research paper on interactive whiteboard

• DOI: 10.21432/T2WW4J
• Corpus ID: 67068282

The Interactive Whiteboard: Uses, Benefits, and Challenges. A survey of 11,683 Students and 1,131 Teachers | Le tableau blanc interactif : usages, avantages et défis. Une enquête auprès de 11 683 élèves et 1131 enseignants

• T. Karsenti
• Published 31 January 2017
• Canadian Journal of Learning and Technology

Figures and Tables from this paper

23 Citations

Elementary teachers’ uses and perceptions of interactive whiteboards for instruction, the impact of interactive whiteboard use on teaching grammar to grade seven efl learners, student behaviour in technology integrated classrooms, the impact of interactive touchscreens on physics education in upper secondary school – a systematic literature review.

• Highly Influenced

INTERACTIVE WHITEBOARD AS A MEANS TO ENHANCE LEARNERS’ ENGAGEMENT IN LANGUAGE LEARNING

An exploratory study of student online collaborative writing with teacher metacognitive prompts, investigating the impact of interactive whiteboards in higher education. a case study., the nature of classroom environment for drawing in the acquisition of communication skills in early years education in kenya, teachers attitudes towards liquid crystal display (lcd) panel interactive board applications, constraints and performance gap analysis (cpga) of lecturers in utilizing smart boards for improving students’ engagement, motivation and learning outcomes (semlos), 43 references, 'the visual helps me understand the complicated things': pupil views of teaching and learning with interactive whiteboards, the interactive whiteboard: a literature survey, interactive whiteboards: real beauty or just "lipstick", the effects of interactive whiteboards (iwbs) on student performance and learning: a literature review, attitudes of students and teachers towards the use of interactive whiteboards in elementary and secondary school classrooms..

• Highly Influential

Understanding an Elementary School Teachers' Journey of Using Technology in the Classroom from Sand Table to Interactive Whiteboard.

The features of interactive whiteboards and their influence on learning, keeping in touch with learning: the use of an interactive whiteboard in the junior school, disparity in practice: diverse strategies among teachers implementing interactive whiteboards into teaching practice in two swedish primary schools, pedagogical strategies for using the interactive whiteboard to foster learner participation in school science, related papers.

Showing 1 through 3 of 0 Related Papers

The impact of interactive whiteboards on education

New citation alert added.

This alert has been successfully added and will be sent to:

You will be notified whenever a record that you have chosen has been cited.

To manage your alert preferences, click on the button below.

New Citation Alert!

Please log in to your account

Information & Contributors

Bibliometrics & citations.

• Mhlongo S Mbatha K Ramatsetse B Dlamini R (2023) Challenges, opportunities, and prospects of adopting and using smart digital technologies in learning environments: An iterative review Heliyon 10.1016/j.heliyon.2023.e16348 (e16348) Online publication date: May-2023 https://doi.org/10.1016/j.heliyon.2023.e16348
• Oulaich S (2020) Pedagogy in the Digital Age: Making Learning Effective and Engaging for Students Innovations in Smart Cities Applications Edition 3 10.1007/978-3-030-37629-1_14 (168-182) Online publication date: 1-Feb-2020 https://doi.org/10.1007/978-3-030-37629-1_14
• Oulaich S (2019) Pedagogy in the digital age Proceedings of the 4th International Conference on Smart City Applications 10.1145/3368756.3369058 (1-9) Online publication date: 2-Oct-2019 https://dl.acm.org/doi/10.1145/3368756.3369058
• Show More Cited By

Index Terms

Applied computing

Computer-assisted instruction

Computer-managed instruction

Distance learning

Recommendations

Embedding interactive whiteboards in teaching and learning: the process of change in pedagogic practice.

This paper draws on research carried out for the UK government during 2004---2006 to evaluate the impact of interactive whiteboards for teaching and learning in primary schools in England. Multilevel modelling showed positive gains in literacy, ...

A national digital library for undergraduate mathematics and science teacher preparation and professional development

The primary goal of the National Digital Library for Undergraduate Mat hematics and Science Teacher Preparation and Professional Development, funded through the NSF Division of Undergraduate Education National Science Digital Libraries Initiative, is to ...

Interactive Whiteboards: Worth the Investment?

T. McDonough Central School District, located in a small city in upstate New York, is looking to invest some of its budget in new resources for Hawkins Elementary School. The School Board is strongly considering equipping each classroom in the school ...

Information

Published in.

• General Chairs:

Institute for Infocomm Research, Singapore

University of Augsburg, Germany

• Program Chairs:

National University of Singapore

Hong Kong Polytechnic University

National Institute of Informatics, Japan

• National Science Foundation of China
• CCNU: Central China Normal University
• Daqian Vision: Daqian Vision
• Microsoft Research: Microsoft Research
• Beijing ACM SIGMM Chapter
• NEC: NEC Labs China

Association for Computing Machinery

New York, NY, United States

Publication History

Permissions, check for updates, author tags.

• instructional effectiveness
• interactive whiteboards
• language teaching and learning
• mathematics education
• science education
• teaching strategies and methods
• Research-article

Funding Sources

• Ministry of Science and Technology of the People's Republic of China
• Ministry of Education of the People's Republic of China
• Intel Corporation
• Daqian Vision
• Microsoft Research

Acceptance Rates

Contributors, other metrics, bibliometrics, article metrics.

• 7 Total Citations View Citations
• 663 Total Downloads
• Downloads (Last 12 months) 19
• Downloads (Last 6 weeks) 2
• Shi Y Yang H Zhang J Wang S Yang H (2019) The Effects of Interactive Whiteboard-Based Classroom Instruction on Students' Cognitive Learning Outcomes: A Meta-Analysis 2019 International Symposium on Educational Technology (ISET) 10.1109/ISET.2019.00065 (276-280) Online publication date: Jul-2019 https://doi.org/10.1109/ISET.2019.00065
• Shi Y Peng C Yang H MacLeod J (2018) Examining interactive whiteboard-based instruction on the academic self-efficacy, academic press and achievement of college students Open Learning: The Journal of Open, Distance and e-Learning 10.1080/02680513.2018.1454829 33 :2 (115-130) Online publication date: 23-Mar-2018 https://doi.org/10.1080/02680513.2018.1454829
• Clayphan A Martinez-Maldonado R Tomitsch M Atkinson S Kay J (2016) An In-the-Wild Study of Learning to Brainstorm: Comparing Cards, Tabletops and Wall Displays in the Classroom Interacting with Computers 10.1093/iwc/iww001 28 :6 (788-810) Online publication date: 3-Mar-2016 https://doi.org/10.1093/iwc/iww001
• Shi Y Yang H Wu D Zhu S (2015) Interactive Whiteboard-Based Instruction versus Lecture-Based instruction Proceedings of the 2015 International Conference of Educational Innovation through Technology 10.1109/EITT.2015.56 (231-235) Online publication date: 16-Oct-2015 https://dl.acm.org/doi/10.1109/EITT.2015.56

View Options

Login options.

Check if you have access through your login credentials or your institution to get full access on this article.

Full Access

View options.

View or Download as a PDF file.

View online with eReader .

Share this Publication link

Copying failed.

Share on social media

Affiliations, export citations.

• Please download or close your previous search result export first before starting a new bulk export. Preview is not available. By clicking download, a status dialog will open to start the export process. The process may take a few minutes but once it finishes a file will be downloadable from your browser. You may continue to browse the DL while the export process is in progress. Download
• Download citation
• Copy citation

We are preparing your search results for download ...

We will inform you here when the file is ready.

Your file of search results citations is now ready.

Your search export query has expired. Please try again.

The Use of Interactive Whiteboards in Education

Opportunities and Challenges

Cite this chapter

• Marissa Saville ,
• Kim Beswick &
• Rosemary Callingham  

Part of the book series: Bold Visions in Educational Research ((BVER))

1665 Accesses

1 Citations

Interactive white boards (IWBs) have been heralded by many as a valuable teaching tool offering innumerable opportunities for increasing student engagement and learning (Campbell & Kent, 2010; Glover, Miller, Averis, & Door, 2005). Although research clearly shows IWBs have the potential to transform the way in which teachers teach (Glover et al., 2005), this potential is not realised simply by their installation into a classroom setting.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Subscribe and save.

• Get 10 units per month
• Download Article/Chapter or eBook
• 1 Unit = 1 Article or 1 Chapter
• Cancel anytime
• Available as PDF
• Read on any device
• Instant download
• Own it forever

Tax calculation will be finalised at checkout

Purchases are for personal use only

Institutional subscriptions

Unable to display preview.  Download preview PDF.

Similar content being viewed by others

Interactive whiteboards in primary schools: a Vietnamese language arts classroom with a quasi-experimental study

Towards a better understanding of the potential of interactive whiteboards in stimulating mathematics learning.

Effects of interactive whiteboard training programs on teacher efficacy and student outcomes in kindergartens

Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50 (2), 179–211.

Article   Google Scholar  

Armstrong, V., Barnes, S., Sutherland, R., Curran, S., Mills, S., & Thompson, I. (2005). Collaborative research methodology for investigating teaching and learning: The use of interactive whiteboard technology. Educational Review , 57 , 455–467.

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory . Englewood Cliffs, NJ: Prentice Hall.

Google Scholar  

Baylor, A. L., & Ritchie, D. (2002). What factors facilitate teacher skills, teacher morale, and perceived student learning in technology-using classrooms? Computers & Education, 39 (4), 395–414.

Beauchamp, G. (2004) Teacher use of the interactive whiteboard in primary schools: Towards an effective transition framework. Technology, Pedagogy and Education, 13 (3), 327–348.

Bennett, S., & Lockyer, L. (2008). A study of teachers’ integration of interactive whiteboards into four Australian primary school classrooms. Learning, Media and Technology , 33 (4), 289–300.

Betcher, C., & Lee, M. (2009). The interactive whiteboard revolution: Teaching with IWBs . Melbourne, VIC: ACER.

Bretscher, N. (2009, January/February). Dynamic geometry software: The teacher’s role in facilitating instrumental genesis . Paper presented at the Sixth Congress of the European Society for Research in Mathematics Education, Lyon, France.

British Educational Communications and Technology Agency. (2007). Evaluation of the primary schools whiteboard expansion project . Manchester: Education and Social Research Institute, Manchester, Metropolitan University. Retrieved from http://downloads01.smarttech.com/media/research/international_research/uk/becta_executive_expansion_summary.pdf

Campbell, C. (2010). Interactive whiteboards and the first year experience: Integrating IWBs into pre-service teacher education. Australian Journal of Teacher Education, 35 (6), 68–76.

Campbell, C., & Kent, P. (2010). Using interactive whiteboards in pre-service teacher education: Examples from two Australian universities. In M. Thomas & A. Jones (Eds), Interactive whiteboards: An Australasian perspective. Australasian Journal of Educational Technology , 26 (Special issue, 4), 447–463. Retrieved from http://www.ascilite.org.au/ajet/ajet26/campbell.html

Chuttur, M. Y. (2009). Overview of the technology acceptance model: Origins, developments and future directions . Sprouts: Working Papers on Information Systems, 9, Article 37. Retrieved from http://sprouts.aisnet.org/9-37

Compeau, D. R., & Higgins, C. A. (1995). Computer self-efficacy: Development of a measure and initial test. MIS Quarterly, 19 (2), 189–211.

Cuthell, J. P. (2003). Interactive whiteboards: New tools, new pedagogies, new learning? Some views from practitioners . Retrieved from http://www.virtuallearning.org.uk/?s=interactive+whiteboards%3A+new+tools%2C+new+pedagogies%2C+new+learning%3F

Dishaw, M. T., & Strong, D. M. (1999). Extending the technology acceptance model with task-technology fit constructs. Information and Management, 36 (1), 9–21.

Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention, and behaviour: An introduction to theory and research . Reading, Mass: Addison-Wesley.

Friel, T., Britten, J., Compton, B., Peak, A., Schoch, K., & Kent VanTyle, W. (2009). Using pedagogical dialogue as a vehicle to encourage faculty technology use. Computers & Education, 53 (2), 300–307.

Futuresource. (2011). Interactive whiteboard market valued at $1.4bn according to new Futuresource report . Retrieved from http://www.widepr.com/press_release/15473/interactive_whiteboard_market_valued_at_1_4bn_according_to_new_futuresource_report.html

Gillen, J., Kleine Staarman, J., Littleton, K., Mercer, N., & Twiner, A. (2007). A ‘learning revolution’? Investigating pedagogic practice around interactive whiteboards in British primary classrooms. Learning, Media and Technology, 32 (3), 243–256.

Glover, D., & Miller, D. (2001). Running with technology: The pedagogic impact of the large-scale introduction of interactive whiteboards in one secondary school. Journal of Information Technology for Teacher Education, 10 (3), 257–277.

Glover, D., Miller, D. J., & Averis, D. (2004, July). Panacea or prop: The role of the interactive whiteboard in improving teaching effectiveness . Paper presented at the Tenth International Congress of Mathematics Education, Technical University of Denmark. Retrieved from http://www.icmeorganisers.dk/tsg15/Glover_et_al.pdf

Glover, D., Miller, D., Averis, D., & Door, V. (2005). The interactive whiteboard: A literature survey. Teaching, Pedagogy and Education, 14 (2) 155–170.

Goodwin, K. (2011). Engaging students in literacy learning with interactive whiteboards . Sydney, NSW: Primary English Teaching Association.

Greiffenhagen, C. (2000). A report into whiteboard technologies . (unpublished) Oxford, Computing Laboratory. Retrieved from ftp://ftp.comlab.ox.ac.uk/pub/Documents/techreports/TR-16-00.pdf

Guin, D., & Trouche, L. (1998). The complex process of converting tools into mathematical instruments: The case of calculators. International Journal of Computers for Mathematics Learning, 3 (3), 195–227.

Hammon, M., Reynolds, L., & Ingram, J. (2011). How and why do student teachers use ICT? Journal of Computer Assisted Learning, 27 (3), 191–203.

Harris, S. (2002). Innovative pedagogical practices using ICT in schools in England. Journal of Computer Assisted Learning, 18 (4), 449–458.

Hartson, R. (2003). Cognitive, physical, sensory, and functional affordances in interaction design. Behaviour & Information Technology, 22 (5), 315–338.

Hennessy, S. (2011). The role of digital artefacts on the interactive whiteboard in supporting classroom dialogue. Journal of Computer Assisted Learning , 27 (6), 463–489.

Jewitt, C. (2006). Technology, literacy and learning: A multimodal approach . London & NY: Routledge.

John, P. D., & Baggott La Velle, L. (2004). Devices and desires: Subject subcultures, pedagogical identity and the challenge of information and communications technology. Technology, Pedagogy and Education, 13 (3), 307–326.

Joint Information Systems Committee. (2009). Effective practice in a digital age: A guide to technology enhanced learning and teaching . Retrieved from http://www.jisc.ac.uk/elearningprogramme.aspx

Kennewell, S. (2005, September). Researching the influence of interactive presentation tools on teacher pedagogy . In the proceedings of the British Educational Research Association Annual Conference, University of Glamorgan, Wales. Retrieved from http://www.leeds.ac.uk/educol/documents/151717.doc

Kennewell, S. (2006, November). Reflections on the interactive whiteboard phenomenon: A synthesis of research from the UK . In Proceedings Australian Association for Research in Education Conference, Adelaide. Retrieved from http://www.aare.edu.au/06pap/ken06138.pdf

Knight, P., Pennant, J., & Piggott, J. (2004). What does it mean to ‘use the interactive whiteboard’ in the daily mathematics lesson? MicroMath, 20 (2), 14–16.

Legris, P., Ingham, J., & Collerette, P. (2003). Why do people use information technology? A critical review of the technology acceptance model. Information and Management, 40 (3), 191–204.

Levy, P. (2002). Interactive whiteboards in learning and teaching in two Sheffield schools: A developmental study (Department of Information Studies, University of Sheffield) . Retrieved from http://dis.shef.ac.uk/eirg/projects/wboards.htm

Mayo, N., Kajs, L., & Tanguma, J. (2005). Longitudinal study of technology training to prepare future teachers. Educational Research Quarterly, 29 (1), 3–15.

McCormick, R., & Scrimshaw, P. (2001). Information and communication technology, knowledge and pedagogy. Education, Communication and Information, 1 , 37–57.

McCoy, C. (2010). Perceived self-efficacy and technological proficiency in undergraduate college students. Computers & Education , 55 (4), 1614–1617.

McCoy, S., Galletta, D. F., & King, W. R. (2007). Applying TAM across cultures: The need for caution. European Journal of Information Systems, 16 (1), 81–90.

McKenzie, J. (1999). How teachers learn technology best . Bellingham, WA: FNO.

McKenzie, J. (2002). Just in time technology: Doing better with fewer . Bellingham, WA: FNO.

Miller, D. J., Glover, D., & Averis, D. (2004). Developing pedagogic skills for the use of the interactive whiteboard in mathematics . Retrieved from http://www.keele.ac.uk/media/keeleuniversity/fachumsocsci/sclpppp/education/interactivewhiteboard/BERA%20Paper%20Sep%202005.pdf

Prensky, M. (2005). Adopt and adapt: School technology in the 21st century. Edutopia . Retrieved from

Rosenfeld, B., & Martinez-Pons, M. (2005). Promoting classroom technology use. Quarterly Review of Distance Education, 6 (2), 145–184.

Serow, P., & Callingham, R. (2011). Levels of use of interactive whiteboard technology in the primary mathematics classroom. Technology, Pedagogy and Education, 20 (22), 161–173.

Slay, H., Siebörger, I., & Hodgkinson-Williams, C. (2008). Interactive whiteboards: Real beauty or just “lipstick”? Computers & Education, 51 (3), 1321–1341.

Smith, H. J., Higgins, S., Wall, K., Miller, J. (2005). Interactive whiteboards: Boon or bandwagon? A critical review of the literature. Journal of Computer Assisted Learning, 21 , 91–101.

Sweeney, T. (2008). Transforming learning with interactive whiteboards: Towards a developmental framework. Australian Educational Computing, 23 (2), 24–31.

Teo, T. (2008). Pre-service teachers’ attitudes towards computer use: A Singapore survey. Australasian Journal of Educational Technology, 24 (4), 413–424.

Teo, T. (2009). Modelling technology acceptance in education: A study of pre-service teachers. Computers & Education, 522 (2), 302–312.

Thompson, R. L., Higgins, C. A., & Howell, J. M. (1991). Personal computing: Toward a conceptual model of utilization. MIS Quarterly, 15 (1), 124–143.

Warren, C. (2003). Interactive whiteboards: An approach to an effective methodology. Computer Education , 103 , 11–12.

Warwick, P., Mercer, N., Kershner, R., & Kleine Staarman, J. (2010). In the mind and in the technology: The vicarious presence of the teacher in pupil’s learning of science in collaborative group activity at the interactive whiteboard. Computers & Education, 55 (1), 350–362. www.marcprensky.com/…/Prensky-Adopt_and_Adapt-Edutopia-01

Download references

You can also search for this author in PubMed   Google Scholar

Rights and permissions

Reprints and permissions

Copyright information

© 2014 Sense Publishers

About this chapter

Saville, M., Beswick, K., Callingham, R. (2014). The Use of Interactive Whiteboards in Education. In: The Future of Educational Research. Bold Visions in Educational Research. SensePublishers, Rotterdam. https://doi.org/10.1007/978-94-6209-512-0_17

Download citation

DOI : https://doi.org/10.1007/978-94-6209-512-0_17

Publisher Name : SensePublishers, Rotterdam

Online ISBN : 978-94-6209-512-0

eBook Packages : Humanities, Social Sciences and Law Education (R0)

Share this chapter

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Publish with us

Policies and ethics

• Find a journal
• Track your research

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Elsevier - PMC COVID-19 Collection

Using a digital whiteboard for student engagement in distance education

The COVID-19 pandemic transformed educational processes across different educational levels. As institutions and faculty members effort focused on guaranteeing academic continuity, the challenge was how to translate the learning methodologies applied in the classroom to virtual settings. A digital whiteboard was integrated to synchronous class sessions to complement the educational experience. During these sessions, students and teachers interacted to co-construct ideas and socialize learning. The objective of this study was to assess the impact of introducing a digital whiteboard in student engagement. The quantitative approach integrated student's perception through an online survey with 12 items. The results show that the students enjoyed the dynamic(4.56), students believe that the incorporation of digital whiteboard contributed to understanding abstract concepts(4.83), and perceived the resource contributed for class engagement(4.72). The design of educational projects that incorporate these resources translate to active learning dynamics which foster student engagement.

Graphical abstract

1. Introduction

By March 2020, SARS-CoV19, a novel Coronavirus first described on december 2019 in the Chinese province of Wuhan was declared as a pandemic. The global impact of this virus was inimaginable as global economic and social activity stopped, carrying out a radical change on the dynamics of daily life. Education of course was not an exception. Courses in different educational levels, from elementary to higher education had to be rapidly migrated to digital platforms, and teachers were forced to innovate and find new digital alternatives for educational dynamics. [1] . Particularly in higher education settings, the main challenge was to successfully provide the curricular content, skill development and a comparable educational experience for learners.

By April 2020, the Digital Flexible Model (MFD, by its initials in spanish), a proposal for distance education by Tecnologico de Monterrey University was developed. This model described the incorporation of digital tools such as Zoom for synchronous sessions and some educational technologies that allowed to recreate a similar learning experience for students during the pandemic [2] . The beginning of the MFD model constituted a challenge for the educational community, which quickly innovated with dynamic and alternatives for student engagement.

The first section of the paper describes the impact of innovation for teaching and learning, and its determining role for the challenge of the COVID-19 pandemic. The following section describes the introduction of the digital whiteboard to spark student engagement in distance education, and presents a method to assess its effects. Then the results section presents findings and analysis of students' perception of the educational experience. The discussion section presents arguments and views of the different factors that impact on student engagement. Finally, the conclusion section presents alternatives to incorporate educational innovation strategies to improve classroom dynamics besides the distance or remote format for learning.

1.1. Innovation in teaching and learning

Educational innovation is the application of an idea, methodology or process to produce a change in the educational experience [3] . These planned changes surpass the achievement of educational goals, and strive to integrate nobel proposals to nurture learning environments [4] . In higher education, models are transitioning away from the massive lecture halls where the students are passive subjects who receive the knowledge of the expert [5] . Educational institutions are focusing on active learning methodologies and innovations that prepare graduates for an uncertain future [6] . But it has also been adjusted to teaching a new generation of students, the Generation Z (Gen Z), which through a generational lens can be described as a generation that values diversity, are optimistic about their future and are highly persistent [7] .

As educational technologies reach new disciplines and develop new applications, these trends have gained acceptance and credibility, as well as an impact on the training process [5] . As new generations arrive into the university, there is a need to update and redesign course materials and methodologies, as well as to assess the contributions of traditional strategies [8] . This has encouraged educators to implement educational innovation projects that integrate technology; however, the question is the extent in which institutions, teachers and students are prepared to implement them [9] .

In medicine, although the prevailing way of teaching has been the see one, do one, teach one for surgical and clinical skills [10] , the educational processes have evolved to incorporate new strategies where the educator guides, facilitates and accompanies the teaching-learning process and students take their role as active learners [11] . Some trends in educational technology that have been incorporated are the use of augmented and virtual reality for teaching anatomy, online evaluation supported by feedback [12] and the use of mobile devices to trigger interaction and discussion.

1.2. Teaching in the COVID-19 pandemic

The challenges of teaching amid the pandemic have emphasized the importance of building on the lessons learned from the previous implementations of innovation projects, as well as in the creativity from teachers for a successful transition. Some of the dynamics that were threatened in educational virtual settings is the interactions of learning with peers and the experiences with faculty [13] . Enablers to foster these interactions are video conferencing tools such as Zoom, Google Meet, Youtube Live, Facebook Messenger, and Whatsapp Rooms. However there must be an intentional incorporation and design to leverage these advantages. These structural elements recreate the campus environment for content delivery; some designs have relied on Learning management systems (LMS) such as Canvas, Google Classroom, Blackboard, Edmodo, Moodle, to name a few. If the structure of these learning environments is well defined, it can favor engagement of the students through challenging tasks and clear guidelines for learning [14] . However, a poor instructional design can become an impeding or maladaptive cognition for engagement [15] .

At first, the temptation was transferring face-to-face classes to short online conferences where teachers would present a monologue of each topic, but it became a setback on the actual model in which students are actors responsible for their own learning process. The face-to-face interactions are a factor for nostalgia in many cases, a constant threat for learning amid the pandemic is behavioral disaffection [16] . Thus, courses sessions benefit for activities that target skill development and interactions between the members of the learning community designed [17] . The redesign underwent needed to create scenarios for active and collaborative learning, where students could experience the emotional and behavioral engagement to manage their own learning [5] .

Providing a comparable experience than the one recieved face-to-face requires the integration of resources which recreate powerful dynamics of students' active role. Although the teaching-learning process has been challenging for teachers and students alike, it has also contributed to visualize elements that under normal conditions would go unnoticed, such as building community and a safe-space for students to interact.

For the Immunology course, the usual dynamics consisted in developing diagrams to explain interactions between concepts and processes. The building process is guided by thought-provoking questions to engage in discussion and to help students to identify key concepts, the conclusions and synthesis, as well as these key points are represented on the whiteboard. This is one of the best valued formats by students that have previously participated in class, as they highlight the opportunity to co-construct ideas and socialize learning. To recreate this dynamic when migrating to the distance education model, an innovation was developed to address two elements: 1) the use of connectivity and distance education platforms to promote a dynamic and active class, and 2) promote collaboration between students and teachers. To do so a virtual whiteboard application was integrated to synchronous class sessions. During these sessions, students and teachers participated and discussed specific topics of the program. The objective of this study was to assess the impact of introducing a digital whiteboard in student engagement on a distance learning experience..

2.1. Methods

In order to assess this implementation, a quantitative approach was implemented [18] . The methodology described by the first level of Kirkpatrick model was incorporated as it focuses on the assessment of student satisfaction and reaction to innovation [19]. A 12-item questionnaire was used to assess different factors that impact student engagement: 1) interaction with peers and faculty [13] , 2) structure and educational environment [14] , 3) emotion and behavior [16] . Each factor explored different elements for engagements described in table 1 . The instrument incorporates 10 items using a five-point Likert scale where 1 corresponds to total disagreement, and 5 total agreement, and 2 additional open-ended questions to understand the engagement factors that students considered the most important in a face-to-face format, and in distance settings.

Instrument design for student engagement

The sample strategy was a convenience sample, since participants were volunteers [18] . It consisted of 39 fourth-semester medical students from the Immunology course which gave consent for the results to be used for educational research purposes.

This distance education model started implementation in early April 2020. In order to achieve an active class, the implementation required prior planning work in which the topics were agreed to be discussed in each session, clear rules of etiquette were established for interaction in the virtual course, as well as materials to be completed before class.

Regarding technical preparations, the teacher logged into the Zoom video conferencing tool on two different devices: computer and tablet. The purpose of the computer session is for the teacher to periodically review the Zoom chat with questions or comments that the students may have, to have an extra screen to corroborate the transmission of the class and to manage the waiting room of the Zoom session. The tablet was used to share the screen where the diagrams were being worked using the Goodnotes app. As the session progressed, the teacher used questions to guide the students' discussion. Together they built the graphic representation that included drawings or annotations. Sometimes screenshots of figures from a book, paper or videos were overlapped into the diagram in Goodnotes to complement the explanation. Altogether, this ensured the class remained interactive, favoring student's engagement and this digital whiteboard was the keystone on achieving it.

In this app, the key concepts and arrows that demonstrate the interaction between the various immunological or hematological processes are integrated. Fig. 1 presents an example of these sessions, it particularly depicts a sequence of events on the platelet activation process. First a table was made (upper left corner) comparing the main glycoproteins on the platelet's surface and their ligands. The bottom right corner showcases a diagram showing step by step platelet activation and involvement of these glycoproteins from adhesion to agregation.

Platelet activation

The construction of the diagram starts from the top left corner, continues towards the right side, and finishes at the bottom of the board. Colors complement the presentation of information in an organized way, helping students to achieve knowledge organisation. Students can take screenshots as the class progresses, but they can also access the diagrams through an online shared-folder where each class is documented.

The items that received the most favorable responses were: “10. I think my teacher showed great commitment making the transition to the distance education model”, “4. The use of graphic resources (whiteboard, drawings, mental maps, integration of text figures) helped me to understand abstract concepts that I find difficult to understand in books”, and “1. I enjoyed the methodology in which my class was taught in the face-to-face format”, with mean of 4.94, 4.83 and 4.8, and variances of 0.053, 0.31 and 0.33 respectively.

The items that received a less favorable evaluation correspond to the items of "7. Switching from the face-to-face diagram construction to a digital version of the whiteboard made it difficult for me to follow the course content.", "8. I felt more involved with the course in the distance course", and "10. I was more motivated to participate in the course in person”, with a mean of 2.3, 3 and 3.63, and variances of 2.22, 1.77 and 1.55 respectively. Table 2 presents the results obtained by each engagement factor.

Student engagement assessment in the implemented innovation

*Reverse scored items

In the open-ended questions, students identified that in face-to-face settings the most relevant engagement factors were 51.2% interaction with peers and faculty, 41.4% structure and educational environment factors, and 7.4% referred to emotion and behavior factors. In distance education settings , students described that the most relevant engagement factors were 0% interaction with peers and faculty, 78.1% structure and educational environment factors, and 21.9% declared emotion and behavior factors. Exemplary quotes of students reflections in open-ended questions are presented in Table 3 .

Students reflections in open-ended questions

4. Discussion

The innovations implemented in the course were focused in fostering interaction with peers and faculty . These adaptations presented a challenge because some of the strategies were not the best to promote interaction and engage students. Tools such as the digital whiteboard were considered useful because they helped to preserve some of the usual conditions and class dynamics. The results of this study show that students felt the class as if nothing had changed from the presence based interactions. That perception of quality was deeply valued since the migration to remote learning was done fast and efficiently given the short-time there was to plan and adapt contents to strategies that were already designed. However, it posed the question to consider if some of the teaching rituals are strictly necessary. Practices such as delivering paper-based assignments, organizing synchronous team-discussions, and long lectures, are to be replaced by the incorporation of some technologies. These additions could contribute to protecting class time for the important elements described above: describing examples of specific processes, discussion with peers, posing questions, and overall constructing knowledge with previous conceptions as learning takes place.

The structure and educational environment elements are shown in the results obtained demonstrate an adequate transition from the face-to-face model. One of the main strengths of this innovation refers to the successful migration to the digital model where the students expressed that the quality was comparable to the one they had in the face-to-face model. A lesson learned in this implementation was to assess the project by the students' voices, not as customers that needed to be satisfied but rather as partners that have to be interested in their learning process in order to succeed. For instance, the first configuration performed by the teacher received feedback from students which ended providing alternatives to make the setting up of the sesion easier. To achieve this, it is crucial that the teacher has presented clear objectives for the class, as these are discussed and clarified with the student there is an alignment of expectations on both stakeholders. One strategy is to ask students to read the course materials, before each class since it is key for students to have that previous knowledge to be able to interact. Although remote sessions were guided by the teacher with directed questions during this process, student preparation is important to the dynamic in a remote model.

A result of the assessment that was quickly recognized was the emotional and behavioral impact that the implementation was fulfilling. The COVID-19 pandemic has demanded an extra effort from teachers to deliver not only the excellence students are used to, but also to provide a bit of normality amid uncertainty and stress. To foster these learning environments educators need to continually assess their own performance to recognize the contribution of their teaching efforts, a skill that needs to be nurtured by faculty development programs even after the pandemic crisis has passed.

5. Conclusions

Students seem to respond well to active learning dynamics besides the distance. Some elements still require a continuous effort to impact on student engagement, for example in the resistance of participants towards opening their microphone to ask questions or share a comment, they still perceive that by making an oral contribution, it's an interruption of the teacher's explanation. There is a need to develop alternatives where all students can participate and engage in the most natural and effective way. This could be achieved by holding dedicated times for discussion, scheduling online forums or by making students work in small groups where they interact with their peers. Unfortunately, there is still not a way to emulate the totality of a face-to-face classroom and the live interaction within. However, these types of dynamics provide a sense of activeness and normality of the classes before the pandemic, with the elements that now make us nostalgic.

Traditionally, a large portion of teachers still limit themselves to conduct lecture-based sessions that are supported with Powerpoint presentations. This strategy is time consuming and could become tedious for Gen-Z students, who are used to receiving multiple stimuli and have shorter attention spans [7] . Implementing educational innovations allows that students stay engaged and active by collaborating in the sessions. Surely, this strategy requires that the teacher performs some additional work, due to the setup, planning and implementation. However, the results obtained make it worthwhile.

Author contributions CRediT roles

Aniela Mendez-Reguera: Investigation, Methodology, Writing – original draft, review & editing.

Mildred Lopez: Methodology; Formal analysis; Resources; Project administration; Roles/Writing - original draft, review and editing

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Biographies

Aniela Mendez-Reguera is currently the Associate Director of the Medical program at Tecnologico de Monterrey, School of Medicine and Health Sciences, where she participates as Faculty in the immunology and microbiology courses. She is an M.D. with a Ph.D in Immunology. She has been enjoying life as an educator for the last four-years. Her contributions have taken her to participate in several international conferences in medicine and educational technology. Her educational innovation projects are published in the leading journals on medical education.

Mildred Lopez is the author of more than 40 articles and 11 book chapters. Currently, she is the Director of Educational Innovation at Tecnologico de Monterrey, School of Medicine and Health Sciences. Phd in Educational Innovation. Fellow of medical education at FAIMER Institute, and of the Association of Medical Education Europe (AMEE). Member of the Latin American Federation of Clinical Simulation and Patient Safety (FLASIC), and the National Academy of Medical Education in Mexico. Founding member of the Healthy Living for Pandemic Event Protection (HL - PIVOT) Network.

Editor: Dr. M. Malek

This paper is for special section VSI-tei. Reviews processed and recommended for publication by Guest Editor Dr. Samira Hosseini.

Academia.edu no longer supports Internet Explorer.

To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to  upgrade your browser .

Enter the email address you signed up with and we'll email you a reset link.

• We're Hiring!
• Help Center

Student Engagement, Visual Learning and Technology: Can Interactive Whiteboards Help

Related Papers

ELIJAH L JOHN

Hserin Serin , Hamdi Serin

Recently there have been a growing number of researches on the influence of interactive whiteboard on engagement and achievement, and many of them have yielded positive results. Learner engagement has emerged from the connection between involvement and achievement and now quite many researchers agree on the correlation between learner engagement and learner achievement. This study aims to show whether the implementation of interactive whiteboard in Mathematics classes affect learner engagement and achievement. A questionnaire survey was conducted including 60 Mathematics department students at a private university in Iraq. The results indicated that the employment of interactive whiteboard impressively influences learners' engagement and achievement in Mathematics.

Recently technology-enhanced applications have become an increasingly important component of education. There is no doubt that the use of technology in education has yielded improvements in learners' achievement. Interactive whiteboard, regarded as one of the most effective educational tool, has the potential to revolutionize classroom instruction. Moreover, interactive whiteboard supports classroom management through motivating learners to participate in classroom activities. The use of interactive whiteboard enhances learners' engagement in the classroom which facilitates classroom management. This study focuses on the impacts interactive whiteboard makes on classroom management. A questionnaire including 100 participants was conducted and the study found that the use of interactive whiteboard in the classroom largely influences classroom management.

Human Systems Engineering and Design (IHSED2021) Future Trends and Applications

Omar Cóndor-Herrera

Nowadays, the teaching – learning process is characterized by the application and implementation of technological resources that look for the improvement of students´ attention and motivation within the educative process. In this context, one of those resources looking to accomplish this objective are digital whiteboards, that allow to make cooperative work, interact in audio and video in real time, closing the gap with the real world. The objective of this paper is to analyze the benefits that interactive whiteboards offer to the learning process, as well as to review previous studies that validate their application and the positive effects given to the educational process.

Journal of Educational Computing Research

Shu Ching Yang

In recent years, the interactive whiteboard (IWB) has been regarded as the most prominent information and communication technology auxiliary instruction device. It is touted as elevating the traditional teaching environment to a digital teaching environment because of its highly interactive features. The purpose of this study is to investigate elementary students’ perceptions of their teachers’ use of the different IWBs’ interactive functions in class and the effect of different IWB’s interactive uses on students’ learning attitudes. The findings showed that teachers’ use of the IWBs’ basic interactive function helps students develop positive learning attitudes, including enjoyment of learning, usefulness of IWB, and willingness to learn. However, such effects from the use of the IWB’s advanced interactive functions were not as clear, particularly the effects on usefulness of IWB and willingness to learn. Moreover, the students’ personal experience operating the IWB had an effect on...

R&E-SOURCE

Juraj Mistina

Modern electronic devices and teaching aids are constantly innovating education. Education has recently undergone many changes. Currently, the latest trend in the modernisation of teaching is represented by an interactive whiteboard. When used correctly, it represents a modern didactic tool that contributes to innovation and the efficiency of teaching a specific subject. This contribution aims to provide up-to-date information on using interactive whiteboards in secondary school teaching. In the paper, the authors describe the results of the conducted research. They focus on using the interactive whiteboard from the student’s point of view and the teacher’s. Using several research methods, the authors investigated the frequency of use of the interactive whiteboard and the learners’ opinions towards its use. They also investigated the possibilities of streamlining the teaching process.

Tạp chí Khoa học

Yehuda Peled , Mandy Medvin , Linda P Domanski

This research examines teacher attitudes and fears about interactive whiteboard (IWB) use as related to perceived classroom implementation to enhance student engagement and achievement. The research took place in four western PA school districts. Nearly 78 percent of all teachers surveyed reported using the IWB either "Often" or "All the time" and 75 percent of them reported using an IWB for two or more years. These combined data suggest that the districts in this study are investing in IWB technology and the majority of teachers are using IWBs on a regular basis for instruction. Number of years and frequency the IWB technology was used in the classroom was strongly related to levels of training and support the teachers believed they received, teachers' sense of self-efficacy and the perceived value of IWB technology as a useful tool, and teachers' perceptions about the positive effect that integrating IWBs had on student achievement.

keshav jadhav

Pooja Kumra

Globalization has permitted technical progress in communication field which enable users to access and exchange information at any time and from any place in the world. Technology has played a vital role in education field too. Learners of today are more familiar with technology and they are growing up in today's world which relies heavily on information technology. Information Technology (IT) has entered in teaching learning process too. When used effectively, IT can play a role in stimulating curiosity, interest and in facilitating and sustaining purposeful engagement in students. Moreover, technology can play a role in triggering and addressing personal, situational, and contextual factors that support autonomy, competence and enhance active and deep learning. Smart technologies are being used because they are said to " enhance learning " by increasing student " interest " through " active engagement ". Researchers have observed that Interactive White Board (IWB) or Smart Board technology affects all three forms of engagement (affective, participative, and cognitive). Since IWB's claims to increase student engagement, there is the possibility that it even increases student interests through the affective and participative domain. Moreover Interactive Whiteboard, regarded as one of the most effective educational tool, has the potential to revolutionize classroom instruction. IWB facilitates the teaching learning process and has the potential to provide a wide range of activities for learners. There is no doubt that use of technology in education has yielded improvements in learner's achievement, engagement, interest and retention. But on the other hand, Technology imposes few challenges too like incompetent and irrelevant e-content, lack of sufficient teacher's training, technical problems of hardware and software, lack of teacher's confidence and unavailability of proper technology.

Loading Preview

Sorry, preview is currently unavailable. You can download the paper by clicking the button above.

RELATED PAPERS

Tuyet Hayes

Omer Faruk Sozcu

YALIN KILIÇ TÜREL

From Kindergarten to Workplace Training

Damian Maher

Tufan Aytaç

Yalın TÜrel

Australasian Journal of Educational Technology

Jacqueline Tinkler

Information Technology, Education and Society, 12(2).

Philip Clarkson

Avances en Interacción Humano-Computadora

Vivian Jimenez Estrada

siti balqis

Dr. Saleena S Gil

Nihan Arslan Namlı , Mehmet Can ŞAHİN

Journal of Business and Social Review in Emerging Economies

ahmad shabdin

Maria Northcote

Australian Educational Computing

Sue Gregory

Halit Arslan

Proceedings of the 2019 8th International Conference on Educational and Information Technology (pp. 137-140). ACM.

Robert Stroud

maureen richards

Piroska Biró

British Journal of Educational Technology

Interdisciplinary Journal of e-Skills and Lifelong Learning

Rachel Zorman

International Journal of Emerging Technologies in Learning (ijet)

Brian Drayton

Universitas Tarraconensis. Revista de Ciències de l'Educació

Josep Holgado-García

•   We're Hiring!
•   Help Center
• Find new research papers in:
• Health Sciences
• Earth Sciences
• Cognitive Science
• Mathematics
• Computer Science
• Academia ©2024

When A.I.’s Output Is a Threat to A.I. Itself

As A.I.-generated data becomes harder to detect, it’s increasingly likely to be ingested by future A.I., leading to worse results.

By Aatish Bhatia

Aatish Bhatia interviewed A.I. researchers, studied research papers and fed an A.I. system its own output.

The internet is becoming awash in words and images generated by artificial intelligence.

Sam Altman, OpenAI’s chief executive, wrote in February that the company generated about 100 billion words per day — a million novels’ worth of text, every day, an unknown share of which finds its way onto the internet.

A.I.-generated text may show up as a restaurant review, a dating profile or a social media post. And it may show up as a news article, too: NewsGuard, a group that tracks online misinformation, recently identified over a thousand websites that churn out error-prone A.I.-generated news articles .

In reality, with no foolproof methods to detect this kind of content, much will simply remain undetected.

All this A.I.-generated information can make it harder for us to know what’s real. And it also poses a problem for A.I. companies. As they trawl the web for new data to train their next models on — an increasingly challenging task — they’re likely to ingest some of their own A.I.-generated content, creating an unintentional feedback loop in which what was once the output from one A.I. becomes the input for another.

In the long run, this cycle may pose a threat to A.I. itself. Research has shown that when generative A.I. is trained on a lot of its own output, it can get a lot worse.

Here’s a simple illustration of what happens when an A.I. system is trained on its own output, over and over again:

This is part of a data set of 60,000 handwritten digits.

When we trained an A.I. to mimic those digits, its output looked like this.

This new set was made by an A.I. trained on the previous A.I.-generated digits. What happens if this process continues?

After 20 generations of training new A.I.s on their predecessors’ output, the digits blur and start to erode.

After 30 generations, they converge into a single shape.

While this is a simplified example, it illustrates a problem on the horizon.

Imagine a medical-advice chatbot that lists fewer diseases that match your symptoms, because it was trained on a narrower spectrum of medical knowledge generated by previous chatbots. Or an A.I. history tutor that ingests A.I.-generated propaganda and can no longer separate fact from fiction.

Just as a copy of a copy can drift away from the original, when generative A.I. is trained on its own content, its output can also drift away from reality, growing further apart from the original data that it was intended to imitate.

In a paper published last month in the journal Nature, a group of researchers in Britain and Canada showed how this process results in a narrower range of A.I. output over time — an early stage of what they called “model collapse.”

The eroding digits we just saw show this collapse. When untethered from human input, the A.I. output dropped in quality (the digits became blurry) and in diversity (they grew similar).

How an A.I. that draws digits “collapses” after being trained on its own output

If only some of the training data were A.I.-generated, the decline would be slower or more subtle. But it would still occur, researchers say, unless the synthetic data was complemented with a lot of new, real data.

Degenerative A.I.

In one example, the researchers trained a large language model on its own sentences over and over again, asking it to complete the same prompt after each round.

When they asked the A.I. to complete a sentence that started with “To cook a turkey for Thanksgiving, you…,” at first, it responded like this:

Initial A.I. output

Even at the outset, the A.I. “hallucinates.” But when the researchers further trained it on its own sentences, it got a lot worse…

After two generations, it started simply printing long lists.

And after four generations, it began to repeat phrases incoherently.

“The model becomes poisoned with its own projection of reality,” the researchers wrote of this phenomenon.

This problem isn’t just confined to text. Another team of researchers at Rice University studied what would happen when the kinds of A.I. that generate images are repeatedly trained on their own output — a problem that could already be occurring as A.I.-generated images flood the web.

They found that glitches and image artifacts started to build up in the A.I.’s output, eventually producing distorted images with wrinkled patterns and mangled fingers.

When A.I. image models are trained on their own output, they can produce distorted images, mangled fingers or strange patterns.

A.I.-generated images by Sina Alemohammad and others .

“You’re kind of drifting into parts of the space that are like a no-fly zone,” said Richard Baraniuk , a professor who led the research on A.I. image models.

The researchers found that the only way to stave off this problem was to ensure that the A.I. was also trained on a sufficient supply of new, real data.

While selfies are certainly not in short supply on the internet, there could be categories of images where A.I. output outnumbers genuine data, they said.

For example, A.I.-generated images in the style of van Gogh could outnumber actual photographs of van Gogh paintings in A.I.’s training data, and this may lead to errors and distortions down the road. (Early signs of this problem will be hard to detect because the leading A.I. models are closed to outside scrutiny, the researchers said.)

Why collapse happens

All of these problems arise because A.I.-generated data is often a poor substitute for the real thing.

This is sometimes easy to see, like when chatbots state absurd facts or when A.I.-generated hands have too many fingers.

But the differences that lead to model collapse aren’t necessarily obvious — and they can be difficult to detect.

When generative A.I. is “trained” on vast amounts of data, what’s really happening under the hood is that it is assembling a statistical distribution — a set of probabilities that predicts the next word in a sentence, or the pixels in a picture.

For example, when we trained an A.I. to imitate handwritten digits, its output could be arranged into a statistical distribution that looks like this:

Distribution of A.I.-generated data

Examples of initial A.I. output:

The distribution shown here is simplified for clarity.

The peak of this bell-shaped curve represents the most probable A.I. output — in this case, the most typical A.I.-generated digits. The tail ends describe output that is less common.

Notice that when the model was trained on human data, it had a healthy spread of possible outputs, which you can see in the width of the curve above.

But after it was trained on its own output, this is what happened to the curve:

Distribution of A.I.-generated data when trained on its own output

It gets taller and narrower. As a result, the model becomes more and more likely to produce a smaller range of output, and the output can drift away from the original data.

Meanwhile, the tail ends of the curve — which contain the rare, unusual or surprising outcomes — fade away.

This is a telltale sign of model collapse: Rare data becomes even rarer.

If this process went unchecked, the curve would eventually become a spike:

This was when all of the digits became identical, and the model completely collapsed.

Why it matters

This doesn’t mean generative A.I. will grind to a halt anytime soon.

The companies that make these tools are aware of these problems, and they will notice if their A.I. systems start to deteriorate in quality.

But it may slow things down. As existing sources of data dry up or become contaminated with A.I. “ slop ,” researchers say it makes it harder for newcomers to compete.

A.I.-generated words and images are already beginning to flood social media and the wider web . They’re even hiding in some of the data sets used to train A.I., the Rice researchers found .

“The web is becoming increasingly a dangerous place to look for your data,” said Sina Alemohammad , a graduate student at Rice who studied how A.I. contamination affects image models.

Big players will be affected, too. Computer scientists at N.Y.U. found that when there is a lot of A.I.-generated content in the training data, it takes more computing power to train A.I. — which translates into more energy and more money.

“Models won’t scale anymore as they should be scaling,” said ​​ Julia Kempe , the N.Y.U. professor who led this work.

The leading A.I. models already cost tens to hundreds of millions of dollars to train, and they consume staggering amounts of energy , so this can be a sizable problem.

‘A hidden danger’

Finally, there’s another threat posed by even the early stages of collapse: an erosion of diversity.

And it’s an outcome that could become more likely as companies try to avoid the glitches and “ hallucinations ” that often occur with A.I. data.

This is easiest to see when the data matches a form of diversity that we can visually recognize — people’s faces:

A.I. images generated by Sina Alemohammad and others .

This set of A.I. faces was created by the same Rice researchers who produced the distorted faces above. This time, they tweaked the model to avoid visual glitches.

This is the output after they trained a new A.I. on the previous set of faces. At first glance, it may seem like the model changes worked: The glitches are gone.

After two generations …

After three generations …

After four generations, the faces all appeared to converge.

This drop in diversity is “a hidden danger,” Mr. Alemohammad said. “You might just ignore it and then you don’t understand it until it's too late.”

Just as with the digits, the changes are clearest when most of the data is A.I.-generated. With a more realistic mix of real and synthetic data, the decline would be more gradual.

But the problem is relevant to the real world, the researchers said, and will inevitably occur unless A.I. companies go out of their way to avoid their own output.

Related research shows that when A.I. language models are trained on their own words, their vocabulary shrinks and their sentences become less varied in their grammatical structure — a loss of “ linguistic diversity .”

And studies have found that this process can amplify biases in the data and is more likely to erase data pertaining to minorities .

Perhaps the biggest takeaway of this research is that high-quality, diverse data is valuable and hard for computers to emulate.

One solution, then, is for A.I. companies to pay for this data instead of scooping it up from the internet , ensuring both human origin and high quality.

OpenAI and Google have made deals with some publishers or websites to use their data to improve A.I. (The New York Times sued OpenAI and Microsoft last year, alleging copyright infringement. OpenAI and Microsoft say their use of the content is considered fair use under copyright law.)

Better ways to detect A.I. output would also help mitigate these problems.

Google and OpenAI are working on A.I. “ watermarking ” tools, which introduce hidden patterns that can be used to identify A.I.-generated images and text.

But watermarking text is challenging , researchers say, because these watermarks can’t always be reliably detected and can easily be subverted (they may not survive being translated into another language, for example).

A.I. slop is not the only reason that companies may need to be wary of synthetic data. Another problem is that there are only so many words on the internet.

Some experts estimate that the largest A.I. models have been trained on a few percent of the available pool of text on the internet. They project that these models may run out of public data to sustain their current pace of growth within a decade.

“These models are so enormous that the entire internet of images or conversations is somehow close to being not enough,” Professor Baraniuk said.

To meet their growing data needs, some companies are considering using today’s A.I. models to generate data to train tomorrow’s models . But researchers say this can lead to unintended consequences (such as the drop in quality or diversity that we saw above).

There are certain contexts where synthetic data can help A.I.s learn — for example, when output from a larger A.I. model is used to train a smaller one, or when the correct answer can be verified, like the solution to a math problem or the best strategies in games like chess or Go .

And new research suggests that when humans curate synthetic data (for example, by ranking A.I. answers and choosing the best one), it can alleviate some of the problems of collapse.

Companies are already spending a lot on curating data, Professor Kempe said, and she believes this will become even more important as they learn about the problems of synthetic data.

But for now, there’s no replacement for the real thing.

About the data

To produce the images of A.I.-generated digits, we followed a procedure outlined by researchers . We first trained a type of a neural network known as a variational autoencoder using a standard data set of 60,000 handwritten digits .

We then trained a new neural network using only the A.I.-generated digits produced by the previous neural network, and repeated this process in a loop 30 times.

To create the statistical distributions of A.I. output, we used each generation’s neural network to create 10,000 drawings of digits. We then used the first neural network (the one that was trained on the original handwritten digits) to encode these drawings as a set of numbers, known as a “ latent space ” encoding. This allowed us to quantitatively compare the output of different generations of neural networks. For simplicity, we used the average value of this latent space encoding to generate the statistical distributions shown in the article.

• Share full article

Advertisement