Encyclopedia of Education and Information Technologies

Living Edition
| Editors: Arthur Tatnall

Computer-Assisted Learning

  • Pedro De BruyckereEmail author
  • Paul A. Kirschner
Living reference work entry
DOI: https://doi.org/10.1007/978-3-319-60013-0_73-1

Synonyms

Computer-assisted instruction; Computer-aided instruction; Computer-based learning; Computer-mediated learning

This contribution explores computer-assisted learning (CAL) from a number of different angles. First, it discusses the many synonyms that exist for the use of information and communication technologies (ICTs) in education. Within these synonyms are concepts which often hide subtle but important differences from both a technological as well as a pedagogical/educational point of view. It proceeds with a short history of CAL. While CAL as an entity was first used in the early 1960s of the previous century (1960s), the concept underlying its use has roots dating back to the beginning of that century. The third and final part will focus on the effectiveness of CAL in education which will lead to an adaptation of the current definition.

What’s in a Name?

In their Dictionary of Computing, Daintith and Wright (2008) describe CAL as

[a]ny use of computers to aid or support the education or training of people. CAL can test attainment at any point, provide faster or slower routes through the material for people of different aptitudes, and can maintain a progress record for the instructor. (Online entry)

This definition has changed through the course of time as the computer itself has changed from a bulky, building-large behemoth of tubes and wires to something that fits into your pocket and can be even smaller. In the 1960s, computer use was first introduced in education. In this first phase of CAL, the computer was a mainframe which was connected to learners in specifically designed “classrooms” via terminals that were relatively far away from the mainframe computer. In this phase, the computer was primarily used as a knowledge-bank of questions which students could use for self-assessment of their knowledge. Using computers was extremely expensive and cumbersome and was only available for a “happy few.”

In the 1980s and 1990s, CAL was often used to refer to the development of either a single computer program or a series of programs which replaced more traditional methods of teaching. This was the direct result of the introduction in August 1981 of the IBM PC (personal computer) followed by the Apple II shortly thereafter. A mainframe computer was no longer needed to use computers for educational purposes and each student could work at her/his own PC in school or, for those pioneers, at home. This period also saw the introduction of what is now called computational thinking/coding with the launch of LOGO by Seymour Papert and his group at the Massachusetts Institute of Technology Artificial Intelligence Laboratory where, with the aid of PCs/microcomputers children in classrooms received an initiation to computer programming skills, turning it from an abstract activity only carried out by specialists into a concrete, real activity that could be carried out by students to program the movement of a robotic turtle.

This second wave was followed by a third wave ushered in by the introduction and availability of the Internet/World Wide Web where computers and people worldwide became connected to each other. This third wave also brought us tools such as email, webpages, blogs, etc. This broadened the concept of CAL to include, for example, web-based tools such as online learning environments, opening the door to online databases, making learning analytics a possibility.

The present wave is shaped by the availability of mobile devices (e.g., smartphones, tablets), semi-intelligent peripherals (e.g., augmented and virtual reality glasses, motion sensing input devices, such as the Kinect or the WII), and seamlessly integrated devices (e.g., the Internet-of-things) where CAL literally can take place anytime and anywhere where there is Internet (and that is virtually anywhere in the world).

Going back to Daintith and Wright’s (2008) definition, when it is examined in detail one can distinguish a number of core elements:
  • Use of computers: while there were analogue computers up to the 1960s, this now implies the use of many different digital media interconnected via the World Wide Web. To update this, it might be better to speak of the “use of multimedia and the Internet.”

  • Support the education or training of people: the “computer” will take up one or more roles and functions of the teacher and/or trainer to offer the learner necessary support and guidance during the instructional or learning event.

  • Test attainment at any point: the “computer” will keep track of the learning process to determine whether the learner has achieved the goals of the CAL. It often does this by comparing the different products during the learning period with the beginning and end situations. This is related to what is now known as “learning analytics” which can be described as “the measurement, collection, analysis and reporting of data about learners and their contexts, for purposes of understanding and optimising learning and the environments in which it occurs.” (Ferguson 2012, p. 305)

  • Provide faster or slower routes through the material for people of different aptitudes: the “computer” provides possibilities to adapt the content of, pace of, and/or approach to learning based on learner progress which is presently determined by learning analytics. The terms now being used for this are adaptive learning and personalized learning.

  • Maintain a progress record for the instructor: the “computer” is not only facilitated to let the technology adapt to the learner but also to inform the teacher/instructor to be able to adapt his or her approach based on the learning analytics.

Daintith and Wright (2008) also state that CAL is one of several terms used to describe this application of computers. Indeed, there have been many different names given to the use of ICT in education. Some names focus on the kind of technology being used, such as computer-mediated learning, computer-based learning, intelligent tutoring systems, web-based learning, and mobile learning. In these aforementioned cases, there can be an overlap between which technology is used and how the technology is used. And although the terms are often used as almost synonyms, at the same time it is perfectly possible to use mobile technology that is not web-based, to use computer technology that is not mobile or to use web-based tools that are not mobile or other web-based tools that cannot be used on personal computers.

The many names for the use of information and communications technology (ICT) in education can also hide differences between pedagogical and educational views, for example, the not so subtle difference between CAL and computer-assisted instruction (CAI). The difference between learning and instruction is the difference between a focus on the task of the pupil (learning) and a focus on the task of the teacher/teaching (giving instruction). This distinction goes far beyond the use of technology and has deep historical roots going back to the Romantic era (eighteenth century), but this does not mean they can be disregarded. Which name people use for the use of technology in education often discloses a certain vision on or philosophy of education. When using CAL, one comes close to a paradigm of social-constructivism where a pupil must build her or his own knowledge, while CAI could be regarded as a more cognitivist or instructivist approach which puts the emphasis on the need for an expert who can give the necessary support and guidance (Kirschner et al. 2006).

Finally, some of the names are a combination of the two. Take, for example, computer supported collaborative learning (CSCL). On the one hand, there is an emphasis on the technology, namely, the computer or, better stated, computers connected to each other through a network. On the other hand, there is emphasis on the pedagogy, namely, collaborative learning. This is an approach to learning whereby a shift is made from learning individually and competitively to learning with others in the context of a team carrying out a task or solving a problem such that the members of the team share authority and accept responsibility for the group’s actions.

The Pedagogical and Educational Roots, History, and Near Future of CAL

The oldest roots of thinking about machines helping someone to learn in a way resembling the definition of CAL date back to the first half of the previous century. Edward Thorndike (1912) was one of the first to think along the lines of providing faster or slower routes through the material for people of different aptitudes. He specified three conditions or laws that maximized learning. His law of effect held that the likelihood of a recurrence of a response is affected by reward or punishment. The law of recency held that the more recent something is, the more likely that it will recur. Finally, the law of exercise held that stimulus-response associations are strengthened through repetition. In his eyes, and based upon his laws, “[i]f by a miracle of mechanical ingenuity, a book could be arranged that only to him [sic] who had done what was directed on page one would page two become visible and so on, much that now requires personal could be managed by print instruction” (p. 165). But Thorndike did not make such a – mechanical – learning machine himself. The first to do so was Sidney Pressey who developed a mechanical testing machine which he, oddly enough, called the Automatic Teacher. What it did was present students with multiple-choice questions, one at a time, that were sequences from least to most difficult. The student could only move on to the next question by answering the previous one correctly. This was intended to lighten the load on the teacher with respect to grading students so she/he could focus on more interaction with the student. While his machine was originally designed to make the testing and scoring of simple multiple-choice achievement tests possible, he soon discovered that it also had instructional qualities closely related to what later became known as programmed instruction as the device could show a predetermined sequence of questions one at a time, through an aperture. The student selected an appropriate answer from a card and pressed a key representing that answer; the selection of the correct answer resulted in the machine prompting a new question for the student (see Fig. 1).

After Pressey, more and different teaching machines were trialed and tested until in 1954 B.F. Skinner wrote his The Science of Learning and the Art of Teaching. In this seminal work, Skinner (1954) laid down the basics of his vision on programmed instruction which was then later translated through different incarnations of teaching machines. The most important difference with Pressey’s teaching machine – though similar to Thorndike’s law of effect – was the strong emphasis that he placed on reinforcement during the learning process (Skinner 1961).

Skinner was inspired by the Socratic teaching method and, thus, he broke down complex lessons into a series of small questions. Each question required the student to respond to it in a way that built upon the correct response to the previous one. In this way, Skinner’s machine (see Fig. 2) required students to master one concept before they were able to move onto the next, a type of learning that eventually came to be known as mastery learning.
Specific to all of these early teaching machines was their basis on a behaviorist view of learning (actually conditioning) where a stimulus is presented (a unit of preferably new information), a reaction (a response) is required, followed by immediate feedback as to the correctness of the response and which is often repeated to strengthen the connection. Arthur Lumsdaine (1959, p. 164), who researched the use of media and programmed learning, wrote that teaching machines have three basic characteristics:
  • First, continuous active student response is required, providing explicit practice and testing of each step of what is to be learned.

  • Second, a basis is provided for informing the student with minimal delay whether each response he makes is correct, leading him directly or indirectly to correction of his errors.

  • Third, the student proceeds on an individual basis at his own rate – faster students romping through an instructional sequence very rapidly, slower students being tutored as slowly as necessary, with indefinite patience to meet their special needs.

The devices thus represent a way of providing a preprogrammed study-practice combination which simulates, in partially or fully automated fashion, the functions of a private tutor in recitation and practice, with immediate correction of errors and feedback to the student.

This first phase of learning machines can be regarded as the behavioristic phase. Though different forms of “computers” were around long before the 1960s – the Antikythera mechanism, for example, has been described as a computer from ancient Greek times (Freeth et al. 2006) – and there have been different learning and teaching machines, the concept of “computer-assisted learning” stems, as stated, from the 1960s of the previous century. While scientific references are older (e.g., Coulson et al. 1962; Glaser 1965; Suppes 1965), the first real mentions of CAL in scientific journals can be found in 1966, after its mention at conferences in 1965 (e.g., The Conference on Computers in Physics Teaching, University of California at Irvine, November 1965) (Based on searches in ERIC and Scholar which resulted in Fowler (1966), Hirsch (1966), and Trump (1966) as oldest sources).

Probably due to a combination of resources, perceived affinity of the domain for CAL and political forces that often drive changes in education, the natural sciences (including mathematics and medicine) were the early adopters of CAL. A good overview of this can be found in a special issue of Computers & Education in 1992 (volume 19, numbers 1–2). For example, the integration of the computer into medical education was recommended in 1979 by Daniel Tosteson, then Dean of Harvard Medical School, when he proposed that the use of what he called “information-processing devices” be developed as a core component of the medical school curriculum. Michael Hewson, in his seminal work on the use of microcomputers in science education (1984) summed it up as follows: in science education, microcomputers allow students to play a more active role in learning, get individual attention for their specific difficulties, and allow students to control the pace at which they work. Also, physics educators showed a particular interest in CAL (e.g., in Fowler (1966)). Another specific form of CAL, computer-assisted language learning (CALL) stems from the same period (Marty 1981), though the use of computers for language learning was at first limited to universities. CALL makes use of computer technology to aid in the presentation, reinforcement, and assessment of language material – usually a second or a foreign language – to be learned. While originally based upon stand-alone applications on microcomputers, CALL now uses a broad range of technologies and approaches ranging from traditional drill-and-practice programs to virtual learning environment and web-based distance learning with its most recent being mobile-assisted language learning (MALL; Shield and Kukulska-Holme 2008).

After these initial often behavioristic inspired approaches of CAL – see also the link with Skinner and Thorndike – in the 1970s and 1980s a different approach to CAL and CALL could be seen with more Intelligent Tutoring Systems introducing a second rather cognitivist phase. The advocates of this approach argued that all CA(L)L courseware and activities “should build on intrinsic motivation and should foster interactivity – both learner-computer and learner-learner” (Han 2009, p. 41). In this cognitivist phase, research and development now were directed at finding, studying, and implementing methods and strategies that would bring about specific and desired cognitive processes, for example, assisting learners in selecting information, organizing it into internally consistent concepts, and integrating new with existing knowledge in cognitive schemata so that it would be personally relevant and meaningful. This second phase coincided with the second wave discussed earlier that came from outside education as, from the 1980s onwards; the microcomputers made computer technology more accessible to a broader audience and to educational institutes besides the universities. Warschauer and Healey (1998) describe a third phase as social-constructivist with a strong emphasis on agency (Warschauer 2000). But it can be argued that for this objective of agency, a democratization of microcomputers was not enough; the third wave described as integrating the worldwide web was key for this.

A fourth phase benefited also from the technological evolutions which came with the development of network-based technology, through which people can share whatever and communicate with each other whenever and wherever (Tafazoli and Golshan 2014), which made an evolution possible from a social-constructivist approach to a connectivist approach. This phase coincided with computer technology becoming more mobile, which led to mobile devices such as laptops, smartphones, iPads, and Augmented Reality-devices, with the possibilities of a steady growth of computing power, making it possible to handle large amounts of data, now referred to as Big Data, or as we call it today in education: learning analytics. These kinds of mobile devices, combined with technology as the original definition stated that is able to test attainment at any point, can give the user both teacher and learner agency.

A further evolution of CAL can come in the form of the use of artificial intelligence (AI), with smart assistants helping both teachers and students in their teaching and learning process. A first example was delivered by “miss Jill Watson,” an AI-assistant powered by IBM’s supercomputer Watson that answered email-questions of students of Professor Eric Wilson (Korn 2016).

The Effectiveness of Computer-Assisted Learning

The first law of Kranzberg (1986) states that “technology is neither good nor bad; nor is it neutral.” This is also the case when discussing the effectiveness of CAL. In 2015, the OECD (2015) concluded that more computers do not necessarily mean more learning; it is more important to see how the technology is being used. In his review of the many different meta-analyses on computer-assisted instruction, John Hattie (2009) described a reasonable degree of variability across the overall effect sizes. The advice Hattie gives is that the use of computers is more effective when:
  • There is a diversity of teaching strategies.

  • There is teacher pretraining in the use of computers as a teaching and learning tool.

  • There are multiple opportunities for learning (e.g., deliberative practice, increasing time on task).

  • The student, not the teacher, is in “control” of learning.

  • Peer learning is optimized.

  • Feedback is optimized.

In other words, it is the pedagogy and not the technology that can affect outcome. This is an echo of Richard Clark (1983) who found that research on media showed that when differences between media on learning are found, it is the method that causes the difference and not the medium. He summed this up in his famous quote, “[M]edia are mere vehicles that deliver instruction but do not influence student achievement any more than the truck that delivers our groceries cause changes in our nutrition” (p. 445).

A more recent literature-review for The National Bureau of Economic Research by Escueta, Quan, Nickow, and Oreopoulos discusses the evidence to the present time (2017) on the use of technology in the classroom. For their review, the authors compiled publicly available quantitative research that used either randomized controlled trials or regression discontinuity designs (where students qualify for inclusion in a program based on a cut-off score at pretest). All the studies that were used in the working paper examined the effects of an educational technology intervention on any education-related outcome, which makes this work broader than studies only looking at a learning effect. The working paper included not only the areas of technology access, computer-assisted learning, and online courses but also the less-often-studied technology-based behavioral interventions.

What are the conclusions?
  • Access to technology may or may not improve academic achievement at the K-12 (4 to 6-year-olds through 12th grade (12) for 17 to 19-year-olds) level, but does have a positive impact on the academic achievement of college students.

  • Online learning courses had the least amount of research that could be examined and showed the least promise of the four areas. However, when online courses were accompanied by in-person teaching, the effect sizes increased to scores comparable to fully in-person courses.

But the most relevant conclusion of the paper was that:
  • Computer-assisted learning, when equipped with personalization features, was an effective strategy, especially in mathematics (Escueta et al. 2017).

This last conclusion means that one element of the definition of CAL is key for its effectiveness: Provide faster or slower routes through the material for people of different aptitudes, although personalization can go much further than adapting the speed of the learning routes; in this regard, the 2010 National Educational Technology Plan (Office of Educational Technology 2010) describes personalized learning as “broader than just individualization or differentiation in that it affords the learner a degree of choice about what is learned, when it is learned and how it is learned; this may not indicate unlimited choice, since learners will still have targets to be met.” This element of personalization, its role in effectiveness of computer-assisted learning, and the new possibilities from technology have warranted an adaptation of the definition by Daintith and Wright. A possible amendment could be the following (the modification is in italics):

Any use of computers to aid or support the education or training of people. CAL can test attainment at any point, personalise the learning routes both in speed, complexity and other possible forms of personalisation of learning, and can maintain a progress record for the instructor.

Conclusion

Computer-assisted learning (CAL) is older than one may think, dating back to the early 1960s of the previous century, with roots dating back to the beginning of the previous century. The meaning and focus of CAL has changed throughout time, shifting from behavioristic, through cognitivist and social-constructivist to connectivist visions, often fuelled by technological evolutions. While many people have put hope in technological breakthroughs making it possible to have better learning results, the actual research shows mixed results.

Cross-References

References

  1. Clark RE (1983) Reconsidering research on learning from media. Rev Educ Res 53:445–459CrossRefGoogle Scholar
  2. Coulson JE, Estavan DP, Melaragno RJ, Silberman HF (1962) Effects of branching in a computer controlled autoinstructional device. J Appl Psychol 46(6): 389–392CrossRefGoogle Scholar
  3. Daintith J, Wright E (2008) A dictionary of computing. Oxford University Press. Retrieved from http://www.oxfordreference.com/view/10.1093/acref/9780199234004.001.0001/acref-9780199234004-e-946
  4. Escueta M, Quan V, Nickow AJ, Oreopoulos P (2017) Education technology: an evidence-based review (No. w23744). National Bureau of Economic Research, Cambridge, MACrossRefGoogle Scholar
  5. Ferguson R (2012) Learning analytics: drivers, developments and challenges. Int J Technol Enhanc Learn 4(5–6):304–317CrossRefGoogle Scholar
  6. Fowler JM (1966) Progress report of the commission on college physics. Am J Phys 34(9):834–861CrossRefGoogle Scholar
  7. Freeth T, Bitsakis Y, Moussas X, Seiradakis JH, Tselikas A, Mangou H (2006) Decoding the ancient Greek astronomical calculator known as the Antikythera mechanism, Nature. 444(7119):587–591CrossRefGoogle Scholar
  8. Han W (2009) Benefits and barriers of computer assisted language learning and teaching. US-China Foreign Lang 6(9):40–43Google Scholar
  9. Hattie JA (2009) Visible learning: a synthesis of 800+ meta-analyses on achievement. Routledge, AbingdonGoogle Scholar
  10. Hewson PW (1984) Microcomputers, conceptual change and the design of science instruction: examples from kinetic and dynamics. S Afr J Sci 80:15–20Google Scholar
  11. Hirsch WZ (1966) Planning education today for tomorrow. Urban Aff Q 2(1):3–18CrossRefGoogle Scholar
  12. Kirschner PA, Sweller J, Clark RE (2006) Why minimal guidance during instruction does not work: An analysis of the failure of constructivist, discovery, problem-based, experiential, and inquiry-based teaching. Educ Psychol 41(2):75–86CrossRefGoogle Scholar
  13. Korn M (2016) Imagine discovering that your teaching assistant is really a robot. Wall Str J 119:6Google Scholar
  14. Kranzberg M (1986) Technology and history: “Kranzberg’s Laws”. Technol Cult 27(3):544–560CrossRefGoogle Scholar
  15. Lumsdaine AA (1959) Teaching machines and self-instructional materials. AV Commun Rev 7:163–181.  https://doi.org/10.1007/BF02767482CrossRefGoogle Scholar
  16. Marty F (1981) Reflections on the use of computers in second language acquisition. System 9(2):85–98CrossRefGoogle Scholar
  17. OECD (2015) Students, computers and learning: making the connection. PISA, OECD Publishing.  https://doi.org/10.1787/9789264239555-en
  18. Office of Educational Technology (2010) National educational technology plan. Office of Educational Technology, Washington, DCGoogle Scholar
  19. Shield L, Kukulska-Hulme A (2008) An overview of mobile assisted language learning: from content delivery to supported collaboration and interaction. ReCALL 20(3):249–252.  https://doi.org/10.1017/S095834400800013XCrossRefGoogle Scholar
  20. Skinner BF (1954) The science of learning and the art of teaching. Harv Educ Rev 24:86–97Google Scholar
  21. Skinner BF (1961) Why we need teaching machines. Harv Educ Rev 31:377–398Google Scholar
  22. Suppes P (1965) Computer-assisted instruction in the schools: potentialities, problems, prospects. Institute for Mathematical Studies in the Social Sciences, Stanford University, Stanford, p 17Google Scholar
  23. Tafazoli D, Golshan N (2014) Review of computer-assisted language learning: history, merits & barriers. Int J Lang Linguist 2(5–1):32–38Google Scholar
  24. Thorndike EL (1912) Education: a first book. Macmillan, New YorkGoogle Scholar
  25. Trump JL (1966) Secondary education tomorrow: four imperatives for improvement. Bull Natl Assoc Second Sch Princ 50(309):87–95Google Scholar
  26. Warschauer M (2000) CALL for the 21st Century. In: IATEFL and ESADE conference (Vol. 2)Google Scholar
  27. Warschauer M, Healey D (1998) Computers and language learning: an overview. Lang Teach 31(2):57–71CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  1. 1.ArteveldehogeschoolGhentBelgium
  2. 2.Leiden UniversityLeidenThe Netherlands
  3. 3.Open UniversiteitHeerlenThe Netherlands
  4. 4.University of OuluOuluFinland

Section editors and affiliations

  • Don Passey
    • 1
  1. 1.Department of Educational ResearchLancaster UniversityLancasterUK