Pedagogy og technology

Piaget

[ A Pedagogical Foundation ] [ Logo ] [ Nordic Experiences ]
[ Evaluating Results ]

The difference from the Skinner period is the intention of making open learning situations. The student or pupil should have the freedom to act, experience and learn after his or hers own plan. The programs should be open and without sequences. This approach set the focus on motivation, while control is getting less important.

The openness does of course rely on the technology of the period which allowed graphical interfaces with dynamic menus, multiple windows, direct manipulation, access to many programs at the same time etc.

So what has this to do with Piaget ? The arguments are well put by Alan Kay. Kay is in many ways the main architect behind graphical user interfaces as we know them today, to the extent that it is possible to point at one person. Kay's reasoning, as it is put forward in a video distributed lesson [7], is basis for parts of the arguments in this section. In addition I discuss briefly an other interesting approach, Logo, although this as pointed out above is older than the Macintosh. I do also discuss some of the experiments and results from the Nordic approach in the 80's.

A Pedagogical Foundation

The key sentence in Kay's contribution is: Doing with Images makes Symbols. The cornerstone is Piaget's theory of a child's development, which (too coarsely) can be illustrated like this according to Kay's sentence:

Bodily experience (Doing)	Images (Images)	Symbols (Symbols)
Early development	Childhood	Grown up

We usually think of this as three stages in a development, and consider the symbolic phase as the normal adult way of operating. We tend to think that we should strive to be as operable in this phase as early as possible, and that we do not need the earlier phases for intellectual work.

Kay claims that all people, regardless of age, skill and task, will gain by communicating on all this "channels" simultaneously. He finds support for this view in surveys and observations that show that people which we consider theoretically, or symbolically, competent, often thinks in images and often has an almost physical experience of their models.

Einstein is often mentioned as an example of a scientist who reasoned in images. An other example close at hand is chess masters who are not able to give explicit logical arguments for their choice of the next move. In older technology history Maxwell and Faraday are often mentioned as an example of two complementary approaches to the same domain. Faraday as an exponent for the intuitive approach and Maxwell for the descriptive. See Bernal [8].

Kay's sentence is clearly illustrated and realised in graphical user interfaces as we know them on todays computers, with the possibility for direct manipulation of images.

One can always ask whether this type of reasoning actually has been an explicit basis for the development of user interfaces, or if is an after rationalisation. This is not so important. The important issue is that it gives us a foothold for extending the reasoning to learning environments.

Logo

An other person who has contributed to the understanding of the dependence between technology and pedagogy is Seymour Papert [6]. His Logo was developed before 1984 on simpler technology than what is typical for the period. Logo attracted a lot of interest, also in the Nordic countries, i the first part of the 80's. Papert's work is explicitly based on Piaget's theories.

The metaphor for Logo is a turtle that can be commanded to walk around and leave a track.

forward 100 right 90 forward 100 to square forward 100 right 90 forward 100 right 90 forward 100 right 90 forward 100 end square Repeat 4 times square end

There are two main points in Papert's reference to Piaget.

The first is a direct parallel to Kay's reasoning. The user should so to speak identify herself with the turtle, and "do" the geometry before and after it is written as commands. It was even produced "turtles" who could be commanded to walk and draw on the floor so you could literally plan the track by going ahead.

The second is the possibility to make aggregate "verbs" that can be reused. This is an approach to support the development of concepts and understanding. It should be possible to build a powerful set of concepts for different purposes. Papert coined the phrase "microworlds" for such sets of concepts..

Logo was very popular for a period of time. To many pedagogues Logo was the first computer program which was explicitly based on a pedagogical idea. Logo was easy to learn and created enthusiasm in the classroom for a period. It was however some problems that appeared rather fast. For most users, children and teachers, the source of ideas dried up after a few hours. The microworlds became too small and too difficult to expand. It was difficult to design a teaching plan that lasted for a longer period. Logo was an interpreted Lisp-dialect and was a complete programming language with tools for treating lists as well as geometry. It was however few, if any, teachers or pupils who took the time and effort to learn other concepts of the language than the initial geometrical commands.

Nordic Experiences

What we have called the Piaget period coincides in time with a period of a planned and rather powerful effort for developing computer based learning material in the Nordic countries. A summary of some of he work done in this period is found at the IDUN centre in Copenhagen[9].

Important components in the Nordic effort in this period was the development of software and the education of teachers. National and Nordic design seminars were interesting laboratories for developing, understanding and testing tools and methods. The main method for design that was used is thoroughly documented in the book Brukerorientert Programdesign [10].

A central concept in the methodical approach was metaphors. It was evident that building an environment for learning had to be based on an understandable and motivating environment. Such metaphors for learning were of many types:

• Extensions of the desktop metaphor.
• Tool metaphors for laboratory-like programs.
• Game metaphors.
• Metaphors based on time, place, map and clock/calender.
• Theme specific metaphors that created strong associations to worlds existing outside the computer.

Common to all these approaches was that they described a limited, controllable world. Even though the structures within this worlds were relatively open, they were planned and controlled by the designer and had an absolute barrier against the rest of the world.

Dynamic simulation

Dynamic simulation is one of many areas that was developed as a tool for learning in this period. The area can serve as a good example of both possibilities and limitations. The group of pedagogues at Stord Haugesund College[11] was the main contributor to this development. The software and the ideas have been developed further and commercialised by PowerSim [12]. The concept build on a simple metaphor which mainly consists of valves to control flow and containers to accumulate flow.

Modelling of the temperature fall in a cup of coffee can be illustrated like this:

The result of running the model, the simulation, can be illustrated like this::

This tool is apparently very simple, flexible and useful for analyses of a wide rang of problems. The ambitions of those who get in touch with the system grows fast. It turns out to very difficult for untrained designers to estimate the complexity of a model. The wish to take on real world problems, in say ecology, almost without exception led to incomprehensive or oversimplified models. This was due to lack of domain knowledge or lack of competence in general system dynamics. The problems with interpreting the results are of course difficult, and will in many cases lead to erroneous conclusions.

The interesting aspect is however the process of building the model. Used on a model with limited complexity and a relatively clear interface to the surroundings, the learning involved in building the model is an interesting pedagogical endeavour. It does however, as is the case with Logo, presuppose a certain level of mastering, creativity and insight from the teacher.

Problems

A general problem associated with the use microworlds of different types is that they have a tendency to become too simple seen from a learning point of view and to difficult from a users point of view. It takes a lot of design and programming effort to include easily available flexibility to the extent that the product can be useful in a longer period of time. Many of the products have a lifespan of a couple of hours in the classroom before they are exhaustively investigated and further use tends to be boring and predictable repetitions.

General, extendible metaphors, like Logo and PowerSim, are a challenge to teachers that too often is not met. The assumed pedagogical value is not considered great enough to invest the necessary time in planning, understanding and mastering the tool. This assumption may very well be wrong, but it is common.

An example of an attempt to meet this challenge is the program suite "Vi på vindusrekka" [13]. This is a series of programs that is designed to be useful in mathematics education in a three year period for children in the age 10-13. The programs are laboratories, simple games, assignment generators and hypertextual texts for explaining central themes in the curriculum. The programs are designed for different purposes in a long term pedagogical plan. This concept runs into similar problems as those mentioned above. The threshold for evaluating the program suite as a useful tool is to large and the programs are at best used at random with very limited effect.

The programs with the longest lifetime in the classroom are probably simple game-like programs based on drill and repetition, programs with a rather simple pedagogy and which can benefit from an advanced technology.

Evaluating Results

Surprisingly small efforts have been made to evaluate the effect on learning of the products developed during the Piaget period. There may bee many reasons for this. One obvious reason is that we do not have good methods to carry out such evaluations. An other reason is that most programs have a too narrow learning objective.

Some have asked for this kind of evaluation to justify investments in technology in schools, while others have emphasised the meta learning that takes place when children get used to the technology as a tool. The latter point of view has been dominant in this period.

Three problems are evident in evaluating learning effects from use of technology:

1. The obvious Hawthorne effect when technology is introduced in a learning environment. As long as technology is considered a scarce commodity, the access to technology will be a motivation for increased activity, and learning.
2. The effect that that takes place when a situation is established in a way which is not feasible as a daily situation. There are examples of experiments where a combination of technology, staff, planning and working conditions gives astonishing results. The reported experiments with Smalltalk at Xerox Parc is one example, see for instance [7]. Own experiments with software design seminars with 16 year olds is an other example [14]. (What can be learned from this, is of course that it is possible to achieve results if you are willing to pay the costs)
3. The results are connected to a certain product or a certain technology which are outdated at the time the results are published. This weakens the interest in the results, and impede the sharing of experiences.

It is my general impression that the learning effect of the pedagogical software on the market to day, as it is used in day to day work in the school society, is low. Some of the reasons for this is discussed above. The access to technology, and the way the access is organised, is also an obvious reason.