Book review
Learning by Doing
A comprehensive guide to simulations,
computer games and pedagogy in e-learning and other educational
experiences.
Edited by Clark Aldrich
Review by Donald Clark, Epic
Published by Pfeiffer 2005
It's heartening that experts such as Schank and Aldrich,
who have been pushing for the 'learn by doing' model for years,
are now in the ascendancy over the likes of Masie and Brandon-Hall,
who regurgitate whatever's being sold this year - LMSs, LCMSs, Reusable
Learning Objects and so on. Academic learning theory puts learning
by doing at the heart of the learning game. James, Dewey, Kolb,
Schank, and now Aldridge, are right to keep this flame alive against
those in education and training who see their job as simply loading
people up with knowledge.
Aldrich is an analyst turned simulations producer.
He understands that real performance needs real performance, not
just classroom lectures or knowledge-based e-learning and this book
is a case for the measured use of simulations to teach such skills.
It is in no way near being a definitive text, but it's a good effort.
The chatty style and plentiful diagrams make the book
very readable. Analysts need this skill, as their audience, investors,
are lazy readers. At times he gets carried away with abstruse classificatory
diagrams and background theory that has little to support it in
terms of actual evidence. In fact the book is woeful when it comes
to real evidence for the efficacy of simulations. There's data galore
on this issue, especially from the military, including models for
evaluating simulations. However, he tends to use material from people
and companies he knows, rather than the real world of actual simulations
work. This is a weakness.
Building and buying the right simulations in the corporations
and higher education today
The book moves nicely from the simple to the complex
and from present to future through a tapestry of simulations theory
and examples. This works well but the underlying taxonomy is, at
times, a little confused. He starts with his four main simulation
types:
- branching stories
- interactive spreadsheets
- game-based models
- virtual labs/products
This confuses the genres. Indeed, he makes some simple
category errors. Interactive spreadsheets use a specific tool (spreadsheets)
and labs are specific real-world places - neither are basic types
of simulations. How, for example, would non-lab based simulations
of physics and maths be classified? One could also argue that games
are different from simulations, leaving only branching stories as
a generic type. For a full discussion of the typology of simulations
see the Epic White Paper on Simulations
and e-learning and Games
and e-learning). Aldrich's book is much better when it discusses
real examples and techniques rather than this type of pedagogic
theory.
The Broader Opportunities for Simulations
In moving beyond the sort of simulations we see in training he
turns his attention to broader educational simulations and folds
in new types of content, which he classifies as systems, cyclical
and linear. I found these concepts somewhat confusing as they overlap
and are not wholly explained either by definition or example. This
is compounded by a confusing triangular classificatory diagram at
the end of the chapter under the heading of convergence. This is
followed by more muddled thinking under the guide of 'the three
essential elements to successful educational experiences' (simulations,
games and pedagogy). This is all very overblown and he completely
lost my interest with the pompous heading 'universal truths', which
are explained using a three ringed Venn diagram. If you're going
to play God, at least have a well-researched model.
Luckily he gets back on track with some chapters on
role-playing in simulations. This is useful as it has real applications
in single and multiplayer training simulations. However, the paucity
of examples turns it into a set of notes and reflections rather
than a serious treatment.
Chapters 13, 14 and 15 explore computer games but
only cover the basics. There's no real mention of Gee's work in
identifying the principles one can learn from games to apply in
learning and Prensky isn't mentioned at all. In truth, others have
a lot more to say about games and learning. Aldrich moves beyond
his area of expertise when he moves beyond simulations into the
world of console games. The chapter on mini-games, interestingly,
brings him back to his core area of expertise as these play a significant
role in learning, especially in education.
top
Next generation simulations
"If we make simulations accurate, they will be too
hard. If we make simulations easy, they will be irrelevant," he
says. I agree. This key issue is discussed using a real example;
First Flight - The Wrights Experience Flight Simulation. This simulation
was so accurate that the guy who successfully flew the 2003 centenary
flight on an exact replica used this simulation and made two flights.
The air force pilot who refused to use the simulation crashed on
his first attempt.
A second example, Virtual University, sponsored by
The Sloan Foundation, is a simulation that allows you to play a
role within a higher education establishment controlling variables
such as faculty, students, courses, sports, performance indicators
and so on. The first version was so real that it proved to be too
challenging. Version two added more pedagogic structure and support
for the learner leading to significant improvements. Anyone interested
in e-learning in Higher Education should check this out.
A good discussion follows on Full Spectrum Warrior
and Full Spectrum Command, the Xbox and PC games produced by the
military. FSW is a first person shooter that manipulates 30 entities
with lots of AI techniques. FSC takes things further controlling
200 entities. These, and other military games, such as America's
Army (amazingly - no detail on this in the book) have a lot to teach
us about how games can be used as simulations for training (see
Epic White paper on Defence
and e-learning).
Both show how the realism of games and the reality
of learning have to be balanced. Modelling, in itself, is often
not suitable for learning, which has to build momentum and provide
support as the learner moves from a state of ignorance to being
skilled. These two examples made you want to explore these issues
further.
Managing the simulation process
I expected more on the process of documentation and
prototyping but again the analyst came through with too much emphasis
on resource data. This varied from 1 person month to 30 person years,
with additional figures for second simulations, where, with the
right tools and processes in place, make the process faster and
cheaper. This is absolutely right. He is also spot on in hammering
home the message that scope creep is a huge risk. In many of the
examples he quotes, some of the complexity was unnecessary. His
point about the role of coaches as a form of learner support in
simulations is also helpful. He also brings in the whole underlying
rationale for simulations work, quoting Will Thalheimer, a military
trainer with a startling critique on traditional learning, "Most
forgetting happens quickly. Repetition is the most powerful of learning
factors. And by spacing that repetition over time, you can significantly
minimise that forgetting…Within forty days of leaving Fallon detachment,
a pilot's bombing accuracy returns to the accuracy he had just before
reporting to Fallon." This is an astounding piece of research. How
much training leads to precisely this zero improvement outcome?
A lot, I would guess. Simulations are one of the few ways in which
we can tackle this enormous drag on efficiency in learning. As Aldrich
notes, this will take courage and leadership.
A final plug for Appendix four. It should have been
in the main body text, as it has some excellent advanced techniques
for branching stories, such as node scores that add up to an overall
performance score, recording your voice, timers, subtle choice design,
interviewing panels and so on. Well worth reading.
Do it!
This book is a good start for those curious about
simulations and e-learning. It is a readable survey of most of the
major issues and has some nice comments and examples from Aldrich's
extensive contacts in the industry. It is not a book for those looking
for research based material, his work on taxonomy is weak and he
could have drawn on a wider set of examples from medicine and the
military, where the evidence for the efficacy of simulations has
been well proven. You'll learn plenty from this book but you'll
learn more from doing it!
|
