Kurzweil 3000: A UDL Solution
Kurzweil 3000: A Universal Design Solution to
Text to Speech Software
Research Proposal
Sandra L. Magnussen
ETEC 500 65-C
Professor Bartosh
University of British Columbia
April 6, 2014
Kurzweil 3000, developed by Kurzweil Educational Systems is reading, writing and learning software for struggling students. Software such as Kurzweil 3000, and other text to speech tools can make a significant difference in supporting the development of decoding and fluency skills, as well as providing meaningful access to curriculum materials (HehirKatzman, 2012). Universal Design for Learning builds in support for diverse student needs into the curriculum as part of the original lesson, rather than adding these supports in afterwards. Universal Design for Learning (UDL) takes the burden of requesting adaptations away from the student, by having them built into the pedagogy of the classroom. Digital materials, which are highly flexible, current, and engaging, are used extensively in UDL classrooms. UDL maximizes the opportunities to participate while minimizing the impact of disabilities.
The purpose of this research proposal is to study the use of Kurzweil educational software in our local school district of 3000 students. While previously only available to a few students, our district purchased universal licensing for Kurzweil 3000 in the Fall of 2013, offering access to all enrolled students to the program on any laptop, mobile device, and at home through the web.
Statement of problem
How does use of the Kurzweil text to speech software along with available digital curriculum content, improve students’ independence, confidence and attitude towards reading?
Review of Literature
To educate all of our learners in the public school system, we must look at how we design for that inclusion. Currently, only one third of students enter high school proficient in reading (NAEP, 2009). Brain research shows us that struggling readers actually use different parts of their brains while reading, making learning the mechanics of decoding more difficult (Shaywitz, 2012). “Reading is about learning-it is about thinking and listening, considering and inferring” (Katz, 2012, p.102). Low achievers show improved literacy scores when educated in inclusive settings, and technology is invaluable in supporting students who are struggling. When information is presented through visual and auditory channels simultaneously, the speed of processing and memory recall are enhanced (MontaliLewandowski, 1996).
This literature review is subdivided as follows: I will first review literature providing background and a history of text to speech reading software. Universal Design for Learning and its role in building in support for students will follow. The importance of training and support for the integration of technology will then precede ananalysis of research on students ranging from Grade 3 to adult using Kurzweil software in a variety of settings.
History of Kurzweil Text to Speech Software
Over the last several years, software that helps to support individuals with reading disabilities has become widely available (Hecker, Burns, Katz, Elkind, & Elkind 2002; Elkind, 1998). Software, which converts printed text into speech,can support students with dyslexia and other reading difficulties, as well as English language learners, gifted learning disabled students and studentswith attention concerns. (Heckler et al., 2002; Elkind 1998; Floyd & Judge, 2012; Weber & Cavanaugh, 2006; Montali et al., 1996; Chiang, Liu, Lee & Shih 2010). Students who are poor readers due to any of the above disabilities find that reading is slow, inaccurate and very hard work (Elkind, 1998).
To extract meaning from text they usually find thatthey have to proceed slowly, re-read passages frequently, struggle to decode unfamiliar words, and interrupt their reading frequently to recover from fatigue and stress. Since they read slowly and intermittently, reading takes them much more time than it does their peers at school. As a result the quantity of material that they read is small, and since they get much less practice reading they have less opportunity to improve their reading skills (Elkind, 1998, p.1).
A number of studies have shown that when information is presented through visual and auditoral channels simultaneously (bimodal presentation), the speed of processing and memory recall are enhanced (Montali et al., 1996; Elkind, 1998; Hecker et al., 2002; Fletcher et al., 2006). Screen readers were originally designed for students who were blind or had low vision, but have now led to an increase in decoding rate, comprehension, and confidence for students with reading difficulties. The reading software Kurzweil was originally created by Cambium Learning in the early 1970’s to support students who were visually impaired. In 1992 a product called Book wise was created by Cambium as well and was the first text to speech solution for students with dyslexia. The updated Kurzweil 3000 followed in 1997. It overcame previous limitations in text to speech technology by providing full color illustrations with the text when scanned, creating an exact replica of the book. Kurzweil 3000 also presents text in a bimodal format (auditorily and visually) using different colors to highlight the reading passage as well as the exact spoken word. It includes dictionary definitions, choice in voice speed, size and type, word prediction writing support, and a variety of other study supports and translation tools. (Hecker et al., 2002; Floyd et al., 2012; Chiang et al., 2010; Kanitkar, Ochoa & Handel; McCulley 2012; Recently, Kurzweil 3000 has developed a web-basedsystem, which enables users to access their own folder from any Internet computer. The application Firefly makes this possible as well on mobile devices.(
Text to Speech Software and Universal Design for Learning
Computer driven technologies such as Positron Emission Tomography (PET) scans and fMRI’s demonstrate learning in the brain in ways never before seen. We know that the brain operations responsible for our cognitive abilities occur primarily in the cerebral cortex(KandelJessell, 2000), and radiographic imaging techniques have made it possible to visualize these areas in living subjects. What has been a fascinating advance is that the brain can be imaged while performing specific tasks, and those tasks can be related to discrete regions of the brain. PET scanning identifies specific regions in the cortexthat light up when a word is spoken, listened to, or read.These contrasting images illustrate how recognition of sensory information is distributed across different areas called networks. Dr. David Rose (Rose & Mayer, 2002),stresses the importance of interpreting brain research from an educational perspective to forge connections between Neuroscience and the classroom. Within the large network of the cerebral cortex there are many smaller networks specialized for performing particular kinds of processing and managing particular learning tasks. In his research, the brain is identified by the networks that reflect their functions: the recognition, strategic and affective networks. “The activation of these networks parallel the three prerequisites for learning described by the Russian psychologist Lev Vygotsky”. (Rose, 2008, p. 12). Understanding this as an educator is extremely important in planning instruction and technology tools for diverse learners in the classroom environment (Katz 2012; Hehir et al., 2012). Very briefly,
1. Recognition Networks located at the back of the brain enable us to interpret all information from our senses. PET scans have demonstrated that recognition processes are distributed laterally across many brain regions, so leading to vast individual differences when processing information
2.Strategic Networks: located in the frontal lobe these help to identify goals, make plans and to follow through. We sometimes refer to these skills as our executive functions and they vary greatly between learners
3. Affective Networks: attends to the learners’ engagement with materials, tools, people and contexts.This research is the backbone of the principles of Universal Design for Learning (
The literature studied reinforces this researchers belief that understanding brain research validates that students are spread across the spectrum in terms of the way they process information, and as educators we need to change our teaching methods, assessment strategies and access to curriculum materials to truly educate the whole child. We know now that there is not one typical learner, but a great variety of learners. A student who appears very disabled when asked to read a traditional print book may look absolutely brilliant in a video creation class or when given the tool that he needs to decode the book, software such as Kurzweil 3000. “Instilling flexibility into methods and materials maximizes learning opportunities, not only for students with identified learning disabilities, but for all students” (Rose, 2002).
In their research on students in post secondary with reading disabilities, Floyd et al (2012) found that the students in their study felt comfortable using text to speech technology in front of their peers. In fact having universal access to these software tools decreases the stigma of them being just helpful fordisabled learners. “With some research suggesting that one of the strongest indicators of AT abandonment is lack of peer acceptance, this high social acceptability is very encouraging for the likelihood of continued use (Floyd et al., 2006,p.60).The research by Fletcher (2006) made claims that whenaccommodations were given to two groups; one average readers and one struggling, which included the oral reading of proper nouns and conversation stems, the results show that only the students with a disability benefited from the support. This is a typical resultas it is students who are most delayed in decoding whobenefit most both academically and emotionally, as shown in the studies byMontali (1996); Elkind (1998) and Hecker et al., (2002);and in Iowa, where a satisfaction survey showed that low skilled readers felt more successful in terms of comprehension when using bimodal software ( Maurer, 2006).
In the Fletcher study (2006), arandom assignment was done with Grade 3 students, both those who had dyslexia and who were average decoders. Contrary to other findingsthough, this study claims that it is important to look at these results as, "Accommodations that benefit students with and without disabilities are a concern because when all students benefit from an accommodation and its provided to only those with a disability, the accommodation provides an unfair advantage to those with a disability" (Fletcher et al., 2006, p. 137).
I disagree with this statement. In work by Richard Lavoie, treating individuals equally is the unfair thing to do. Fairness is not giving everyone the same thing; fairness is giving individuals what they need. (Lavoie & Clarke, 1990). Thus the accommodation, which in this case, was oral reading, not bimodal software, is not providing an unfair advantage, it is levelling the playing field for the student with a disability. Fletcher usesthe term "interaction hypothesis" (p.138) to describe the term that accommodations should benefit only those with a disability. Construct -irrelevant variance then can be demonstrated when an accommodation does not alter a test validity or inflate the score in an unfair manner.An accommodation should eliminate at least one source of the variance that is not fundamental to the underlying construct, affecting the measurement of theconstruct of studentswith a disability, but not instudents without. With a UDL approach we want to reduce the construct–irrelevance variance. Building in flexible digital media options for all students to use when and how they want will help students with disabilities feel comfortable to use the tool that would best benefit them at the time. The true goal of assessment is to determine what the student understands about a concept, not how well he can read (decode) the test question.
Training and Support for Technology Systems Change
Technology use linked with the pedagogy of the classroom can produce remarkable results for learners, when they have access to, and training to use the technology tools available. In the past technology has been used as an accommodation and not impacted schools. (Rose & Mayer 2002; HehirKatzman 2012; Katz, 2012).Fullen’s work describes the criteria for new learning systems to be engaging for teachers and students, easy to use, include ubiquitous access to technology and be steeped in real life situations (2013). Teachers of activators or change agents will be part of the solution. When in that role a teacher as activator has an effect size of .60 as opposed to the facilitator role with an effect size of .17 Exploring the relationship between technology to pedagogy and how this integration has to use change knowledge so that the focus is on whole system reform. Fullen also discusses how technology support needs to be timely and effective on all aspects: software, hardware, maintenance, electricity and conductivity. It was noted in a number of studies that when students were provided training and ongoing support in using bimodal software, the results were positive in the students’ satisfaction and feeling of success and independence (Maurer, 2006; Montali & Lewandowski, 1996; McCulley, 2012; Hecker et al., 2002; Weber & Cavanaugh, 2006). In contrast, in the study by Hsin –Yu Chang (2010) , there was not a growth shown in compensatory attributes ashad been proven inthe other studies with longer timelines.Their study with only a two-week span showed a low motivation to participate by teachers and students.“It is reasonable to conclude thatwithout intensive intervention there’s no significant difference on psychological variables, which might have needed time to form the changes such as the participants learning adjustments and academic self-perceptions after using bimodal software”(Chiang et al., 2010, p. 478).
Kurzweil Software Usage in a Variety of Settings Analysis
Some educators may have worries that a text reader such as Kurzweil 3000 might become a crutch to prevent struggling students from learning to read and write on their own.Research findings show quite the opposite. Assistive technology like Kurzweil 3000, for example, not only allows struggling students to read grade level material, it also helps to improve word recognition and decoding. The standards must be kept high, but the way the students achieve these goals may vary. In the Boston Arts Academy in Boston, MA, students are allowed to use screen readers or audiobooks to access books that may be well above their current reading level. “We asked an English teacher if dyslexic students were allowed to use screen readers. Yes they are. All dyslexic students? No, I mean all students. They can all use Kurzweil. I don’t care how they read Shakespeare, I only care that they read Shakespeare” (HehirKatzman, 2012, p.69). As students using the program become more active readers, they are developing the skills that support independent reading without the assistive technology. The literature reviewed included studies from the following populations: students with dyslexia and other reading disabilities (Hecker et al., 2002; Elkind, 1998;McCulley 2012; Montali & Lewandowski, 1996; Fletcher et al., 2006; Floyd et al., 2012; Chiang et al., 2010; Maurer, 2006); students with attention deficit disorder (Hecker et al., 2002), and students who were gifted (Weber et al., 2006). These learners ranged in age from Gr. 3 to adults, in settings from public elementary and secondary schools to college and prison. In an analysis of data, themes emerge in 3 areas: reading speed, comprehension, and attitudes towards reading and ability.
Reading Speed. In the ability to use the software as a text reader, all studies reviewed showed that most students were able to read faster and for longer periods of time using the software. In the case study of one female sophomore student at Indiana University, she found that using Kurzweil helped her to participate fully in class and her reading speed with and without the software has improved.
An unintended side effect of Kurzweil is that my reading speed has increased. After I passed the initial learning curve of navigating around the program I was able to increase the Reading speed from 170 to 230 words per minute. This has allowed me to finish reading assignments with more time to devote to other school projects. Additionally, when I am reading text that is not on Kurzweil, my eyes have been trained to read at a faster pace and I’m proud tosay my reading speed without Kurzweil has also increased (Kanitkar, p. 650).
In the Elkin study (1998) pre-test results showed reading rate was clearly a major problem for the participants. He found unaided reading rate is a very good predictor of who will benefit in terms of increased reading speed using text to speech technology. In the Iowa study 2006 -2007, notable findings were that students could access twice the amount of text in the same amount of time (Maurer, 2006). In all of the studies where students were at a post secondary level, the benefits of bimodal reading help to support the vast amounts of content material that has to be read was clear (Chiang et al., 2010; Floyd et al., 2012; Hecker et al., 2002; Kanitkaret al).
Comprehension. Studies show that when information is presented through visual and auditory channels simultaneously the speed of processing memory recall are enhanced. Less skilled readers comprehended more with bimodal versus unimodal presentations (Montali & Lewandowski, 1996; Hecker et al., 2002; Elkind, 1998; McCulley 2012; Fletcher et al., 2006; Floyd et al., 2012; Chiang et al., 2010; Maurer, 2006). In a study with gifted learners by Weber (2006) , research on gifted students with reading disabilities shows comprehensionimprovement when text-to-speech software was used. It is important to note that learningdisabilities in the population of gifted students are 10 to 15%. “Larger font size and spacing actually cause the eyes to move more slowly while reading, allowing students to track their reading more easily and giving them more processing time.” (Weber & Cavanaugh, 2006, p. 61). The characteristics of students whose results showed a comprehension improvement show they had a poor unaided reading rate, good oral language capabilities and a good ability to integrate auditory and visual information.