Taking Notes Together: Augmenting Note Taking

Taking Notes Together: Augmenting Note Taking Laurian C. Vega, Margaret Dickey-Kurdziolek, Lauren Shupp, Manuel A. Pérez-Quiñones, John Booker, Ben Congleton Center for Human Computer Interaction and Department of Computer Science Virginia Polytechnic Institute and State University (Virginia Tech) {lhobby, mdickey, lshupp, perez, jobooker, bc}@cs.vt.edu ABSTRACT Sufficient tools for students with Learning Disabilities and Attention Deficit Disorder have not yet been established. We believe that the current tools these students can use call for a drastic change in traditional learning paradigms by either the instructor or the pupil. To fill this gap, we propose our tool, Taking Notes Together (TNT), as a collaborative note taking tool that will help in equalizing the classroom for students with disabilities. This tool allows students to collaboratively tag classroom lecture/discussion in real time through synchronized transcription and audio recording. TNT provides a visualization that highlights the important classroom points and we argue facilitates better recall and a deeper understanding of the classroom material. Through our evaluation we were able to prove that all students can benefit from this tool. We also present a case study of one student with ADD and how they benefited. The tool makes the learning experience, particularly for students with special needs like LD and ADD, less stressful while still being active in the notetaking. KEYWORDS: Collaborative Learning, Collaborative Note Taking, Attention Deficit Disorder, Learning Disorders, Tagging

1. INTRODUCTION Supporting students with Learning Disabilities (LD) and with Attention Deficit Disorder with Hyperactivity (ADD/ADHD) can be difficult. Proper teaching techniques should be used to support these specialized needs in order to minimize the time spent in the classroom and for later recall. In this paper, we discuss the need for a tool that facilitates the collaborative tagging of transcribed audio – one that will promote equity in the classroom. We argue that our proposed multi-media tool, Taking Notes Together (TNT), overcomes the shortcomings of traditional paper-based notes, stand-alone audio recordings, and transcriptions. We make the case that TNT supports a natural learning method that LD/ADD students will be able to use coherently with their lectures and class-based education.

Also, by allowing students to take advantage of the growing use of technology within the classroom TNT can support students in a collaborative, faster, and more automated process. Current research has explored many technologies to enhance traditional paper-based note-taking either by transcription or digital ink [8, 16, 19]. Yet, much of the research community is reluctant to step outside of the paper-based paradigm of taking notes. Paper or ink notes alone are insufficient because users often miss important information and participate less in discussion [10]. Some research has examined audio as data, either alone or in conjunction with digital ink notes [8, 16, 18, 19]. Nevertheless, the focus is usually on a single note-taker. Davis et. al. recognized the need for collaborative notetaking, where different notes can be merged [8]. Their NotePals tool seemed to do it all by integrating ink notes with audio across multiple users. However, their observations revealed the desire to tag important information: "There was also unanticipated usage. Several researchers independently used the device as an audio editor to isolate key verbatim quotes" [10]. This is where TNT combines the advantages of audio as data, collaborative efforts, and transcription. TNT hopes to alleviate users of the need to manually record audio data via transcription and audio recording. The tool also promises quick access to the chronologically ordered information by collaborative tagging. In the rest of this paper we address a significant amount of background work done with LD/ADD students and in general on note taking collaboration. We then talk about a tool we propose will help all students, in particular LD/ADD student, and discuss our experiment with results.

2. RELATED WORK 2.1 Background on LD/ADD ADD/ADHD and LD have been and are still a growing concern for educational institutions. This is because between three to five percent of the student population has ADD. Of these students, ADD is linked to an additional learning disorder by thirty-three percent [1, 2].

Thus, tailoring an educational program to the special needs of this significant minority while still maintaining a structured educational method for the rest of the student population is important [15]. For memory we look at the Encoding Specificity principle to help with the in class recall: “specific encoding operations performed on what is perceived determine what is stored and what is stored determines what retrieval cues are effective in providing access to what is stored“ [7]. This means that a student is going to develop various cues when learning to be able to recall this information at a later time. Thus, technology could help by having a way for students to work with these predefined cues. However, technology should not be simply thrown at the problem [17]. Instead, careful consideration of what the classroom needs are and how to support them should be evaluated. Specifically, Bramer found after evaluating seven college students that self perception of his or her disorder had a large effect on the overall success of the student [6]. Also, in a lecture based classroom ADD/LD students find that they cannot keep up with the incoming information – due to a range of possible disabilities – and thus miss the mental and physical encoding of the class material. Later, when the student goes back to review the lecture notes, the student is unable to comprehend the disorganized annotations due to recall problems [9]. An article by Joseph Boyle and Mary Weishaar showed that LD students were able to effectively recall and comprehend lecture information by having students take notes with a structured note taking system [5]. As a result, the students’ test grades increased. The authors went on to argue that the two current ways for LD students to take notes in the classroom (using note takers and having the teacher change methodology) were ineffective. However, the structured note taking system has had limited results. While disappointing, this result is encouraging for trying a new method for teaching students with special needs. In summary, during a traditional lecture-based class LD/ADD students can have serious educational difficulties. Past methods have had limited success and can cause passive learning. Using a tool that engages the students and allows the student to learn in a more natural way will be successful for helping LD/ADD students.

2.2 Note Taking We looked at memory-aids tools [4, 12, 13], tools for people with disabilities [20], memory tools [4, 12, 13, 20], and transcription tools [11], to help design TNT. In one article by Wu, Baecker, et al., they describe the codevelopment of a tool to help those suffering with chronic

amnesia called “OrientingTool”. The developers chose the participatory design paradigm because they felt, “it can be extremely difficult for designers to imagine the experience of coping with a cognitive impairment, resulting in a gulf of understanding.” When evaluated, the OrientingTool proved to be very useful for amnesiacs and their caretakers. For TNT, we collaborated with the Assistive Technologies Lab to facilitate a better understanding of how to design for LD/ADD. In other related work, tools have been designed to facilitate audio recording and indexing. As one example, Kubala, Colbath, et al. designed “Rough and Ready”, a tool to record, transcribe, and catalog CNN newscasts [11]. The authors felt that, “speech is not valued today as an archival information’s source because it is impossible to efficiently locate information in large audio archives.” Their tool, called “Rough ‘N’ Ready,” used current transcription tools to produce text from newscasts and then pick out names, places, and other key words. They found that their participants could “understand the contents of a news broadcast from a small set of highly descriptive labels.” Building on this work, we also wanted to include a transcription and index. Current transcription technologies, however, do not produce transcriptions with enough reliability. We hope as transcription technologies improve, that they could be used with assistive tools like TNT. In a similar effort [3], the Classroom 2000 project explored the use of different classroom technologies to study whether ubiquitous tools enhanced teaching and learning. The project broke the experience into three phases: pre-production (e.g. teacher preparation), live recording (e.g. taking notes, group exercises), and postproduction (accessing notes). In the pre-production phase, teaching styles were categorized into presentation, public notes, private notes, and discussion; learning styles were categorized into verbatim recording, highlighting, and none. The project then evaluated three courses with different teaching styles. In the evaluation they observed a strong liking for their tool and the Web access to recorded class materials. In an article by Leggett, he said, “the design of a system will need to accommodate the needs of the ‘memory worker’, whether as an individual or part of a closed or open working group.” [12] Students certainly fit the definition of “memory workers”, and class lectures are rich with information that could be lost to students that do not have effective note-taking abilities. This is why we have chosen the classroom setting to be the launching point for our tool design. Dr. Bell, in his paper describing the CyberAll tool, predicted, “in ten years, systems should be able to recall every personal lifetime conversation.” [4] The development of tools for class

lecture recording and indexing seems like a good first step to designing systems to record and index information in any audio form leveraging collaboration.

This facilitates easy use of TNT. Students are able to see the collaborative tagging along the top of the recording.

3. Taking Notes Together (TNT) 3.1 Motivation Our goal is to create a multi-media tool that supports a more active and natural education for all students. We plan on doing this by creating a system that is noninterruptive to the lecture session. To do this we have created our tool, TNT. We found that people with LD/ADD can feel stressed because of differing abilities to do combined activities such as write and listen. Therefore, we incorporated collaborative tagging, so that students can actively listen and participate without manually transcribing information. It is very easy to imagine how a student could take ineffective notes: missing information, recording less important information, not recording more important information. By supporting a way for students to collaboratively determine what lecture parts are important, a student will not only be able to compare and contrast what others found important, but train themselves to pick up verbal weight themselves. TNT enables student to record information in such a way that students feel capable and not rushed.

3.2 Usage TNT starts by taking an audio recording of a classroom lecture with palm pilots. At the start of each lecture the students will use their palm pilots to log into our online tagging system. TNT then shows buttons that correspond to the categories of class lecture material (e.g. see Figure 1). As the lecture progresses, students tag material that is important by selecting one of these buttons. These timestamped tags are then uploaded to a server where they can be collected. During the class, students could also take notes; however, normal cognitive stress that is caused by taking notes is reduced because the student will be able to use TNT as an alternative. When the student needs to review the lecture, the audio file, the transcription, and tags are integrated together for easy review. At this point, students can see an aggregation of all class tags. In the future, students may benefit from seeing which tags were their own. TNT was developed using Flash with tagged XML. The transcript is tagged by discussion breaks, which are visually shown as paragraphs. These are about a minute or two in length. You can see a working version of TNT in Figure 2. When the audio is playing you can see the current position of the transcription highlighted in blue and indicated on the timeline with the gray horizontal bar.

Figure 1. The Online Tagging System

Hypothetically, the higher the bar the more students tagged at that time, indicating importance. We are currently considering aggregation formulas for students who are misbehaving in class and are tagging to cause confusion with the visualization. Also, supported in TNT is a searching tool. Here students can enter a word or phrase that they remember from class, and the audio and transcription will skip to that point. Lastly, the audio and transcription are tightly coupled. The student can click on any section of text in the transcription and the audio jumps to that portion of the recording. Likewise, the student can either click on the visualization or drag-anddrop the audio marker to the desired section and the transcription jumps to that part as well. This supports students jumping around in TNT to the tagged sections. Eventually, the interface will be developed to support a note taking schemas such as the Cornel Note-taking system [14]. This is where the left hand, one-third of the interface is open for notes that the student might want to take, while the transcription is placed on the right twothirds of the page. This way the students can take subsequent notes using TNT, and this information would be logged for later viewing.

4. EXPERIMENTAL METHODOLOGY This study was designed to generate both qualitative and quantitative results about the usefulness and usability of this multi-media tool. In designing our study we realized that it would be very difficult to find and recruit a significant number of students with LD/ADD that were all within one class. Because of this, our study consisted of two phases: a controlled study using an in-classroom

setting, a case-study with a participating with ADD from said classroom, and an expert cognitive walk-through of TNT. The user evaluation consisted of recruiting a number of participants from a Virginia Tech class to evaluate our tool with a variety of metrics. One of the participants had ADD, and we describe his/her experience with TNT in greater detail as a case-study. We then asked experts from the VTAssist lab to do perform a cognitive walk-through of our tool. This study had IRB approval at Virginia Tech.

4.2

Procedure

One lecture from a Virginia Tech Business Information Technology 2406 Quantitative Methods class was audiorecorded. The following is the class description:

4.1 Materials

“Study of quantitative techniques used in managerial decision-making. BIT 2405: Data collection, descriptive statistics, probability theory, and statistical inferential procedures. BIT 2406: Linear regression and correlation analysis, forecasting, mathematical modeling, and network models.”

Each participant in the user evaluation used a Windows XP desktop machine to fill out the online questionnaires and class quiz. Participants in two experimental conditions were allowed to use any notes they had taken to complete the class quiz. Participants in one other condition were only allowed to refer to TNT to complete

This is a lecture based class taught using chalkboard, transparencies, handouts, quizzes, tests, and online material. The recorded class was on sensitivity analysis, which determines how much of a product you need to generate a margin of profit. Students were not informed of the purpose of the recording so that note taking would be normal. In addition, a handout was provided to the

the class quiz.

students based off of the projector transparencies used.

Figure 2: A screen shot of TNT. The audio timeline with a histogram of tags is at the top. Users can click on the timeline or transcripit on to activate audio playback. The timeline is tagged by students as either Example, Important, Test/Homework, and Software. The transcript is divided into sections of about one to two minutes.

The material from the lecture was transcribed and tagged by the experimenters. Experiments were scheduled for two weeks after the recorded lecture and during this time the students had not had any tests. Each study participant was given approximately an hour to complete the experiment. Before beginning the experiment, participants were asked to complete a consent form and a demographic questionnaire. Participants then had the remainder of the time to answer the online class quiz. At the end, participants completed a summative evaluation based on their experience with TNT.

Factor

Experimental Design

The independent three-level variable tested is note aid (see Table 1). We did not run a “No Notes” condition because sufficient research proves that notes enhance performance. We also did not test student performance with only the transcript because we wanted to evaluate the existing indexing method (student notes) against TNT. In the “Notes Only” condition students were only allowed to have access to their notes and handouts. In the “TNT Only” condition, students were allowed access to TNT. In the third condition, students were allowed access to their notes, handouts, and TNT. The dependant variables measured were time and accuracy. The accuracy was measured by the quiz scores. The quiz questions were designed to test the student’s understanding and memory of the lecture material. Seven of the nine questions were recall questions based on the lecture in the recording. The remaining two questions were prepared by the Professor based off of material from that lecture. All questions had exactly one correct answer. The order of the questions was the same for each participant, following the traditional test design in academia. Participants were given the quiz in its entirety, giving them the freedom to answer questions in any comfortable order. Our hypothesis for this study was that students with TNT support would perform better. To set up this study for later statistical testing we set our null hypothesis to be that there would be no difference between the three groups.

4.4

Table 1: The factor and its levels tested in this experiment.

Demographics

For each condition we used nine participants (total=27). This was approximately half of the people enrolled within the class. There were six 19-year-old students, sixteen 20-year-old students, four 21-year-old students, and one 22-year-old student. There were eighteen sophomores, seven juniors, and two seniors. We had one ADD student, one student with color blindness, and one student with a

Note aid

Level

{

Notes Only TNT Only TNT + Notes

5. USER EVALUATION RESULTS 5.1 Quantitative 5.1.1 Quiz Results Accuracy and time to completion were measured. For accuracy a 2-way ANOVA was performed on own notes and use of TNT. Analysis of variance showed a main effect for note aid (F(2,16)=64.14, p<.001) proving that the note aid affects accuracy. Tukey’s HSD post-hoc analysis shows the “Notes Only” condition was significantly less accurate than the others (p<.001). “TNT Only” and “TNT + Notes” were not statistically different (see Figure 1). This clearly shows that having a multimedia tool helps students re-access and re-find information. We believe “TNT + Notes” performed less accurately than “TNT Only,” although not significant, because students rely on their personal notes to answer questions. Accuracy

Correct Answers (%)

4.3

hearing impairment. (These were self reported.) There were eleven student note reorganizers and sixteen said that they never reorganized. When asked what methods they used to recover class information, one responded that (s)he asks for clarification in class, two responded that they use the book, and the remaining responded that they borrow notes.

100 80 60 40 20 0 Notes

TNT

TNT+Notes

Condition

Figure 3. Average accuracy for class quiz. Non-

overlapping error bars show statistical significance (sig. different conditions also linked by arrows). Time to Completion 40 30

nine students in the “Notes Only” condition elected to not answer. Six of the nine students from the “TNT Only” condition described sensitivity analysis, its use. Seven of the nine students from the “TNT + Notes” condition described the outcomes of sensitivity analysis and how they appreciated the professor given example.

For time to completion, we performed a 1-way ANOVA finding no effect for note aid. Although not significant, we observed students in “No Aid” group skipping or guessing the questions they knew they could not answer. This is reflected in the accuracy results (see Figure 3).

Almost all of the participants in the “Notes Only” and “TNT + Notes” conditions had trouble locating the notes from this particular lecture. Many had saved the handout and had annotations. Only one of the “Notes Only” participants had notes they had taken from the book.

5.1.2 Summative Evaluation

The participants in the “TNT Only” and “TNT + Notes” conditions usually spent some initial time exploring the software. Some participants would use the histogram of tags at the top of the screen to locate important pieces of the lecture while several other participants used the “Find Next” tool to seek to specific words. Ten of the eighteen participants in both TNT conditions seemed to prefer reading the text in the transcript without listening to the audio. On the other hand, three of the participants seemed to prefer the other extreme of just listening to the audio. One participant in particular let the audio play while they looked through the entire class quiz, filling in answers as they heard them. Five of the eighteen participants seemed to prefer reading the transcript while listening to the corresponding audio. One participant even used her finger to trace the words on the screen while the audio for that section of the lecture was being played. Six of the nine participants in the “TNT + Notes” condition referred to their notes at some point in answering the class quiz, while the remaining three did not refer to their notes at all.

Minutes

Figure 4. Average time to completion for the class quiz

During the experiments, we also observed and recorded how participants made use of their materials (e.g. notes & handouts, scrap paper, TNT). In the two conditions where students had their notes & handouts available to them, we recorded how they referred to them as well as any other interesting comments students made. In the two conditions where students had TNT available to them, we recorded which functionalities of TNT the participants were using and whether we observed any patterns in how the quiz.

20 10 0 Notes

TNT

TNT+Notes

Condition

We asked students to complete a short questionnaire to evaluate TNT. The questionnaire consisted of eleven general usability questions using a Likert-scale: eight tool navigation strategies questions, and three questions gauging student interest in collaborative tagging. We found very few differences between students using only TNT and students using TNT and their own notes. Not surprisingly, students who only used TNT were more likely to find the audio useful (3.75/5) when completing the quiz than students who used TNT and their notes (2.44/5) (p = 0.006). Additionally, students who only used TNT (4.28/5) were more likely to desire the ability to tag content from the lecture during class than students using TNT and their notes (3.77/5) (p = 0.092). Both groups of students primarily used the find box (11/16), but almost equally choose the tagged graph (6/16) and transcripts (7/16) as their secondary navigation technique. This was not a surprise; however, future work should make use of less reliable computer generated transcripts to more accurately evaluate the usefulness of our tool as an automated memory aid. In general, students found TNT to be useful, interesting, stimulating, and easy to use. They felt the navigation and data organization was intuitive, and felt that they would use TNT to study if it was available for their classes.

5.2 Qualitative In addition, we also asked our participants to list the three most important lecture topics discussed. Three out of the

5.3 Participant Case-Study One of our participants had ADD, and was randomly assigned to the “TNT Only” condition. This participant was of senior standing and twenty-two years old. When asked what types of things the participant took notes on, the student said that (s)he tried to note what was not in the book. The participant also liked to organize their notes in outline form. If the participant ever had to miss a class, he/she would often ask to borrow the notes of another classmate.

To complete the class quiz, the participant decided to listen and read the entire lecture. While listening to the lecture, the participant would take notes on paper. Occasionally the participant would back-track and listen to a section again. Once (s)he had finished reviewing the lecture, the participant started the quiz. During the quiz the participant referred only to taken notes and did not use TNT. The score was perfect. On the summative evaluation, the participant said that the transcript of the lecture was the most useful feature of TNT and that the audio of the lecture was the second most useful feature. (S)He also expressed the desire to create tags to use in TNT such as “test question” or “not in text book.” This is a very interesting request. Generic tags, like keywords, help for finding generic materials. However, as with any collaborative technology, the need to personalize within the context of action is also important. This might be particularly crucial for students with special needs and different learning schemas. Therefore, personalized tagging of audio material is a key need for TNT. When asked for any additional comments about TNT the participant said, “I really liked how it had both audio and visual notes for the class. It was very easy to take notes and to organize them as well.”

5.4 Expert Cognitive-Walkthrough In order to validate our tool with an expert, we conducted a cognitive walkthrough with two administrators of the Assistive Technology Lab. The qualitative measures we were looking for were intuitiveness and overall user satisfaction. We found that our tool was a good start on adding collaborative aspects to note taking tools. There are tools like OneNote, NoteTaker, and NoteShare which help with taking notes. However, as was pointed out earlier, simple note taking as supported by these technologies can be an overly stressful process that disconnects the student from active learning. Therefore, a tool that allows the student to go back and click through the audio recording with tagged information would be extremely useful. There were three main improvements our experts thought should be handled by our tool: highlight words or sentences instead whole paragraphs, show interval times, and have the importance bars be textured to help students who may be colorblind. There were also many suggestions for future work including but not limited to the following: a summary of what the individual student tagged at the bottom of the screen, downloadable notes for offline use, a synthesized voice option with variable speeds, an option to resize the transcript, and a text pad on the left of the transcript to take notes in. Lastly, our experts brainstormed a list of generic categories for the tagging and came up with the following: vocabulary,

critical keywords, important points made by the instructor, points to come back to, critical dates, and a classroom specific tag (i.e. in Computer Science courses: syntax, in Math courses: theorem). Overall this walkthrough showed that we have made significant progress in developing this tool, but that there are many other aspects we could add to make it better.

6. CONCLUSION These results show us that the null hypothesis, TNT conditions will perform equally as well as the “Notes Only” condition, can be discarded. We have shown that students will perform more accurately on a class quiz derived from lecture material using TNT than those students using only their own notes. The statistical difference between the “Notes Only” and both TNT conditions shows that multi-media, note-taking and retrieval tools such as ours can be a more effective way of retrieving lecture material than traditional note taking practices. However, we realize that with such a small sample size future studies will have to repeat this experiment to confirm our findings. Potentially, students will no longer have to rely on incomplete, tedious note taking; rather, one can tag important events and reaccessing any of the indexed information with greater ease. We feel that this has implications on the way educational tools should be developed in the future, specifically web based tools. Currently, many teachers in all ranges of education post some sort of lecture material to the Web for their students. Imagine if teachers could post the audio from the lecture with the class materials to the Web, and have it dynamically indexed in such a way that it was easy for students to located specific information and review the content. Although we only had one ADD student participate in our study, we were able to see that TNT was effective for this student. We saw some students strictly use the audio component of TNT, while others strictly used the text, and the rest of the students were some where in between. Students with ADD/LD need varying modes to access class information, we feel that multi-media tools such as TNT will serve as a “first step” to developing future educational technologies for students with special educational needs.

7. FUTURE WORK Whether using a tool like TNT or some other method, finding a way to make the classroom an equal learning place for all students should be and is an important focus for many education systems. However, many secondary schools and universities employ a broad range of

educational software packages with few aimed towards students with learning disabilities or collaborative student involvement. Research should continue to investigate assistive note-taking tools to improve upon our design and evaluation of multi-media tools like TNT. Specifically, we hope to see what implications a collaborative tool like TNT will have on the classroom and what visualizations and features are best suited for audio data. For instance, how do you provide a visualization that highlights what you found important but also one that highlights what other students found important? Would filtering the tags improve information retrieval? Also, what are the social implications of collaborative note taking? Would a tool like TNT facilitate social loafing? How could we discourage this behavior? Furthermore, we see this as a tool for students to potentially become more closely linked with the professor. How could professors use this tool to understand what topics their students are struggling with? The development of tools such as TNT opens up a host of new questions and research areas to be discovered and explained.

8. ACKNOWLEDGEMENTS We would like to thank the Virginia Tech Assistive Technology Lab who helped us with the development of this project.

REFERENCES [1]. NIMH research on treatment for attention deficit hyperactivity disorder: The Multimodal Treatment StudyQuestions and answers. 2000, National Institute of Mental Health. [2]. Identifying and treating attention deficit hyperactivity disorder: A resource for school and home. 2003, U.S. Department of Education. [3]. Gregory D. Abowd, Christoper G. Atkeson, Cindy Hmelo, Ami Feinstein, Rob Kooper, Sue Long, Nitin "Nick" Swahney and Mikiya Tani. Teaching and Learning as Multimedia Authoring: The Classroom 2000 Project, in Multimedia '96. 1996, ACM: Boston, Massachusetts. p. 187-198. [4]. Gordon Bell, A Personal Digital Store. Commun. ACM, 2001. 44(1): p. 86-91. [5]. Joseph R. Boyle and Mary Weishaar, The Effect of Strategic Notetaking on the Recall and Comprehension of Lecture Information for High School Students with Learning Disabilities. Learning Disabilities Research & Practice, 2001. 16(3): p. 133-141. [6]. Jennifer Bramer, Succeeding in College With Attention Deficit Disorders: Issues and Strategies for Students, Counselors and Educators. 1996. Plantation, FL: Specialty Press. [7]. Stuart K. Card, Thomas P. Moran and Allan Newell, Psychology of Human-Computer Interaction. 1983. Hillsdale, New Jersey: Lawrence Earlbaum Associates, Inc.

[8]. Richard C. Davis, James A. Landay, Victor Chen, Jonathan Huang, Rebecca B. Lee, Frances C. Li, James Lin, Charles B. Morrey, III, Ben Schleimer, Morgan N. Price and Bill N. Schilit. NotePals: lightweight note sharing by the group, for the group, in Proceedings of the SIGCHI conference on Human factors in computing systems: the CHI is the limit. 1999, ACM Press: Pittsburgh, Pennsylvania, United States. [9]. C. A. Z. Dendy and A. Zeigler, Teaching Teens with ADD and ADHD: A Quick Reference Guide for Teachers and Parents. 2003. Bethesda: Woodbine House. [10]. Daniel P.W. Ellis and Keansub Lee. Minimal-impact audio-based personal archives, in International Multimedia Conference: Continuous archival and retrieval of personal experiences. 2004, ACM Press: New York, New York, USA. p. 39-47. [11]. Francis Kubala, Sean Colbath, Daben Liu, Amit Srivastava and John Makhoul, Integrated Technologies for Indexing Spoken Language. Communications of the ACM, 2000. 43(2). [12]. Mike Leggett. Losers and Finders: Indexing Audio-Visual Digital Media, in C&C. 2005, ACM Press: London, United Kingdom. p. 210-217. [13]. Thomas P. Moran, Leysia Palen, Steve Harrison, Patrick Chiu, Don Kimber, Scott Minneman, William van Melle and Polle Zellweger. "I'll Get That Off the Audio": A Case Study of Salvaging Multimedia Meeting Records, in CHI. 1997, ACM Press: Atlanta, GA. [14]. Francis P. Robinson, Effective Study. 1961. New York: Harper & Row, Publishers. [15]. Edward J. Sabornie, Douglas Cullinan, Susan S. Osborne and Lynne B. Brock, Intellectual, Academic, and Behavioral Functioning of Students with High-Incidence Disabilities: A Cross-Categorical Meta-Analysis. Council for Exceptional Children, 2005. 72(1): p. 47-63. [16]. Lisa J. Stifelman. Augmenting real-world objects: a paper-based audio notebook, in Conference companion on Human factors in computing systems: common ground. 1996, ACM Press: Vancouver, British Columbia, Canada. [17]. Nigel Ward and Hajime Tatsukawa. Software for Taking Notes in Class, in IEEE Frontiers in Education Conference. 2003, ASEE/IEEE: Boulder, CO. [18]. Steve Whittaker, Patrick Hyland and Myrtle Wiley. FILOCHAT: handwritten notes provide access to recorded conversations, in Proceedings of the SIGCHI conference on Human factors in computing systems: celebrating interdependence. 1994, ACM Press: Boston, Massachusetts, United States. [19]. Lynn D. Wilcox, Bill N. Schilit and Nitin Sawhney. Dynomite: a dynamically organized ink and audio notebook, in Proceedings of the SIGCHI conference on Human factors in computing systems. 1997, ACM Press: Atlanta, Georgia, United States. [20]. Mike Wu, Ron Baecker and Brian Richards. Participatory design of an orientation aid for amnesics, in Proceedings of the SIGCHI conference on Human factors in computing systems. 2005, ACM Press: Portland, Oregon, USA.