
need to write any text. There are some variations of the picture-based CAPTCHA, for example,
sliding distorted pictures to arrange them (Figure 5).
Figure 4. Picture based CAPTCHA
Figure 5. Sliding picture based CAPTCHA
(from http://www.geekandblogger.com)
Game based CAPTCHA test – includes puzzles or interactive games (Mohamed et al., 2013).
User experience and gamification are some of the “buzzwords” in the last years (Robson, Plang-
ger, Kietzmann, McCarthy, & Pitt, 2015). In order to encourage the users to perform tedious but
substantial tasks, the activity is wrapped with a joyful function. In Figure 6, an example of a game
can be seen, where the user needs to drag the pictures of items that are food to the baby’s mouth.
Figure 6. Game based CAPTCHA
CAPTCHA Tests and Users with Disabilities
In the text-based CAPTCHA test, the user experiences difficulties deciphering and identifying the
characters due to the blurring of the characters displayed and their distortion. According to Fo-
leys study (2012), the text-based CAPTCHA test has many accessibility problems. For example,
visually impaired or almost blind users find the distorted text difficult to decipher, and sometimes
even completely impossible to see. This means that they are not able to pass this test. For people
with dyslexia, the ability to read and understand text can be affected by the way in which text has
been written and produced, therefore, users with learning disabilities might also find it difficult to
CAPTCHA: Impact on User Experience of Users with Learning Disabilities
212
identify correctly the characters displayed in this test, which in turn will cause them to fail the test
as well. Hsu and Lee (2011) found that older users show greater difficulty in passing a text-based
CAPTCHA in comparison to younger users, and that even non-disabled users may encounter dif-
ficulties recognizing and understanding the distorted characters. Bursztein, Bethard, Fabry,
Mitchell, and Jurafski (2010) found that the response times of older users were longer, while they
made fewer mistakes. Furthermore, a botcan have the ability to recognize the character se-
quence using Optical Character Recognition (OCR) software. In order to prevent access from
bots, Azad (2013) suggested to raise the security of text -based CAPTCHA by adding “noise”,
increasing the level of distortion of the characters and aligning the characters more closely; yet,
this would make it more difficult for users to identify the characters, causing more mistakes.
Mostly, CAPTCHA tests require the deciphering of a sequence of deformed characters in Latin
letters. This means that the text-based CAPTCHA test, the most common type being used today,
requires that users be able to know and read Latin letters. For Thai internet users, English is a less
familiar language, so for them, tests in Thai could prove to be a more suitable option. Tangmanee
and Sujarit-apirak (2013) claimed that Thai users are well aware of the existence of CAPTCHA,
but prefer an application using Thai language, which they are more familiar with. The study of
Fidas et al. (2011), explains that Greek users have also experienced difficulties using CAPTCHA.
Users whose mother tongue is not written in Latin letters frequently find CAPTCHA more chal-
lenging. Some CAPTCHA tests were defined in other languages, for example, in Arabic and Per-
sian (Shirali-Shahreza, & Shirali-Shahreza, 2006).
In the audio-based type, the sound clips played are based on the English language, and therefore
the user must understand English; in addition, there is a problematic side to recognizing similar-
sounding letters. Moreover, the text is played together with “noise”, in order to pose a challenge
to “bots, and that makes this solution less than ideal as it poses a problem for users who have a
hearing impairment or suffer from hearing loss, which, according to the World Health Organiza-
tion, consists of more than 5% of the population (WHO, 2015). According to Onwudebelu et al.,
(2010), while audio-based CAPTCHA tests are more commonly used for the visually impaired,
they do not provide full accessibility and are even characterized by a lower degree of security.
Picture-based CAPTCHA tests require recognition and selection of images with a similar or an
exceptional meaning or out of a sequence of images and may cause confusion, as the images can
sometimes be interpreted as having different meanings (Ahn, Kim, & Kim, 2013). Some studies
propose variations of the tests, in order to simplify them, but these propositions are not widely
used yet (Ahn et al., 2013; Gossweiler et al., 2009). Picture-based CAPTCHA tests do not pose
many of the problems faced by users with learning disabilities; however, in these tests, visually
impaired users still come across challenges similar to those of the text-based tests and struggle to
pass the tests (Foley, 2012).
Preliminary results of the study of Madeira et al., (2015), dealing with mobile applications’ usa-
bility for dyslexic users, show that a gamified set of activities allow dyslexics to improve multi-
sensory perception, constituting an added value facilitator of adaptiveness and learning. Thus,
game-based CAPTCHA tests may be the best choice for these users.
CAPTCHA tests must be easy for the user to pass successfully and be sufficiently difficult to
prevent the “bots” from passing them. However, most studies focus primarily on how to make the
tests more difficult for bots, in response to the growing number of security threats.
The CAPTCHA test supposedly provides an efficient method to distinguish between real users
and “bots”. However, the extensive use of CAPTCHA actually impairs the experience of users
with disabilities, and using this method is not the ideal solution in the long-term (Onwudebelu et
al., 2010). It emphasizes the gap between accessibility and security on the Internet, as it poses an
obstacle and a significant challenge for the visually impaired or learning-impaired in the reading
Gafni & Nagar
213
and comprehension of what is presented in the test, seeing that it provides significant challenges
to users who have impaired vision or have learning disabilities.
Research Questions and Hypotheses
The objective of this study is to investigate the differences between various CAPTCHA tests and
examine how they affect user experience among populations with and without learning disabili-
ties. For this purpose three questions were examined in the study.
People with learning disabilities find it difficult to read regular text, so reading and deciphering
distorted letters (Foley, 2012; Hammill, 1990; Katchergin, 2015), such as in the Text-based or
Arithmetic based CAPTCHA test, will prolong their efforts in reading, and therefore the response
time may be longer. However, gamification of the test may be an easier task for users with learn-
ing disabilities (Madeira et al., 2015), taking less time to perform it. Thus, the first research ques-
tion and its hypothesis are:
RQ1: Are there any differences in the response time of users with or without learning
disabilities?
H1: The performance time of users with learning disabilities will be longer in text and
arithmetic based tests
The difficulties that users with learning disabilities need to cope with may cause frustration and
decrease motivation (Katchergin, 2015; McCarthy & Swierenga, 2010), therefore it may result in
failures. However, when using CAPTCHA tests that are not based on letters and numbers, their
performance may be higher (Madeira et al., 2015). Thus, the next research question has two com-
plementary hypotheses:
RQ2: Are there any differences in the success rate of users with or without learning disa-
bilities?
H2.1: The success rate of users without learning disabilities will be greater.
H2.2: Users with learning disabilities will succeed better in tests, which do not include
letters.
According to prior studies, users found it annoying to perform the CAPTCHA tests (Onwudebelu
et al., 2010). Those users with learning disabilities will find these tests more frustrating. Howev-
er, their attitude may be better with the Picture and Game based tests, because in these tests the
text parts are smaller and not distorted. Therefore, the third research question has two hypotheses:
RQ3: Are there any differences in the attitude and user experience of users with or with-
out learning disabilities?
H3.1: Attitude and user experience of users with learning disabilities will be more nega-
tive.
H3.2: Users with learning disabilities will report better experience in tests, which do not
include letters.
Methodology
The data for this study was collected using an experiment, which was embedded into a question-
naire.
The first part of the questionnaire was composed of 11 demographic and general information
questions. The next five parts, each dealt with one of the following CAPTCHA tests:
CAPTCHA: Impact on User Experience of Users with Learning Disabilities
214
1. Text-based CAPTCHA
2. Arithmetic operation-based CAPTCHA
3. Picture-based CAPTCHA, using the version with the slider option, developed by Minteye
Company (www.minteye.com)
4. Game-based CAPTCHA, developed by Are-you-a-human Company
(www.areyouahuman.com)
5. “No CAPTCHA”, developed by Google Company (www.google.com/recaptcha)
The participants were asked to actually perform and solve each test, and immediately after that to
answer 10 questions about their experience using each one of the tests. The same questions were
asked about each of the tests. The responses were based on a five-level Likert-type scale (1
Strongly disagree’, to 5 – ‘Strongly agree’). Altogether, each participant answered 61 questions
and performed 5 different tests.
While the participants tried to solve the tests, a specifically developed hidden script automatically
accumulated the data about the success or failure of the respondent in using each test and the time
it took to complete it.
The questionnaire and experiment, which were built as a single unit using a plug-in based on
WordPress, which is a free and open-source web content management system, were delivered
primarily via the Internet, both through social media like Facebook (Baltar & Brunet, 2012), and
through Websites and forums related to learning disabilities.
The combination of both research methods was based on the study conducted by Abrich, Ber-
benetz, and Thrope (2011), which defined the quality of user experience on whether the user was
correct or not when taking a test, as well as on the level of test difficulty the user reported.
The answers were collected during one week (December 2014), and then gathered and analyzed
using IBM® SPSS® Statistics.
Results
In the first part of the questionnaire, the respondents had to give demographic data and infor-
mation about their previous familiarity with the different CAPTCHA tests.
There were 212 respondents, 60 reporting having learning disabilities or thinking they have but
not diagnosed yet (28%) and 152 without learning disabilities (72%).
The frequency of Internet usage was similar and high in the two groups (4.7 in LD and 4.88 in
non-LD).
Table 1 supplies demographics descriptive statistics of the participants in the questionnaire and
experiment.
Gafni & Nagar
215
Table 1: Demographics of the survey participants
N
Gender 99 men
(47%)
(53%)
Age 18 and under - 6 participants
19-30 – 141 participants
31-45 - 46 participants
46-59 - 14 participants
(3%)
(66%)
(22%)
(7%)
(2%)
Education 21 - high school
118 - Undergraduate students
46 - Bachelor degree
(10%)
(56%)
(22%)
(13%)
Table 2 summarizes the familiarity of the users (with and without learning difficulties) with each
CAPTCHA test. The mean value and standard deviation are shown. As seen, the text-based
CAPTCHA is the most familiar to the users, and the arithmetic operation based one is also
known. Most of the users are not so familiar with the other three types.
Table 2. Familiarity with the CAPTCHA tests (1 – ‘Never, 5 – ‘Very often’)
CAPTCHA type
LD users
(n=60)
Non-LD users
(n=152)
Total
(n=212)
1.
Text based
3.88
(1.01)
3.92
(0.94)
3.91
(0.96)
2.
Arithmetic operation based
2.48
(1.25)
2.19
(1.26)
2.28
(1.26)
3.
Picture based
1.03
(0.18)
1.2
(0.63)
1.15
(0.55)
4.
Game based
1.13
(0.5)
1.1
(0.43)
1.11
(0.45)
5.
NO CAPTCHAbased
1.33
(0.88)
1.69
(1.13)
1.59
(1.08)
The general position of the respondents towards CAPTCHA, according to prior experiences and
before the experiment was conducted, was collected by a set of statements, for which the partici-
pant had to rate his agreement on a Likert based scale. Table 3 summarizes the means for each
statement, for the users with learning disabilities, for those without learning disabilities, and for
the whole sample. As can be seen, the users understand the purpose of the tests (4.19), but this
understanding does not cause them to feel protected (2.83). Moreover, most of the users do not
like to use the CAPTCHA tests.
It is important to emphasize, that according to Table 2, most of the users are familiar only with
the text-based and arithmetic-based tests. Moreover, there is a prominent difference between the
two kinds of users for two types of CAPTCHA: users with learning disabilities are more familiar
with arithmetic-based test, while they are less familiar with the latest type, the NO CAPTCHA
test.
CAPTCHA: Impact on User Experience of Users with Learning Disabilities
216
Table 3: General position towards CAPTCHA tests
Statement
LD users
Non-LD users
Total
“I understand the meaning and the purpose of CAP-
TCHA tests 3.97 4.28 4.19
I feel frustrated / I hate it
3.32
2.94
3.05
“I feel that I spend too much time on it”
3.98
3.41
3.57
I feel protected / safe
2.95
2.79
2.83
I prefer something more comfortable instead of this
test / I prefer it would not exist 3.83 3.64 3.70
Thereafter, the participants were asked to solve the first test and immediately answer some ques-
tions about their experience. This was done with the other four tests as well.
After performing each of the CAPTCHA tests, the participants were asked to rate their experience
with the test, in terms of the extent to which they agree with the following statements:
1. The experience after the CAPTCHA test:
How would you describe your feeling after taking this CAPTCHA test?”
Item 1.1: Frustrating the test was difficult and unclear
Item 1.2: A waste of time It took too much time to do it
Item 1.3: Comfortable/Enjoyable It was a nice test
2. The position towards this CAPTCHA test:
“If you have to take this CAPTCHA test again, what will your position be towards this test?”
Item 2.1: “I would rather do it again – since it’s easy and clear
Item 2.2: “I would rather do it again – since it takes only a short time to do it
Item 2.3: “I would rather do it again – since it’s nice and comfortable
3. The position towards the Website:
“If you have to take this CAPTCHA test again, what will your position be towards the Website
that displays this test?
Item 3.1: “My position will be positive
Item 3.2: I will not cooperate with this site I will leave this site immediately
Item 3.3: I will not cooperate with this site I will not perform actions such as regis-
tration, buying, etc.
The outputs of the respondents’ experience are summarized in Table 4, which gives the items
means and standard deviation, the constructs built according to the statements, and their reliabil-
ity, measured by Cronbach’s alpha. Principal component factor analysis with Varimax rotation
was used to examine construct validity. Three items were intended to comprise one factor, ‘Un-
willingness to use the Website’, but one item (3.1) was statistically excluded. The 1.3 statement
was intended to be included in the ‘Frustratingconstruct (negative), but it was found that it load-
ed a different factor. Finally, four constructs were defined: ‘Frustrating’, ‘Enjoyable’, ‘Readiness
for future use’, and ‘Unwillingness to use the Website’.
Gafni & Nagar
217
Table 4. User experience constructs definition
Item Mean (SD) Cronbachs alpha Construct Mean (SD)
1.1
3.94 (1.28)
.802 Frustrating 3.86 (1.23)
1.2 3.78 (1.40)
1.3
2.94 (1.49)
Separated
Enjoyable
2.94 (1.49)
2.1
3.07 (1.48)
.954 Readiness for future use 3.05 (1.43)
2.2
3.09 (1.48)
2.3
3.00 (1.52)
3.1
2.89 (1.34)
Deleted
3.2
4.29 (1.07)
.904 Unwillingness to use the
Website 4.28 (1.03)
3.3 4.26 (1.09)
The results (Table 5) indicate that there is a significant difference between users with learning
disabilities and users without learning disabilities among the participants in two user experience
constructs: ‘Frustrating’ and Unwillingness to use the Website’.
Table 5. Statistical results of User experience
User Experience construct
LD User
N
Mean
SD
T
df
Sig.
(2-tailed)
Frustrating
No LD
760
3.96
1.19
3.961 508.79
.000 LD 300 3.62 1.29
Enjoyable
No LD
760
2.95
1.49
.412 1058 .680 LD 300 2.91 1.50
Readiness for future use
No LD
760
3.07
1.43
.627 1058 .531 LD 300 3.01 1.43
Unwillingness to use the
Website
No LD
760
4.34
0.96
3.123 463.18
.002 LD 300 4.10 1.18
Table 6 summarizes the findings of the user experience of each of the CAPTCHA tests, according
to the constructs defined. The gamed-based CAPTCHA was found to be the most enjoyable for
users with learning disabilities, while the text-based was the least enjoyable and most frustrating.
Statistical differences between the group of users with learning disabilities and the group of users
without were found in the ‘Frustratingconstruct in the text-based (t = –2.36, df = 210, sig =
0.019) and arithmetic-operation-based (t = –3.294, df = 210, sig = 0.001) tests.
CAPTCHA: Impact on User Experience of Users with Learning Disabilities
218
Table 6. Comparison of means and sd - user experience for CAPTCHA types
CAPTCHA type
LD User
Post-test experience
Frustrating Enjoyable
Readiness for
future use
Unwillingness to
use the Website
1. Text based
No LD
3.56 (1.10)
2.05 (1.17)
2.42 (1.12)
4.30 (0.87)
LD
3.15 (1.17)
1.85 (1.15)
2.09 (1.04)
4.98 (1.03)
2. Arithmetic operation
based
No LD
4.52 (1.03)
3.01 (1.47)
3.30 (1.42)
4.45 (0.85)
LD
4.69 (1.20)
2.72 (1.28)
3.07 (1.27)
4.11 (1.26)
3. Picture based
No LD
3.94 (1.15)
3.34 (1.42)
3.31(1.44)
4.37 (0.95)
LD
3.68 (1.36)
3.22 (1.52)
3.26 (1.51)
4.17 (1.19)
4. Game based
No LD
3.86 (1.21)
3.24 (1.38)
3.07 (1.37)
4.30 (0.97)
LD
3.68 (1.23)
3.55 (1.47)
3.19 (1.39)
4.29 (1.13)
5. “NO CAPTCHA
based
No LD
4.20 (1.31)
3.11 (1.63)
3.24 (1.56)
4.28 (1.13)
LD
3.88 (1.42)
3.20 (1.49)
3.44 (1.53)
3.98 (1.27)
As mention in the methodology section, each time the participant performed a test, two inputs
were collected automatically: the response time to complete the test, and the output, to check if
the test was performed successfully or if the answer was wrong.
As for response time, surprisingly, there were no significant differences between users with learn-
ing disabilities and those participants without learning disabilities for any of the CAPTCHA tests
(Table 7).
Table 7. Comparison of the CAPTCHA types Response Time
CAPTCHA type
Response Time (in seconds)
Mean (SD)
Median
LD
users
(n=60)
Non-LD
users
(n=152)
LD
users
(n=60)
Non-
LD
users
(n=15
2)
t-test
Sig
(2-tailed)
1. Text based
22.25
(12.73)
19.36
(13.69)
18 16 -1.41 0.16
2. Arithmetic operation based
11.4
(10.85)
9.16
(8.3)
9 8 -1.61 0.11
3. Picture based
26.78
(29.53)
24.34
(23.55)
18 18.5 -0.63 0.53
4. Game based
21.07
(13.05)
20.34
(16.33)
17 17 -0.31 0.76
5. “NO CAPTCHAbased
19.32
(22.74)
18.97
(23.6)
11.5 13 -0.10 0.92
Gafni & Nagar
219
The success rate was calculated for each test. Table 8 presents the results and the statistical com-
parison, using the Pearson Chi-square test (p < 0.05), indicating that there is a significant depend-
ence between the test success and the existence of LD only for text -based CAPTCHA, bringing
the LD users to fail more often. The arithmetic test is more difficult for all populations.
Table 8. Comparison of the CAPTCHA types Success Rates
CAPTCHA type
Success
Rate LD
users
(n=60)
Non-LD
users
(n=152)
χ²
value
df
Sig.
(2-
tailed)
1. Text based
%
58.33%
74.34%
Pass
35
113
5.23 1 0.022
Fail
25
39
2. Arithmetic operation based
%
48.33%
50%
Pass
29
76
0.048 1 0.83
Fail
31
76
3. Picture based
%
68.33%
73.03%
Pass
41
111
0.47 1 0.49
Fail
19
41
4. Game based
%
83.33%
82.89%
Pass
50
126
0.01 1 0.96
Fail
10
26
5. “NO CAPTCHAbased
%
86.67%
89.47%
Pass
52
136
0.34 1 0.56
Fail
8
16
Discussion
Findings suggest that users with learning disabilities have more difficulties in solving the tests,
especially those with distorted texts, and have more negative attitudes towards the CAPTCHA
tests than other users. Surprisingly, there was no significant difference found in response times,
between users with learning disabilities and those without in any of the five test types, thus, re-
jecting H1, which claimed that “The performance time of users with learning disabilities will be
longer in text and arithmetic based tests”. The reason H1 was not supported may be the fact that
CAPTCHA tests have become so common that the users are getting used to the twisted letters.
Another reason for this finding may be due to a certain level of impulsiveness in users with learn-
ing disabilities. According to estimates done on children with dyslexia, 30% have at least a mild
form of ADHD, which is characterized by hyperactivity, inattentive and impulsive behavior (Lee,
2015).
However, a significant difference was found in the test success rates between the two kinds of
users for only the text-based CAPTCHA. In all other types of CAPTCHA there were no differ-
ences between the groups. Thus, H2.1, claiming that ‘The success rate of users without learning
CAPTCHA: Impact on User Experience of Users with Learning Disabilities
220
disabilities will be greater’ was partially rejected, supporting the position of Hsu & Lee (2011),
who claim that even the non-impaired population might encounter difficulties with CAPTCHA
tests. Their study was one of the first to examine CAPTCHA tests from the usersperspective.
These findings, in fact, confirm the widespread notion that CAPTCHA tests are difficult for peo-
ple.
In addition, H2.2 which claims that ‘Users with learning disabilities will succeed better in tests,
which do not include letterswas accepted, meaning that users with learning disabilities have
more difficulties in reading the distorted texts, supporting Foley’s study (2012), but they have no
problems performing other tasks. Perhaps, if those users had spent more time to resolve the test,
the success rate would have been better, in spite of their learning disabilities.
A significant difference in the user-experience attitude was found for ‘Frustrating’ and ‘Unwill-
ingness to use the Website’ between users with learning disabilities and the those without, where
users with learning disabilities had a more negative attitude towards the tests, accepting H3.1,
which claims that Attitude and user experience of users with learning disabilities will be more
negative’. The most negative attitude of the users with learning disabilities was found in the text-
based and arithmetic-based CAPTCHA, supporting H3.2, which claims that ‘Users with learning
disabilities will report better experience in tests, which do not include letters’.
Conclusions
The main conclusion of this research is that all five types of tests influenced user experience to a
certain degree, from frustration to enjoyment. Users with learning disabilities found it more diffi-
cult to succeed in the text-based CAPTCHA, and there was a significant difference found be-
tween participants with learning disabilities and those without learning disabilities in most users
experiences constructs examined.
According to the International Dyslexia Association (International Dyslexia Association, n.d.),
overcoming dyslexia and other learning difficulties can be achieved through multisensory re-
education, which involves the use of visual, auditory, and kinesthetic-tactile pathways simultane-
ously in order to enhance memory and written language learning. The Web is a very useful tool
for individuals with disabilities. Therefore, it is important for organizations to design Web sites
that are accessible by all kinds of individuals. The Internet helps improve the life of individuals
with disabilities, and Web accessibility can improve the experience of these populations. In order
to create accessible Web sites, Web developers need to follow the guidelines set by the W3C
(2008). It is easier to develop an accessible site from scratch, than to improve an existing one
(White et al., 2005). Accordingly, designing CAPTCHA tests adapted for users with learning dis-
abilities is necessary.
It appears that appropriate and careful reference to the findings and conclusions of this research
on the part of user experience experts, developers, and web designers can lead in the future to
applying solutions that are more suitable for many users and especially for populations with
learning disabilities. Such solutions could benefit significantly the accessibility of the Internet and
improve the user experience on many Websites.
As described, the game-based CAPTCHA tests may be the best choice for users with learning
disabilities. Still, it should be noted that visually impaired users might take more time to solve a
test of this type.
The solution chosen by the developers of the Website must take into consideration, on the one
hand, the security level needed and, on the other hand, the user experience and frustration in solv-
ing the test.
Gafni & Nagar
221
Limitations and Further Research
The main limitations of this study are the following:
(1) The small number of participants with learning disabilities compared with those without (60
with LD, and 152 without). A larger number of users with learning disabilities might have
provided a better representative sample.
(2) The definition of users with learning disabilities some were diagnosed, some were self-
reported, but not diagnosed, and others may be not aware of having learning disabilities.
Further research is needed, especially for the No CAPTCHAtest, where findings were unclear.
This test seems obvious and easy to perform, for all the population, regardless of any disability.
For instance, it would be good to examine separately how many users were presented with an ad-
ditional task to complete after this test, and in which cases it is being used.
Another recommendation for future research would be to examine the influence of different de-
vices in solving CAPTCHA tests, e.g., through mobile phones or computers, on performance and
especially on user experience. It may be found, that performing those tests on touch-screens, like
in smartphones and tablets, are easier than on a desk-computer, in which a keyboard and a mouse
must be used. However, if typing is needed, a larger screen and keyboard may be more comforta-
ble for users with learning disabilities. A study about CAPTCHA tests has been recently conduct-
ed, using Nielsen’s heuristic evaluation (Reynaga, Chiasson, & van Oorschot, 2015). Their re-
searchs aim was to propose and validate a set of heuristics for evaluating CAPTCHA schemes on
smartphones. However, they did not investigate the issue among specific sensitive populations, in
the way suggested here.
References
Abrich, R., Berbenetz, V., & Thorpe, M. (2011). Distinguishing between humans and robots on the Web.
Retrieved December 1, 2014, from http://richardabrich.com/abrich-berbenetz-
thorpe_thecaptchaexperiment.pdf
Ahn, Y., Kim, N., & Kim, Y. S. (2013). A user- friendly image-text fusion CAPTCHA for secure web ser-
vices. In Proceedings of International Conference on Information Integration and Web-based Applica-
tions & Services, 550-554.
Angre, A. R., Kapadia, M. D., & Ugale, M. (2015). PiCAPTion: Picture CAPTCHAs for internet
authentication. International Journal of Computer Applications, 114(10).
Azad, K. J. (2013). CAPTCHA: Attacks and weaknesses against OCR technology. Global Journal of Com-
puter Science and Technology, 13(3).
Baltar, F., & Brunet, I. (2012). Social research 2.0: Virtual snowball sampling method using Facebook.
Internet Research, 22(1), 57-74.
Bevan, N. (2009). What is the difference between the purpose of usability and user experience evaluation
methods. UXEM’09 Workshop, Upssala, Sweden.
Bursztein, E., Bethard, S., Fabry, C., Mitchell, J. C., & Jurafsky, D. (2010). How good are humans at solv-
ing CAPTCHAs? A large scale evaluation. In Security and Privacy (SP), Symposium on Security and
Privacy IEEE, 399-413.
Fidas, C., Voyiatzis, A., & Avouris, N. (2011). On the necessity of user-friendly CAPTCHA. In Proceed-
ings of the SIGCHI Conference on Human Factors in Computing Systems, 2623-2626.
Foley, A. (2012). Biometric alternatives to CAPTCHA: Exploring accessible interface options. Masters
Dissertation, Dublin Institute of Technology, Ireland.
Ghazarian, A. (2014). CAPTCHAs’ effect on UX and how to fix it. Retrieved December 15, 2014, from
DesignModo: http://designmodo.com/ux-captcha-effect
CAPTCHA: Impact on User Experience of Users with Learning Disabilities
222
Gossweiler, R., Kamvar, M., & Baluja, S. (2009). Whats up CAPTCHA? A CAPTCHA based on image
orientation. In Proceedings of the 18th international conference on WWW, 841-850.
Hammill, D. D. (1990). On defining learning disabilities: An emerging consensus. Journal of learning Dis-
abilities, 23(2), 74-84.
Hande, S. G., & Ali, M. S. (2015). Enhancing the security using CAPTCHA as a graphical password. In-
ternational Journal of Advance Research in Computer Science and Management Studies, 3(4), 346-
352.
Hsu, C.-H., & Lee, Y. -L. (2011). Effects of age groups and distortion types on text-based CAPTCHA
tasks. In J. A. Jacko (Ed.), Human-computer interaction. Users and Applications (pp. 435-455). New
York: Springer-Verlag Berlin Heidelberg publications.
International Dyslexia Association. (n.d.). Multisensory structured language teaching. Retrieved June 7,
2016, from: http://eida.org/multisensory-structured-language-teaching/
Katchergin, O. (2015). How many learning disabled are there? The rhetorical power of statistics in the Is-
raeli discourse on learning disabilities. Open Journal of Social Sciences, 3(09), 155-166.
Lee, R. (2015). Dyslexia font simulates how its like to read if youre dyslexic. TechTimes. Retrieved June
9, 2016, from http://www.techtimes.com/articles/58567/20150608/dyslexia-font-simulates-how-its-
like-to-read-if-youre-dyslexic.htm
Madeira, J., Silva, C., Marcelino, L., & Ferreira, P. (2015). Assistive mobile applications for dyslexia. Pro-
cedia Computer Science, 64, 417-424.
Matej, S. (2013). Strict standards: Introduction to CAPTCHA accessibility. Retrieved June 1, 2014, from
CAPTCHA.com: http://captcha.com/captcha-accessibility.html
McCarthy, J. E., & Swierenga, S. J. (2010). What we know about dyslexia and web accessibility: A re-
search review. Universal Access in the Information Society, 9, 147-152.
Mohamed, M., Sachdeva, N., Georgescu, M., Gao, S., Saxena, N., Zhang, C., & Chen, W. B. (2013).
Three-way dissection of a game-CAPTCHA: Automated attacks, relay attacks, and usability. Cornell
University. New York: arXiv:1310.1540.
Mujumdar, D., & Polisetti, S. (2011). A platform to monetize usable & secure CAPTCHAs for desktop and
mobile devices (PICATCHA). Retrieved Dec 20, 2014, from Kevinwarnock:
http://www.kevinwarnock.com/wp-
content/uploads/2012/05/picatcha_mims_final_report_summary_0.pdf
National Joint Committee on Learning Disabilities (NJCLD). (1994). Learning disabilities: Issues on defi-
nition-revised. In NJCLD (Ed.), Collective perspectives on issues affecting learning disabilities (pp.
61-66). Austin, TX: Pro-Ed.
Onwudebelu, U., Sanjo, F., Obi, N. C., & Alaba, O. B. (2010). CAPTCHA malaise: Users suffer conse-
quences of the anti-spam technology while the spammers adapt. In Proceeding of the 1st International
Conference and Workshop on Software Engineering and Intelligent Systems, 113-125.
Pascual, A., Ribera, M., & Granollers, T. (2015). Impact of web accessibility barriers on users with a hear-
ing impairment. Dyna, 82(193), 233-240.
ReCAPTCHA. (n.d.). In Wikipedia. Retrieved May 8, 2016, from
https://en.wikipedia.org/wiki/ReCAPTCHA
Reddy, K. V., Krishna, D. S. R., & Reddy, D. J. (2015). CAPTCHA and its techniques for providing securi-
ty in web and applications. International Journal of Research, 2(8), 815-821.
Reynaga, G., Chiasson, S., & van Oorschot, P. C. (2015). Heuristics for the evaluation of CAPTCHAs on
smartphones. In Proceedings of the 2015 British HCI Conference, ACM 126-135.
Robson, K., Plangger, K., Kietzmann, J. H., McCarthy, I., & Pitt, L. (2015). Is it all a game? Understanding
the principles of gamification. Business Horizons, 58(4), 411-420
Gafni & Nagar
223
Sagirani, T., Nugroho, L. E., Santosa, P. I., & Kumara, A. (2015). User experience model in the interaction
between children with special educational needs and learning media. In 2015 2nd International Con-
ference on Information Technology, Computer, and Electrical Engineering (ICITACEE) (pp. 72-75).
IEEE.
Saini, B. S., & Bala, A. (2013). A review of bot protection using CAPTCHA for web security, IOSR Jour-
nal of Computer Engineering, 6, 36-42.
Shirali-Shahreza, M. H., & Shirali-Shahreza, M. (2006). Persian/Arabic CAPTCHA. In Proceedings of the
IADIS International Conference on Applied Computing (pp. 258-265).
Singh, P. V., & Pal, P. (2014). Survey of different types of CAPTCHA. International Journal of Computer
Science and Information Technologies, 5(2), 2242-2245.
Tangmanee, C., & Sujarit-apirak, P. (2013). Attitudes towards CAPTCHA: A survey of Thai internet users.
The Journal of Global Business Management, 9, 29-41.
Von Ahn, L., Blum, M., & Langford, J. (2004). Telling humans and computers apart automatically. Com-
munications of the ACM, 47(2), 56-60.
W3C. (2008). Web Content Accessibility Guidelines (WCAG) 2.0. Retrieved December 10, 2014, from
World Wide Web Consortium: http://www.w3.org/TR/WCAG20
White, J., Goette, T., & D. Young, D. (2005). Measuring the accessibility of the U.S. State Government
web sites. Communications of the International Information Management Association 5(1), 31-40.
WHO. (2015). Fact sheet N°300: Deafness and hearing loss. World Health Organization. Retrieved June 7,
2016, from: http://www.who.int/mediacentre/factsheets/fs300/en/
Williams, P., & Hennig, C. (2015). Effect of web page menu orientation on retrieving information by peo-
ple with learning disabilities. Journal of the Association for Information Science and Technology,
66(4), 674-683.
Biographies
Ruti Gafni is the Head of the Information Systems B.Sc. program at
The Academic College of Tel Aviv Yaffo. She holds a PhD from Bar-
Ilan University, Israel (in the Business Administration School), focus-
ing on Information Systems, an M.Sc. from Tel Aviv University and a
BA (Cum Laude) in Economics and Computer Science from Bar-Ilan
University. She has more than 30 years of practical experience as Pro-
ject Manager and Analyst of information systems. She also teaches in
the Management and Economics MBA program at the Open University
of Israel.
Idan Nagar is a graduate of the Information Systems B.Sc. program at
The Academic College of Tel Aviv Yaffo, and a student of the MA
Interdisciplinary design innovation and entrepreneurship at The Col-
lege of Management Academic Studies. He worked as an Information
Systems implementer, and served as a data analyst in the Israeli Air
Force. Idan is a Mobile-Web developer, especially interested in Hu-
man-Computer Interaction and particularly in UI/UX design, develop-
ment and measurement
... With the spread of the Internet and its technologies, it has become necessary to propose and develop online verification methods, CAPTCHA, and facilitate it to all society members in proportion to their special needs, such as people with autism spectrum disorders [7], learning disabilities [8], and blinds or visual impairment [9]. For Internet users with a visual disability, our target in this research, the audio CAPTCHA overcomes limitations of other CAPTCHA types because it is based on what they hear, not what they see [1]. ...
... Findings show that most existing CAPTCHA requires intensive mental and recalling efforts to remember all digits. Unlike the IRemember CAPTCHA method requires less workload efforts [4][5][6][7][8][9][10][11][12][13][14][15][16]. ...
... It has been found that difficult captchas can have a major negative impact on user experience [8]. This effect is even greater when the user is elderly or has accessibility issues [9] . In the worst cases, people with limited sensory or cognitive abilities may abandon their session due to difficulties with a captcha. ...
... As mentioned earlier, a period of enrollment is needed to achieve a high success rate for user classification [9]. The enrollment period refers to the time-period when a user account doesn't have enough data gathered to successfully create a biometric pattern based on the user behaviour. ...
Conference Paper
Full-text available
In this paper, we consider a novel method of mining biometric data for user authentication by replacing traditional captchas with game-like captchas. The game-like captchas present the user with a short game in which they attempt to get a high score. The data produced from a user's game play will be used to produce a behavior biometric based on user interactions, such as mouse movement, click patterns and game choices. The baseline expectation of interactive behavior will be used as a single factor in an intrusion detection system providing continuous authentication, considering the factors such as IP address, location, time of use, website interactions, and behavior anomalies. In addition to acting as a source of data, game-like captchas are expected to deter bots and automated systems from accessing web-based services and improving the user experience for the end-users who have become accustomed to monotonous alternatives, such as Google's re-captcha.
... CAPTCHAs have evolved over time, setting the user different types of challenges (Gafni and Nagar, 2016;Kaur and Cook, 2019) (See Figure 1). Initially, they required the transcription of distorted texts or short audios. ...
... There are a number of usability issues, as CAPTCHAs are more difficult for people facing either the physical decline often associated with older people (Prusty, n.d.) or learning disabilities (Kaur and Cook, 2019). There is evidence that they also discriminate against individuals in peripheral cultures, as the texts, audios and images tend to be a reflection of prevailing cultural Western content (Gafni and Nagar, 2016). In particular, image-based CAPTCHAs require the identification of images containing cultural elements that may differ from one context to another, meaning that people from peripheral cultures will find it more challenging to perform the identification task correctly. ...
Article
Full-text available
Ageism is the most invisible form of discrimination. While there is some awareness of gender, racial, and socioeconomic discrimination on digital platforms, ageism has received less attention. This article analyzes some tools that are frequently embedded on digital platforms from an old-age perspective, in order to increase awareness of the different ways in which ageism works. We will firstly look at how innovation teams, following homophilic patterns, disregard older people. Secondly, we will show how ageism tends to be amplified by the methods often used on digital platforms. And thirdly, we will show how corporate values contradict the usability issues that mainly affect people with a low level of (digital) skills, which is more common among older people. Counterbalancing the abusive power control of the corporations behind digital platforms and compensating for the underrepresentation of groups in less favorable situations could help to tackle such discrimination.
... CAPTCHA (Completely Automated Public Turing test to tell Computers and Human Apart) is a commercially accepted security mechanism on many websites. It is used as an anti-spam test to deter automated programs and bots' such as "Brute-force" or "Spam-bots" [1][2][3]. ...
Chapter
Full-text available
Turing tests are used to secure the human interaction on the Internet. Tests such as CAPTCHA are based on visual or auditory recognition of symbols and are difficult to distinguish by elderly people. A study examining the consistency of a tactile feedback-based Turing test identified an alternative to mainstream tests. This approach examines the vibration-based sensitivity which is detectable through skin surfaces when used to touch the screen of a mobile device. The study concentrated on a range of rough, smooth, sticky and coarse textures as possible differentiators for swipe-based tactile authentication using mobile devices. This study examined the vibration-based touch screen capabilities of 30 elderly people over the age of 65. The results of this study showed that tactile differentiation can be a viable alternative for device and security authentication for Turing tests such as those used for CAPTCHA and reCAPTCHA verification.
Article
Ageism is the most invisible form of discrimination. While there is some awareness of gender, racial, and socioeconomic discrimination on digital platforms, ageism has received less attention. This article analyzes some tools that are frequently embedded on digital platforms from an old-age perspective, in order to increase awareness of the different ways in which ageism works. We will firstly look at how innovation teams, following homophilic patterns, disregard older people. Secondly, we will show how ageism tends to be amplified by the methods often used on digital platforms. And thirdly, we will show how corporate values are often against usability issues that mainly affect people with a low level of (digital) skills, which is more common among older people. Counterbalancing the abusive power control of the corporations behind digital platforms and compensating for the underrepresentation of groups in less favorable situations could help to tackle such discrimination.
Conference Paper
Full-text available
As the bar for authentication is raised, so is the risk that many users may be marginalized. A common method of limiting access to services made available over the Web is visual verification of a bitmapped image-CAPTCHA. This presents a major problem to users who are blind, have low vision, or have a learning disability such as dyslexia or users who for whatever reason cannot answer the puzzle. CAPTCHA authentication is an effective tool in the war on spammers but it discriminates and excludes people, exactly what the internet was not supposed to do. We do not believe weighing spam against a significant percentage of internet users is fair. This paper reports the findings of an experiment conducted to examine users' dissatisfaction on CAPTCHA and provides a number of potential solutions that will help CAPTCHA developers and sites owners to involve the users in the developmental process rather than having the spammers solely in their hearts as they come up with various forms of CAPTCHA systems to test for the humanity of users.
Article
Full-text available
Several user tests were carried out on people with a hearing impairment to evaluate the impact of different web accessibility barriers on two similar web sites, one accessible and the other not accessible. The tests’ focus was to analyze users’ moods when faced with different accessibility barriers. Results show “complex text” and “multimedia content without text alternative” as the most critical barriers for users with this profile. Our investigation contributes to a better understanding of users when confronting accessibility barriers, and to emphasize the need of web content authors to use plain language and to provide captions and sign language alternatives in video content.
Article
Full-text available
The intention of this article is to problematize the current understanding of learning disabilities by scrutinizing the historical and social context in which they are embedded. The first part of the article lays out theoretical assertions from various fields: sociology of knowledge, rhetoric of science, rhetoric of statistics and historical critical discourse analysis. Integrating these constructionist approaches and through a short historical presentation of the evolution of the discourse and the various critiques that are developed from it, the article reveals the obscurity that surrounds the concept of learning disabilities. In the second part, the article examines one important idiom which forms the basis for the Israeli disabilities discourse: the statistical one which deals with the percentage of disabled persons in the population. Through an analysis of major texts of the Israeli disabilities field and interviews with professionals, it becomes clear how central statistical assertions are shaped into “scientific facts”, even when their scientific foundations are quite shaky. The article’s aim is to contribute to the development of a more complex disabilities discourse by uncovering its social, historical and cultural contexts. Another aim is to raise awareness to possible uses of statistical knowledge as a discursive and rhetorical tool.
Article
Full-text available
The ability to read is one of the main skills of a human being. However, some of us have reading difficulties, regardless of social status, level of intelligence or education. This disorder is the main characteristic of dyslexia and is maintained throughout life, requiring early and specialized intervention. Dyslexia is defined as a learning disturbance in the area of reading, writing and spelling. Although the numbers of prevalence rely heavily on the type of investigation conducted, several studies indicate that up to 17% of the world population is dyslexic, and that men have greater prevalence.
Conference Paper
Full-text available
The utilization of Information and Communication Technology (ICT) is not an unfamiliar thing and has reached almost every aspect of human life, including aspects of education and learning. The use of ICT has become a fundamental requirement in supporting the effectiveness and quality of the educational process, especially in supporting creativity and accuracy in the development and utilization of learning media. The need to develop learning media is necessary for special educational needs (SEN), which is to educate and prepare children with SEN to be able to live independently in the society. The development of learning media for children with SEN should be arranged with a clear study on the user experience (UX) model development framework. The existing challenge is how learning media could be developed with the focus on UX aspect and the ability/ limitations of children with SEN, with the hope that the prepared learning media can be used to help these difficulties and limitations in SEN, which in turn motivate children with SEN to learn. UX modeling aims to optimize the development of learning media that is appropriate for SEN, thus is expected to enhance the children with SEN experience in the process of knowledge acquisition and absorption as an independent life provision.
Article
Full-text available
There is growing interest in how gamification—defined as the application of game design principles in non-gaming contexts—can be used in business. However, academic research and management practice have paid little attention to the challenges of how best to design, implement, manage, and optimize gamification strategies. To advance understanding of gamification, this article defines what it is and explains how it prompts managers to think about business practice in new and innovative ways. Drawing upon the game design literature, we present a framework of three gamification principles—mechanics, dynamics, and emotions (MDE)—to explain how gamified experiences can be created. We then provide an extended illustration of gamification and conclude with ideas for future research and application opportunities.
Conference Paper
The utilization of Information and Communication Technology (ICT) is not an unfamiliar thing and has reached almost every aspect of human life, including aspects of education and learning. The use of ICT has become a fundamental requirement in supporting the effectiveness and quality of the educational process, especially in supporting creativity and accuracy in the development and utilization of learning media. The need to develop learning media is necessary for special educational needs (SEN), which is to educate and prepare children with SEN to be able to live independently in the society. The development of learning media for children with SEN should be arranged with a clear study on the user experience (UX) model development framework. The existing challenge is how learning media could be developed with the focus on UX aspect and the ability/ limitations of children with SEN, with the hope that the prepared learning media can be used to help these difficulties and limitations in SEN, which in turn motivate children with SEN to learn. UX modeling aims to optimize the development of learning media that is appropriate for SEN, thus is expected to enhance the children with SEN experience in the process of knowledge acquisition and absorption as an independent life provision.
Conference Paper
Captchas are used as a security mechanism on the web to distinguish human users from automated programs. However, existing captchas are not well-adapted to mobile devices and may lead users to abandon tasks. Although Web developers have many available captchas, they lack the tools to evaluate if these captchas are suitable for their mobile site. In this paper, we present domain specific usability heuristics for evaluating captchas on smartphones. To assess effectiveness, we compared our proposed heuristics against Nielsen's during evaluations of four captcha schemes on smartphones. The custom heuristics revealed more major problems and more detailed feedback on the problems than Nielsen's.
Article
CAPTCHA stands for "Completely Automated Public Turing test to tell Computers and Humans Apart" and has received much attention since it first began appearing on websites. It requires the deciphering of distorted texts, mostly in English which is something that computers still cannot do well. It is also helpful in preventing the abuse of online services. The current text-based CAPTCHA requires users to be able to read English characters. For Thai Internet users who might not be very familiar with English, a Thai language based CAPTCHA may be a more appropriate option. To date, no published work has examined the extent to which Thai Internet users are familiar with CAPTCHA; therefore, this study attempts to survey their awareness of, and attitudes towards, the online test. Based on 340 usable online questionnaire submissions, it was found that Thai Internet users are generally aware of CAPTCHA, but their understanding of it does not go very deep. Using exploratory factor analysis, their attitudes towards CAPTCHA can be classified in two dimensions: (1) the perceived drawbacks of the CAPTCHA test and (2) the feasibility of Thai language CAPTCHA. In addition to providing our insights into the application of CAPTCHA in the Thai Internet user context, online service providers could take certain measures to improve users' attitudes and understanding regarding CATPCHA.
Article
The basic challenge in designing these obfuscating CAPTCHAs is to make them easy enough that users are not dissuaded from attempting a solution, yet still too difficult to solve using available computer vision algorithms. As Modern technology grows this gap however becomes thinner and thinner. It is possible to enhance the security of an existing text CAPTCHA by systematically adding noise and distortion, and arranging characters more tightly. These measures, however, would also make the characters harder for humans to recognize, resulting in a higher error rates and higher Network load .This paper presents few of most active attacks on text CAPTCHAs existing today.