Table of Contents Teaching & Learning with Internet Tools
A Position Paper
presented at the Workshop on "Teaching & Learning with the Web" at the First International Conference on the World-Wide Web, 1994 at CERN, Geneva.
Daniel Schneider, TECFA, FPSE, University of Geneva
email@example.com - http://tecfa.unige.ch/
This position aims to encourage discussion between WWW providers of educational material and researchers in the field of educational technology. This paper is still a early draft and I am afraid it will retain that status for a while
Introductory reference material in the Appendix sections has awful very rough draft status. It is useful only as far as it points to some real reading. The table of Contents is in a separate file (by pure laziness) for the moment). Everything is straight Framemaker to html translation without modifications...)
What is known about human learning relevant for the design of educational material? How should this knowledge be used in the context of more specific educational goals and WWW-supported learning & teaching activities? Generally speaking there are two important statements about learning and instruction:
The learning environment designer must take into account both perspectives. WWW-based courseware must not restrict itself to delivery of educational content. It must be grounded in some model of instruction and learning. Many possibilities exist and haven proven to be effective. However, each paradigm works under certain conditions in certain situations using some set of specific educational technology. For example, it can easily be argued that a good book is better than hypertext version of that book (why do people always print out things?). Also, general rules can be formulated such as "learning without doing is pretty useless in most domains". Consider the following questions:
- One learns by doing something (psychology)
- One learns by pursuing an instructional goal (education sciences)
Some basic misconceptions about learning have to avoided: Reading or seeing does not imply much learning. Even being able to recall knowledge does not mean being able to apply knowledge. Efficiency is not measured by mastery of the exercises and tests of a courseware tool, but my mastery of the task. Also be aware of more subtle knowledge transfer problems: Even "micro"-competencies such as operating an Airbus vs. a Boing or programming in C instead of Pascal are difficult to teach. If more abstract things like "programming" instead of just C programming have to be learned, special teaching strategies have to be used (such exercising skills in variating contexts. Lastly, the complexity of a learning environment must be adapted to the learners skills (Would you teach C as an introductory programming language)?
- Would you take the plane if you knew that the pilot has read all the documentation about flying and successfully passed examination testing his knowledge?
- Would you take a plane if you knew that the pilot has been trained with Microsoft flight simulator? Which one do trust better?
- Would you take an Airbus if you knew that the pilot has been trained on a B-737 and has read the WWW multimedia Airbus documentation?
- If you had to teach flying an Airbus, would you take a would-be pilot and show him how to fly "hands-on"?(1)
Learning appears to be a complex matter. No doubt that this is the reason why all the various branches of learning theory do not even view the problem from a same angle. However, all academic traditions do provide the learning environment designer with important key ideas:
Courseware engineering is concerned with electronic learning environments. Such an environment is a combined system involving tasks, agents, courseware products, etc. which is aimed at supporting learning processes and in which learning takes place mostly in interaction between learners, courseware products, other tools, and to a lesser degree tutors (human or artificial). Courseware is always a combination of elements (cf. de Dinan & van Shaik 93: 193), such as:
- Learning must take place within optional external "conditioning" (behaviorism)
- Learning is related to active problem solving and involves integration, construction and compilation of new content (cognitivism)
- Learning is constrained by human cognitive capacities (experimental psychology, HCI theory)
Each type of courseware architecture organizes those resources in various ways. How complete are the materials supplied to the learner with respect to the information and support needed by the learner to achieve an instructional goal? Several ways of tackling this problem have been proposed.
- textual material (including textual representations)
- simulation models,
- feedback information, etc.
We shall briefly introduce some ideas from (1) Instructional Design and from (2) Advanced Learning Environment Research.
Instructional Design Theory provides detailed prescription on how to organize teaching and learning at the global (curricula), lesson and task level. Most work is also grounded in some learning theory. Despite and maybe also because of the level of details those approaches attempt to formalize instruction their practical use is often debated. Some argument against reading much instructional design theory is that a good teacher with good practice intuitively knows and uses things like Gagné's steps.
Most people agree that instruction needs principles, however some researchers feel that instructional theory should not just be grounded in learning theory but BE applied learning theory and to implement optimal learning conditions according to what we know about learning. This is the way most research in Advanced Learning Systems operates.
At the current state of the WWW technology, it is not possible to implement this kind of advanced learning environment (without making use of external clients). However, there are points that WWW based courseware can adopt.
- The Learner must be active
- A learning environment should be designed as powerful dedicated working environments. It must be rich and complex reflecting the essential properties of what has to be learned.
- The environment must be structured. If the richness of a learning environment is a quality, its complexity may reduce learning. It must provide optimal learning conditions in function of the learner's stage of knowledge.
- Learning environments should be designed as hierarchical knowledge base generators
- Learning environments should present knowledge as a communication system. A learner must interact with agents, tutors.
Here is a (short) list of different kinds of computational learning environments in use. They represetn different learning paradigms and can be classified along several axes like "Instruction - Learning", "External - Internal Control". Each are still appropriate for certain kinds of learning.
Now match that to those more technical items:
- Programmed Instruction (little step by step transfer of content)
- Computer Assisted Instruction (Drills & Tutorials)
- Intelligent Computer Assisted Instruction (ITS Tutorials)
- Computer Based Learning (Simulations, Hypertext & Microworlds)
- Intelligent Learning Environments (Microworlds + tutors, helpers, experts)
- Cognitive Learning Support Environments (some hypertexts)
- Knowledge Construction & Environments & Intellectual Toolkits
The WWW, specially in conjunction with external local or server-side clients offers a lot of possibilities. But not everything can be done and it is important not to use in inappropriate learning paradigm for a given educational goal.
- Information servers to look up information (manuals, books, expositions, bibliographies, programs, etc.)
- Distribute educational material (texts, programs)
- Provide curricula & guidance to lessons and exercises in hypertext format.
- Implement collaborative work (dynamic hypertext, "News like conferencing system", co-writing)
- Implement Jigsaw puzzles
- Question & Answering, (tests, Skinner & Bloom type of learning to some extent)
- Interface to local clients (e.g. simulations, programming environments, tutors, etc.)
- The same thing over the web (e.g. have intelligent nodes, cf. Mallery)
There are 3 aspects:
In more simple and technical terms, courseware engineering (cf. De Dian & van Schaik 93:193) is concerned with:
- Integration into a learning environment setup: What role in learning & teaching does hypermedia have? (see also "What learning activity can we do with the Web?" on page 6) What other tools do we need?
- Integration into a learning environment architecture: How can we make integrate hypermedia with other computational learning tools?
- Computational integration with other programs: How can we build achieve tight integration?
Furthermore, Courseware engineering is not everything. Courseware alone rarely constitutes the full learning environment. Authors of Instructional Material on the Web should be aware of the following:
- transferring educational information
- organizing pedagogically optimized access to this information via an appropriate interface and structuring of the material
- implementation of instructional tactics, e.g.:
- giving examples
- multiple choice questions
- asking the student to perform a task, etc.
- telling the student what learning strategy to adopt with some material
- implementation of instructional strategies, i.e. sequencing of teaching materials
- Teaching and learning involves a learning environment. It is not good enough to hypertextify a text or implement form-based tests. Good Web based courseware is more global in design. Not everything must be built into html or server-side scripts, but somehow instructional tactics & strategies as well as learning strategies have to be "put" upon the learner or communicated to the learner. Last, but not least he needs assistance and very often cooperation with other learners.
- Delivery of instructional text (multimedia and hypermedia) has be structured according to some pedagogical & learning strategies. If the learner has to discover everything himself - fine - but tell him so!
In discussing the role of technological support in education, Sandberg (94:225) identifies the components of a (technologically rich) learning environment (see "Overview of the Learning Environment (Sandberg 94:225)" on page 9). These components must all be there in order to optimize learning. However, they can be "implemented" in many different ways. Each component has functionalities for which we should insure:
Let's examine the Table 1, "Filling in the Functionalities of a Learning Environment," on page 10. The first WS participant who comes up with a good filled in table will get a Swiss beer!
- teacher: His role is provide something between loose guidance & direct instruction. It can be a human agent (present or distant) or an intelligent agent. He provides information from the curriculum to the task level.
- monitor: The Monitor ensures that something is learner. A role taken by either the teacher, the learner (self-control) or by some program.
- fellow learners: Improve the learning process (some research tries to implement artificial ones)
- learning material: Learning material contains what has to be learned in a very broad sense (e.g. knowing what, knowing how). It can be computational in various ways (exploratory hypertext, lesson & task oriented hypertext, simulation software, task solving environments, etc.
- External info sources: All kinds of information which is not directly stored in the learning material (e.g. additional material, handbooks, manuals, etc.)
- tools: Everything which may help the learning process other then the learning material (e.g. calculators, communication software, etc.)
Table 1: Filling in the Functionalities of a Learning Environment
teacher monitor fellow learning Ext. info tools
learners material sources
Simple WWW partly
WWW with forms & badly
server side scripts (Q&A)
WWW with local data partly
processing clients (tutoring
WWW with intelligent partly
server-side computing (tutoring
computer or yes
virtual reality yes
- The question whether to fly with an Airbus at all, concerns more HCI but there is an interesting point concerning transfer of something learned: Would you drive a car without steering wheel (using sticks or something like that instead)?
Table of Contents Next Chapter