Using technology in teaching learning-disabled

Written by: Helen Trugman Ph.D., Edith Gotesman Ph.D.

Abstract: Our paper discusses the use of innovative technological solutions in foreign language teaching for a particular student population, specifically learning-disabled (LD) students. The present research summarizes four academic projects implemented in institutions of higher education in Israel: Holon Institute of Technology, Ashkelon Academic College, Netanya Academic College and Levinski Teacher Training College.

Keywords: dyslexia, learning-disabled (LD) students, assistive technology, presentation accommodations, text-to-speech engines, reading software

[↑]

1. Introduction

The present projects examine ways to assist language learners in developing foreign language literacy with the help of computers. We strongly believe that in a competitive modern environment of the ‘global village’, computer literacy becomes a prerequisite of professional accomplishment for any person in general, and for a learning-disabled one, in particular.  To function efficiently in a global economy, one has to build up and constantly perfect her learning strategies, starting at school, through college and continuing all life. This paper shows how developing reading skills with the help of computer software can assist LD students in achieving this aim.

The main goal of the paper is to raise English teachers’ awareness of the existing assistive technology, specifically, text-to-speech engines (or reading software). We discuss multiple advantages of such reading software over traditional accommodations provided for LD students, mention some logistical and financial considerations involved in setting up language programs using such software, and provide some suggestions on the choice and use of particular software.

[↑]

2. Target population

The projects under discussion targeted a heterogeneous LD population in four different colleges in Israel. The majority of the participants were assessed as dyslexics, with some students diagnosed additionally as having Attention Deficit (Hyperactivity) Disorder (AD(H)D). Some other students had problems with short-term memory. In several colleges visually impaired students and poor readers of English were also involved in the projects.

[↑]

2.1 What is dyslexia?

defines the specific reading disability dyslexia as “a difficulty in learning to read in a person who has good intelligence, strong motivation, and who has received appropriate teaching”. Dyslexia is not a developmental lag, but rather represents a chronic condition. It cannot be cured and often stays with the person through the lifespan (Gerber, Ginsberg and Reiff 1992, Shaywitz and Shaywitz 2001 and references cited therein). However, many LD students have normal or above-normal intelligence and exhibit high-level academic achievement despite their reading deficiency (Shaywitz 2003). Since such students cannot be denied full and equal access to educational opportunities in instruction and testing by law, they have to be provided with appropriate accommodations both in class and on tests. These alternative teaching and assessment options are aimed to measure these students’ abilities rather than disabilities. To do otherwise puts them at risk of joining the grave statistics of underachievers or failures.

[↑]

2.2 Accommodations provided for LD students

In this section we give a brief synopsis of various accommodations that LD students enjoy in institutions of secondary and higher education in the West (see Landau, Vohs, & Romano 1998 for a complete list).

(1)     Accommodations for LD students

  • timing/scheduling accommodations

LD students usually get extension of time on tests and exams. Sometimes, an LD student can be tested at a time most beneficial to her—for instance, tests can be administered in the morning; or students can be allowed short rest breaks during tests to regain concentration level.

  • setting accommodations

Dyslexic students can be tested in a small group in a separate room with a proctor (this is especially beneficial for students with AD(H)D). When problems with concentration are coupled with stress and Test Anxiety syndrome, a student can be tested without a proctor in a separate room.

  • presentation accommodations

The way a test is formatted is not of little importance: large-print editions of texts and question sheets are very helpful, as well as increased spacing between and within test items. It is recommended to use standard instructions for test tasks, consistent with those used in class assignments, in order to reduce tension which might be created by an unusual task instruction. Instructions should be given in simplified language and the key words in test instructions should be highlighted, underlined or boldfaced, so as to help LD students get quickly to the gist of the question. Some dyslexic students get a permission to use assistive technology, such as audio-taped texts and questions, text-to-speech engines, electronic or online dictionaries, etc.

  • response accommodations

When funds permit, a tester can be provided to transcribe the answers of a dyslexic student (or a dysgraphic one) on a test. When assistive technology is available, a dysgraphic student can use a word processor with a spell check to type in her answers on a test. If a test is administered with a pen and paper, then quite often spelling and punctuation requirements are waived. Since many dyslexics appear to have low self-esteem (Riddick, Sterling, Farmer and Morgan 1999), it is advisable to run periodic checks on the student’s progress with a test and to bring students back on track if needed.


[↑]

3. Advantages of computers over traditional methods

[↑]

3.1 What is Assistive Technology?

As we can see in (1), various forms of technology can be provided to LD students as part of response or presentation accommodations. Technology used as an educational aid is known as assistive technology (AT), which is defined in Raskind and Higgins (1998: 1) as “any technology that enables an individual with a learning disability to compensate for specific deficits”. In some cases such technology helps LD students to improve their performance in a given area of disability, whereas in others it can be used to “bypass” their specific deficits entirely.

Recently, researchers have been investigating ways in which computers and other forms of advanced technology can be utilized to support LD students in their effort to succeed academically (Raskind & Higgins 1998, Anderson-Inman 1999, Pisano 2002). In the following section we compare traditional methods used for facilitating LD-students work with text-to-speech engines and discuss the former’s downsides.

[↑]

3.2 Audio aids versus text-to-speech engines

Traditionally, the dyslexic population has been dealt with by providing additional instruction to such students, the so-called ‘private tutoring’, and by using human testers transcribing the student’s answers or testing her orally on tests. Such methods presuppose, first and foremost, availability of qualified tutors and testers, as well as proper scheduling of tutoring sessions and exams (i.e. finding time slots convenient for both a tutor and a student). In addition, the institution has to provide facilities for tutoring and testing sessions, and, last but not least, find funds to pay for the services.

If traditional audio aids are used instead of or in addition to human testers, then one has to take into consideration purchasing necessary equipment (tape-recorders, cassettes, players, etc.), besides paying to English-speaking personnel for taping all the necessary materials, and to proctors supervising students during tests.

Confronted with such considerable investments in time and money, some institutions have been recently replacing audiocassettes created by human readers by audio CDs recorded with the help of speech synthesizers. This solution eliminates some logistical and financial problems mentioned above (for instance, a need for an English-speaking reader to tape texts), but only to some extent. Other problems remain, as shown in (2):

(2)     Some logistical and financial problems with using audio CDs

  • Creation of audio CDs/mp3 files requires computer-literate personnel.
  • Though faster than recording a tape, it is still time-consuming to create an audio CD appropriate for a test, i.e., with each paragraph of the text recorded as a separate track.
  • Multiple CDs are required if several LD students take the same exam.
  • Every minor change in a text or question sheet requires preparation of a new CD, which additionally increases investment in time and money.
  • With the advent of mp3 and mp4 gadgets, CD players become less available on the market.

In addition to the logistical and financial downsides mentioned in (2), audio aids exhibit a number of other serious drawbacks, which led many students to see them as an unsatisfactory answer to their needs. For instance, students find it time-consuming to locate a particular sentence on the cassette/CD even if tracks are created. Nor is it possible to get fast to a particular word on the tape/CD. It is also hard to create an exact correspondence between an audio and a visual sign, i.e. between the taped text and the printed one. Quite often students seem to be dissatisfied with the tempo of reading or the reader’s accent, even when a tape/CD has been created with a human voice.

Irrespective of all these drawbacks, however, it seems that audio support in this or that form (either as audio cassettes or audio CDs) still prevails as a technological solution offered to the LD student across many educational institutions. What is the better alternative? We claim that computer technology, specifically speech-to-text engines converting a written text into speech, can provide a better solution to the LD student’s problems (cf. Norris and Graef 1990). Compared with audiocassettes/ audio CDs, computerized speech synthesis offers several important advantages, summarized in (3):

(3)     Advantages of text-to-speech engines:

  • Computer software allows for more efficient reading comprehension due to simultaneous multi-sensory support—students both see the text on screen and hear it being read.
  • Computer software improves listening comprehension by allowing the user to create personal ‘readers’ and thus eliminating problems with the human reader’s accent and tempo of speech (in TextAssist program); or by choosing a human voice (in TextAloud program), whose speech can also be adjusted to the user’s needs. Importantly, voice editing can be done on the spot and as many times as needed.
  • Using reading software on a test is less time-consuming for the LD student than audio aids since any particular word or phrase can be easily found on screen and repeatedly read by the program. There is no need to constantly wind and re-wind a tape or a CD.
  • Doing a test on a computer enables faster scanning for key words in questions thanks to a smart use of computer functions (Ctrl+F), which also saves the LD student’s time and bypasses his inability to scan.
  • It is more convenient for the LD student to do a test on a computer since it allows replacing several written sources (a test paper and a dictionary) with one medium—a computer. Students can arrange the text, the question sheet and the dictionary in different windows on screen. Organizing in this way their virtual working place helps them concentrate on the task and stay in focus longer.

Even though using computer technology also entails a serious investment of money and time, especially at the initial, setting-up stage, eventually it becomes less expensive and more manpower- and time-efficient than using audio-taped materials. The following section presents the text-to-speech engines used in the projects and discusses some pre-requisites for starting such a project.

[↑]

4. Selecting the appropriate technology

According to Raskind & Higgins (1998) and Raskind (1998), selecting the appropriate technology for an individual with a learning disability requires careful analysis of the interplay between the individual (a dyslexic student in our case), the specific tasks or functions to be performed (decoding and understanding a written text), the specific technology (a text-to-speech engine) and the specific context of interaction (doing a homework assignment or writing a test). Knowing the needs of learners can help language instructors with the choice of a specific technological aid. However, as with the use of audio aids and human testers, there are other factors that might affect the choice of text-to-speech software. Table 1 summarizes technical parameters and features of the three programs used in four different colleges in Israel: ReadPlease, TextAloud and TextAssist. After pinpointing major upsides and downsides of each program, we discuss organizational and financial considerations that should be taken into account when choosing a proper text-to-speech engine for a specific student population.

[↑]

4.1 Comparison of different reading programs

Table 1: Features of text-to-speech engines used in the projects

ReadPlease (ReadPlease Plus) TextAssist TextAloud
Synthesized voices + (only 4 readers) + (28 readers) +
AT&T Natural voices + (should be purchased separately) + (should be purchased separately)
Voice modification possibilities speed only very precise adjustment (speed, (singing) pitch, volume, intonation, echo, frication rate, smoothness, richness, etc.) somewhat limited (only speed, pitch, volume ), only speed with human AT&T voices
File types read e-mail, txt, doc, rtf, (unprotected) pdf, html e-mail, txt, doc, rtf, pdf, html e-mail, txt, doc, rtf,  (unprotected) pdf, html
Uploading the text into the program the text should be copied-pasted into the program’s main window works directly in the original file; no need to copy-paste the text is opened in the program’s main window in .txt mode
Highlights text when reading + + +
The text is read in the program’s main window in a separate window above the original file in the program’s main window
Program window size 640×480 (full screen in ReadPlease Plus) full screen full screen
Change of font and color + + +
Smart keys for word search + (runs search from the original file by using Ctrl+F) + (somewhat limited search functions in the program’s main window)
Ease of operating

easy but time consuming; many very easy and user-friendly; personal reader profile creation requires more training easy but learning  various functions and options, may require more extensive training
Records files for later listening ReadPlease does not have this option, only the paid version can record  .wav files +

can record files in .wav, MP3 and MP4 formats

Availability of free and trial versions ReadPlease is free

no trial version for ReadPlease Plus

no free or trial versions a 21-day trial version without natural voices
Cost of a downloadable version (AT&T voices are extra) $50 ($59.95 for the latest version of ReadPleasePlus)

www.readplease.com

$50

www.textassist.com

$30

www.textaloud.com

We should admit that money considerations often force an institution to prefer a less costly text-to-speech engine at the expense of student convenience. However, financial considerations should not be the primary driving force, for student satisfaction and motivation to use the program ultimately underlie the success of the whole project. We have found that students often begrudge their time spent on learning complicated commands and get easily frustrated if doing a test on screen requires multi-tasking, i.e. simultaneous work with several application windows. The student satisfaction stems mainly from the ease of learning and operating this technology, as well as from the availability of various functions simplifying the work with the test.

Though ReadPlease is the only freeware being used in the project, it turned out to have too many shortcomings to be recommended for institutional use (see also reviews on www.cnet.com website for detailed discussion of shortcomings by the editor and users). The need to cut and paste a body of text into the program’s main window is quite inconvenient. Neither does the program allow for customizing the reader’s voice or for a wide choice of readers. Particular difficulties arise with ReadPlease when a student has to work with two files simultaneously or read different portions of one file. Since there is no option allowing a student to skip to a certain paragraph, she has to upload a selected portion of the text/questions to the ReadPlease window or scroll down manually to locate that section in the file. This significantly complicates a student’s task and might cause frustration.

Similarly to ReadPlease, TextAloud also reads texts in the program’s main window, to which one or several texts may be uploaded simultaneously. And as in ReadPlease, uploading a file changes the format of the original file into a .txt format. Though not a problem for the program users who are just interested in reading long texts, it becomes a predicament on an English test—students cannot highlight or mark in any way topic sentences of paragraphs or key words they might need for the test. In addition, uploading a .doc file with the test into the program’s window automatically converts it into .txt format and eliminates the correspondence between the test on screen and the printed one, thus eliminating one of the major advantages of using reading software.

In this respect, TextAssist program appears to outperform all other programs since it allows its user to operate it from within the original file, be it an e-mail message, an Internet site, or a Word file. The TextAssist floating toolbar is very small and unobtrusive, which floats over any other program window and does not prevent simultaneous use of multiple windows on screen. For instance, a student can simultaneously open a file with a text in a .doc format, the Word Translator window, and the Text Assist program, and have them all appear on screen together. In contrast, TextAloud, which does not have an in-built translator, requires opening the Word Translator window or any other translator program and adjust both windows manually on screen. Moreover, translating words from TextAloud file using a separate translator program is more cumbersome than translating words from within a .doc file using the Word Translator. The same can be said about the search functions allowed by TextAssist and TextAloud programs: while TextAssist does not prevent using standard Word smart keys, such as Ctrl+F, for locating any word or string of words in a text (which is an indispensable tool for LD students, who find it hard to scan for words on their own); TextAloud comes with an in-built search application which searches for a word inside the program’s main window. Unfortunately, this search application differs from the standard Word application and has to be learnt separately; moreover, it does not always perform faultlessly. Hence, very rarely students opted to learn and use this application on tests.

Though TextAssist program seems to be the easiest to operate since it capitalizes on the user’s knowledge of Word and other programs, it has one significant downside in comparison to the other text-to-speech engines—its incompatibility with natural AT&T voices. The program’s voice menu provides a wide range of synthesized voices (28 in-built ones) and allows for their very precise adjustment in tone, pitch, speed, volume, echo, etc. In addition, a user can create her own reader by customizing one of those in the menu and save it for future use. Yet, students tend to complain about the quality of synthesized voices, especially at the beginning.

However, we didn’t find this to be a real stumbling block for the majority of LD students, especially for those who persist in using and practicing the program on a regular basis. Moreover, sometimes ‘mechanically sounding voices may actually be more intelligible’ (Raskind & Higgins 1998: 5), i.e., clearer to understand by non-native speakers than native speakers’ voices. Yet, in several cases the decision not to purchase TextAssist stemmed from the dissatisfaction with voice quality.

To summarize the discussion above, the institution administration and project heads have to consider the following issues, when deciding which text-to-speech program to choose for our students:

(4)     Major considerations to be taken into account

  • financial considerations (availability vs. lack of site licenses, of free downloadable or trial versions, etc.)
  • availability of natural voices
  • students’ time spent on learning and practicing the software
  • ease of using the application
  • time spent on scanning, editing or formatting tests and reading materials for the use with reading software
  • correspondence between the test on screen and on paper
  • possibility of creating audio files (mp3, wav. and other formats)

[↑]

4.2 Pre-requisites for starting the project using reading software

It is clear that the major pre-requisite for starting such a language teaching program is financial. First, the decision should be taken by the administration and necessary funds should be allotted for the project. It entails buying and installing necessary equipment in a computer laboratory (e.g. text-to-speech software licenses, earphones), as well as purchasing a CD-writer and re-writable CDs for creating individual test CDs for the students. Implementing such a project certainly requires time, commitment, and effort on the part of the project head. At the initial stage, it is necessary to convert all teaching materials into digital form, by typing them up or scanning.  Digital course materials can be handed out to LD students on individual CDs, or uploaded to online course sites (see Managed Learning Environments discussed in Barrett, Rainer and Marczyk 2007). To ensure that LD students know how to use the program, it is necessary to prepare manuals on program use for students (translating the instructions into their mother tongue) and to hold regularly instructional workshops for LD students enrolled in English courses. Students should be encouraged to use the program constantly in order to improve their skills and minimize the time needed for organizational matters. If the language program entails use of text-to-speech engines not only in class but also on tests, there is a need to train personnel for program use during tests—such as technicians responsible for computers maintenance, as well as proctors supervising students in a computer lab during tests.

Even though use of reading software in a language program may require a considerable investment of money and time, especially at the initial, setting-up stage, in the long run it becomes less expensive and more efficient than using audio-taped materials.

[↑]

5. Ways to combine traditional teaching with AT

In Israeli universities and colleges English instruction is provided in form of reading comprehension courses in English for Academic Purposes (EAP) which aim to equip students with the skills and strategies needed to meet their English reading requirements in academic courses. EAP courses often require very extensive reading in English, which presents an almost insurmountable task for any student with learning disabilities, especially dyslexic students. Such requirements put high demands on language instructors that frequently face a heterogeneous population in class. As a result, quite often LD students are considered to be a burden or sometimes incapable of foreign language acquisition. In order to ensure that all students, no matter how diversified their abilities are, acquire the proper knowledge of English, several projects were initiated in four academic colleges in Israel, each using a speech-to-text engine, yet within a distinct format.

The information concerning all the projects is summarized in Table 2 below. Note, that in the three out of four projects, regular students learned side by side with LD peers, with frontal teaching being the major kind of instruction in class. In such heterogeneous groups, LD students could manage reading materials in class in several different ways. In some projects, dyslexic students could use a laptop with a text-to-speech engine during class assignments. In another, students had access to the computer lab where they used the reading software, while their peers read the text in class. In still others, LDs students did not enjoy access to the reader in class, yet could do their homework with the help of one. Only in one project conducted in Levinski Teacher Training College, all dyslexic students studied as one group in a computer lab with a permanent access to the text-to-speech engine and other online language resources. Such students enjoyed all the benefits of assistive technology both in class and on tests. However, this class was composed of LD students of three different language levels, which greatly complicated the task of the language instructor.

Another important factor distinguishing the projects is homogeneous versus heterogeneous use of AT in teaching LDs. While two projects, in H.I.T. and Levinski, used only text-to-speech engines in English courses, the other two supplemented the use of computers with audio aids. This heterogeneous use of AT stemmed from logistical problems, for instance, unavailability of computer labs at all time slots when English courses are conducted.

[↑]

5.1 Synopsis of the four projects

Table 2: English programs using AT

ACADEMINC INSTITUTION HOLON INSTITUTE OF TECHNOLOGY

(H.I.T.)

ASHKELON ACADEMIC COLLEGE NETANYA

ACADEMIC COLLEGE

LEVINSKI TEACHER TRAINING COLLEGE
STARTING YEAR 2001-2007 2003 2005 2006
STUDENT POPULATION dyslexics, ADHD dyslexics, visually-impaired; poor readers dyslexics, dyslexics
READER USED TextAssist ReadPlease/

ReadPlease Plus

TextAloud ReadPlease/later TextAloud
ASSIGNMENTS USED FOR homework, unseen tests, final exams classwork, homework,

final exams since 2007

final exams all activities
ADDITIONAL AT none audio cassettes on final exams audio CDs created with TextAloud on tests and quizzes none
MAJOR COMPLAINTS quality of synthesized voices; insufficient  Word dictionary dealing with two ATs, bad voices, need to cut-paste texts speed & quality of voice;  two ATs; lack of experience with TextAloud bad voices of ReadPlease, students of different levels in one class
LOGISTICAL PROBLEMS lack of computer labs and creating personal CDs with tests sending LD students to the lab for reading during classes lack of computer labs, creating multiple audio CDs none
STUDENT

PERFORMANCE & SATISFACTION

improved reading fluency and comprehension;

boosted self-confidence;

higher grades with nearly 100% rate of success (1-2% of failures); more consistent student performance and fewer complaints with homogeneous use of AT

[↑]

5.2 Project results

It was found that in all the projects LD students significantly benefited from the use of assistive technology, both in class and on tests. Though the heterogeneous use of AT (i.e. using both computers and audio aids for language instruction) somewhat complicated the students’ work by requiring them to master two technological aids instead of one, we found no increase in the failure rate in these cases, in contrast to students’ anticipation. Nevertheless, the homogeneous use of assistive technology appears to be preferential for a number of reasons. First, in the absence of other aids provided for LDs students, they have no other choice but to learn and master the reading software. Constant use of the software, in turn, increases students’ confidence and, eventually, their satisfaction with the reader. This leads to improved performance on tests and, in the long run, to success in the course.

It was also found that the homogeneous approach to AT in language courses ensures higher grade consistency: we can anticipate success or failure on the final test based on the student performance at intermediate tests, if the same technological aid is used throughout the course. When confronted with a different technology on the final test, students might perform in a less predictable way, and their final grades can deviate from their class averages to a greater extent.

[↑]

6. Conclusions

Based on the findings accumulated in the four projects, we conclude that the best results are achieved when one kind of assistive technology is used consistently and practiced extensively both for homework, in class and on tests (cf. Levinski Project). It seems that simultaneous use of various technological aids sends LD students a wrong message—passing an English course by any means becomes their primary concern rather than acquiring a new learning tool for life. When a text-to-speech engine is mastered by a student it stops being just an aid for passing a language course and becomes an integral part of a student’s life, helping her to bypass her impairment and integrate into modern society.

Logistically, the best format for a language course using AT is conducting a course in a computer lab.  This both relieves students’ anxiety by exposing them to AT on a permanent basis and prevents scheduling problems. Unfortunately, the best format may be unrealistic due to financial or administrational restrictions. These considerations must be taken into account before a decision is made to start such a project. Though the optimal solution seems to be provided by a language learning center serving the needs of English-learning LD students, this is also the most expensive solution. Therefore, when funds are not forthcoming, a well-thought-out compromise may be needed for the interim period, for a partial solution is better than no solution at all.

[↑]

References

  • Anderson-Inman, L. (1999) “Computer-based Solutions for Students with Learning Disabilities: Emerging Issues”, Reading and Writing Quarterly, Vol 15, pp 239-249.
  • Barrett, R., Rainer, A., and Marczyk O. (2007) “Managed Learning Environments and an Attendance Crisis? “ The Electronic Journal of e-Learning, Vol 5, No. 1, pp 1-10, available online at www.ejel.org.
  • Gerber, P.J., Ginsberg, R., and Reiff, H.B. (1992) “Identifying Alterable Patterns in Employment Success for Highly Successful Adults with Learning Disabilities”, Journal of Learning Disabilities, Vol 25, pp 475-487.
  • Landau, J. K., Vohs J. R., and Romano, C. A. (1998) All Kids Count. Parents Engaged in Education Reform (PEER Project), Federation for Children with Special Needs, Boston, Massachusetts.
  • Norris, M., and Graef, J. (1990) “Screen Reading Programs for Students with Learning Disabilities”. In H. J. Murphy (Ed.), Proceedings of the Fifth Annual Conference on Technology and Persons with Disabilities, Vol 5, pp 491-499.
  • Pisano V. Leonard (2002) “How to Use Technology to Support Students with Learning Differences”, LD Resources, available online at http://www.ldresources.org
  • Riddik B., Sterling C., Farmer M. and Morgan S. (1999) “Self esteem and Anxiety in the Educational Histories of Adult Dyslexic Students”, Dyslexia, Volume 5, Issue 4, pp 227-248.
  • Raskind, M. H., and Higgins, E. L. (1998) “Assistive Technology for Postsecondary Students with Learning Disabilities: An Overview”, Journal of Learning Disabilities, Vol 31, pp 27-40.
  • Shaywitz, S. E., and Shaywitz, B. A. (2001) “The Neurobiology of Reading and Dyslexia”. Focus on Basics, NCSALL, August 2001.
  • Shaywitz, S. E. (2003) Overcoming Dyslexia, Vintage Books, New York.
  • Trugman, H. and Gotesman, E. (2007) “Computer-assisted language teaching for students with learning disabilities”, Proceedings of the 6th European Conference on e-Learning (ECEL 2007), Ed: Dan Remenyi, Oct. 4-5, 2007, Copenhagen, Denmark, pp 639-648.
  • www.cnet.com/downloads/windows/productivitysoftware/voicerecognition

Tags: , ,

No comments yet.

Leave a Reply

You must be logged in to post a comment.