Last changed 23 Oct 2017 ............... Length about 2,000 words (17,000 bytes).
(Document started on 15 Feb 2005.) This is a WWW document maintained by Steve Draper, installed at You may copy it. How to refer to it.

Web site logical path: [] [~steve] [best] [this page]

The real educational issues of learning objects

By Steve Draper,   Department of Psychology,   University of Glasgow.


I first encountered the distinctions made here in a brief mention in a talk by Roy Tasker. He tells me that in fact they came from discussions at ASCILITE 2002 (the Australasian educational technology conference). Almost certainly, different people have somewhat different definitions: this is my own development of the basic idea.

The common overall point, however, is to remind us how little the technology itself does in determining whether any learning in fact occurs.


I'm going to give examples not only of digital cases of each of the 4 levels, but also of text and of physical teaching materials.

Digital asset

  • a videotape
  • a multimedia file
  • an executable physics simulation program
  • In a 3D modelling package, models of a pile of cardboard squares, each a different uniform shade of colour.
    Non-digital cases:
  • The main (expository) text in a textbook.
  • A French-English dictionary
  • The text of Middlemarch (by George Eliot)
  • The chemical sodium bicarbonate
  • A stopwatch.
  • The transcript or tape of a lecture
  • A pile of cardboard squares, each a different uniform shade of colour

    Learning object

  • A modern textbook as sold: not only the exposition, but contents, index, exercises, the covers to protect it and label it, page numbers and a binding to allow rapid flipping through it. Note that in normal life, a textbook is used for a number of different tasks, supported by different features of this object: flipping through it in a shop deciding whether to buy it, learning from it the first time, revising from it, using it as a reference source to look up specific things years later, etc. These multiple learner tasks applied to a single asset are a crucial reason for distinguishing asset from object. Typically modern digital learning objects are much poorer at this than modern printed text books, which have evolved steadily for centuries to serve more and more purposes (learner tasks).
  • A printed copy of Pride and Prejudice (by Jane Austen): either a good read for pleasure or a set book in a Literature course. Modern novels come as themselves; "classic" novels usually come with an "introduction": an essay by a scholar other than the author.
  • Copper sulfate (blue crystals): as a chemical supply; or in suitable small packaged quantities for 100 chemistry students to do a lab exercise with, complete with a label suitable for the students.
  • Sodium bicarbonate: packed and labelled for home cooking; for student chemistry experiments; ..
  • The physics simulation program with some information on what machines it will run on, and what courses it might fit into.
  • A lecture: often their purpose or role ("metadata") are announced orally and/or in handbooks e.g. "Attend if you need revision on ...", "will serve as an introduction to the reading", "essential as it contains material not in the textbook".
  • The set of coloured tiles, plus tags saying a large table top is required, and a tutor trained in their use or with the tutor script.

    Learner's activity Laurillard lists exactly 12 generic activities e.g. expression (exposition by lecture or textbook), re-expression (a student writes an essay, tries to answer a question, tries to tell another student about it), ..etc.

    Tasker's and his colleagues' idea of activity is slightly different. Examples include: Explore, Describe, Apply, Observe, Represent, Refine, Review, Access, Question, Decide, Report, Reflect, Interpret, Construct, Justify, enRole, Research, React, Resolve,

    The simulation program plus a worksheet for students of things to run on it, settings to try, phenomena to set up and observe, ...

    Learning session Neither Tasker nor Laurillard call a lecture or tutorial or going through an online document an "activity", firstly because these are generic formats for assets (like "books" or "videos"): a specific learner task must be added to the object. Of course, skillful students (or researchers at a conference) will apply their own goals: although not always what the author intended. In particular cases however, a lecturer may tell the audience what they think the activity should be for the next bit "Now put down your pens and just think about ....". However in many cases of pedestrian practice, the actual student task in lectures is not thinking nor learning, but collecting material for later possible learning.

    Sitting down a learner with tutor and a set of coloured tiles: in one case, 10 tiles of fully saturated hues. They are asked to arrange them in any order that seems logical to them. When they finish, the tutor will ask what is the rationale for their arrangement; and (if it isn't the arrangement in fact eventually required) ask them probe question e.g. (if this learner put them in a straight line) "Could the two tiles at extreme ends from each other in fact have been placed adjacently?"

    Learning design
    Tasker's own example schemata for learning designs include:

    But a more classic design might be:

    The use of the coloured tiles (whether on a tabletop, or in a 3D digital modelling package) is part of a design where the learner is given a sequence of about 10 tasks, arranging subsets of colours and then merging arrangements, and then finally placing them on a skeleton sphere to form the Runge sphere (the hue, saturation, brightness 3D colour space).


    The above, in my view, is the best current expression of an old lesson that keeps having to be relearned around educational technology: it isn't the technology but the pedagogical or learning design around it that makes a difference to learning. Obviously if you are a technology enthusiast you are liable to think the technology is the essential thing, but in fact experienced teachers often fall into the same error. They see (for example) a fascinating simulation that both excites them, and perhaps teaches them some new aspect of an old topic. For instance I was once present at a demo of a simulation on Taylor series for the members of my university's maths department, and one of them commented that they had been teaching this for years, but it had made them realise something new about Taylor series. It is natural for them to want to share this with their students. Then they put the software in front of the students, and are bewildered when nothing happens in most cases. The teachers have a whole "context" of partial knowledge in their heads, and for them the simulation (the digital asset) alone can be enough for a rich learning enhancement experience. But most students do not have that context: that is why they are students. When Papert and his associates were at the height of "pushing" LOGO as a programming language for children, with huge claims about the educational benefits, they talked about "the blank screen phenomenon" of how nothing happened if a child is just given LOGO by itself. My own weakness of this kind is more to do with diagrams: for the most important ideas I find or develop, I often end up creating a diagram that for me summarises it all (Laurillard's model; learning causes); but my painful experience is that these mean little to students being introduced to the ideas. At best they become useful later, but are useless for introductions.

    Pulling this together: it constitutes in another form the more abstract theoretical point that the learners who do best are generally those who already know the most, using their partial knowledge to gain access to the meaning of new material, and their stock of open questions to direct what they want to learn from it. An expert sees what is interesting where a layperson notices nothing. Having said that, the best interactive museum exhibits succeed in drawing in a wide variety of people: but these are rare.

    Tasker's 4-way distinction first makes the point that technology alone causes no learning. Secondly, it offers a first way to break down the extra work that needs to be done, and so makes a start at planning for it by giving a framework for understanding what needs to be added to naked technology or media. As I say, it is a lesson that has been painfully rediscovered again and again. Tasker's is the clearest and furthest developed statement of this core point that I have come across.

    See also Barney Dalgarno paper

    Roy Tasker

    I don't know Roy well, but I think his work is notable. Below are some pointers to him and his work. Here is some context.

    A starting motivation for him was Alex Johnstone's identification of a key bottleneck for students learning chemistry: learning in three different representations at once and how to inter-relate each new concept or fact in all three domains: the macroscopic (e.g. how chemical phenomena appear to the senses, colour, smell, etc.); the formal or representational (the equations used to represent reactions); and the "submicro": the invisible but 3-dimensional world of molecule's shapes and their dynamic motions, interactions, and kinetics. The third of these is generally the hardest for students, and least well dealt with in teaching.

    This identified a strategic educational problem in chemistry, and Roy took it on. His key idea for a solution was to develop computer animations that can show the shapes and motions of molecules, together with skilled tutorial dialogue to get students to see the problems with their assumptions and prior conceptions for which the animations offered insight.

    The "patter" that came with the computer animations is actually highly skilled socratic dialogue. For me the demo Roy once gave me in 1993? was a notable learning experience, and an exemplar I have always remembered about a mode of learning. I currently am involved in a project that in my mind was inspired by this, although in a very different area (colour theory): creating an effective learning experience around visual exercises and demonstrations, and socratic dialogue from a human tutor that guides the learner into recognising and confronting latent problems in their pre-existing partial knowledge.

    But personal human 1:1 tuition isn't cost sustainable. So Roy had two aims for the next decade of work: more simulations and animations (generalising his early exemplars to cover more of chemistry teaching); and how to replace himself as part of the package. His distinctions above reflect part of his growing analysis and understanding of what he was value-adding to the software itself.

    There is a recent PhD thesis supervised by him on this stuff:
    Rebecca Dalton (2003) The development of students' mental models of chemical substances and processes at the molecular level (University of Western Sydney). Available online: type in "Dalton" in the author box.

    Pointers to Roy Tasker

  • Home page/profile
  • Learning designs web site
  • Publisher demos on the web

    Another set of distinctions

    Simulations. animations (demonstrations); models.

    This area (of using computer animations to teach aspects of science) can raise the need for another set of distinctions e.g. between animations and simulations. A rough go at these might be:

    Having said that, every simulation is only realistic about some properties, and not others that it doesn't attempt to model.

    Web site logical path: [] [~steve] [best] [this page]
    [Top of this page]