10 Feb 2005 ............... Length about 2,000 words (15,000 bytes).
This is a WWW document by Steve Draper, installed at http://www.psy.gla.ac.uk/~steve/3theories.html.
You may copy it. How to refer to it.
Web site logical path:
The relationship of the Perry, deep&shallow learning, and Laurillard models
Stephen W. Draper
Department of Psychology
Triggered by students' questions in a class in November 1997.
Three of the theories of the LTP (learning and teaching process) that I
view as important are the notion of deep and shallow learning (Marton et al.
1984), Laurillard's (1993) model, and Perry's (1968/70) model. In particular,
Marton and others contrast deep and shallow learning, or offer the idea of a
dimension of learning approaches; Perry argues for an ordered set of stages
(that some have simplified to 3 main stages); and Laurillard's model includes
a contrast between the level of public, formal descriptions and private,
personal experience, with a third set of "reflective" activities relating them.
The suggestion was: do these 3 layers in Laurillard's model correspond with the
3 main Perry stages; and Perry's stage A with shallow learning?
These links are probably significant associations in practice, but not
necessary or universal.
Laurillard's levels are to do with the kind of knowledge: public and personal,
expressed in formal language or in personal direct experiences. Someone who
learned only in the top level (of public, formal descriptions) would fail to
link it to personal experience; but might learn rival theories of one domain
and their relationships at that level (and so have a high Perry score), and
might think deeply about what those theories implied rather than how to repeat
key words (and so do deep rather than shallow learning). Conversely someone
could cover all the Laurillard levels to produce rounded knowledge of one topic
(e.g. Newton's laws), yet still think there is just one true theory in this
Deep and shallow learning is about the aim of the learner and the narrowness of
the aspect of what is learned: with shallow learning being about learning only
what is needed for some particular test task, rather than grounding it more
richly with relationships to other aspects. Thus shallow learning often is
about rote learning of technical terms without thinking about their meaning (if
the test task is multiple choice questions), it but could be about learning
procedures without learning the reason for them (if the test is a practical
one), or about learning to churn out a standard essay about pros and cons
without thinking about the connection of the topic to any new situation or
experience (e.g. to pass a history exam consisting of essays). A research
psychiatrist who never thought about how he could be mentally ill or what that
would feel like would be a deep learner with a high Perry score, but with at
best a limited connection to Laurillard's level of personal experience (he
might know what mentally ill people looked like externally).
Perry is about a naive faith in simple truth versus the ability to learn
concepts that might not be true, to learn and reason about supporting and
disconfirming evidence for them, and to come to a reasoned though provisional
view on which theory is best (so far). (In fact, this is an old aim of
education, particularly university education. It is obviously a general
intellectual skill, that should be valuable to the learner even when the
subject matter of their degree is not. Since Perry, people who argue
essentially this often cluster under a banner called "critical thinking"
(e.g. Kuhn; 1991), which
seems to me essentially the same idea, except that it stresses a general
cognitive skill and de-emphasises how it interacts with every piece of teaching
in every subject.) Adopting a high (position C) Perry approach means dealing
with concepts and theories that are not definitely true or false, holding them
simultaneously in mind, and reasoning about their relationship a) with each
other i.e. with rival theories, b) with the evidence in the literature, c) with
one's personal experience. Full coverage of the 12 Laurillard activities only
guarantees (c), not the other two. Thus you could possibly have a high Perry
score but not cover the Laurillard levels; and conversely you might cover the
Laurillard levels but meet only a good fit between experience and theory so
that a low Perry score was not challenged.
Thus in my view these three theories are addressing separate and independent
ideas about learning. However I would expect quite a strong association in
practice. For instance, integrating relevant personal experience with the
concepts you learn is likely to challenge the theory to some extent and push
for a high Perry score; and it is likely to make you wonder more about the
meaning of the theory in a wider sense, and so move you more towards an attempt
to understand, not merely do tests well i.e. to deep learning.
Going further: types of "depth"
The deep and shallow work is founded on
learners' differing "approaches to learning", basically whether their
goal is to understand (associated with deep learning) or to perform
well on some specific anticipated test task (shallow learning). As
such, it is an independent idea from Perry's and Laurillard's levels.
However it is possible to see it in another way as embracing those
This is because when we try to unpack what is meant by "understanding"
or deep learning, we must recognise that there is no end to (deep)
learning or understanding: there is always more that could be done,
because there are always more connections to make, more applications of
a concept to work out. We could list some of the different types of
connection any learner can seek out: different types of connection,
and so in a sense different kinds of "depth". (All are good, more is
better, but none still leaves you with something.)
- Concept to example(s). You read a concept and may feel you
understand it, but can you immediately give examples of it?
Given a suggested example, can you decide whether or not it is
a valid example?
[These are deductive connections.]
- Concept to personal experience. a) applied
to own feelings b) how it looks when you witness it in others.
How does the concept relate to your personal experience i.e.
can you think of examples you have experienced?
[These are connections between the two parts of the Laurillard model:
top and bottom, public/private, conceptual/experiential.]
- Concept to alternative theories of this topic.
What are the rival theories/models? (For this topic,
the topic is "theories of the learning and teaching process".
The 3 theories discussed here are independently offered, but as
this essay discusses, it seems they are not really rivals but
alternative partial truths that could probably be synthesised
into a single consistent whole. That is not accepted in the
literature, but seems to be a tenable position.)
[Pure Perry/critical thinking type links between alternative views of
- Status of the concept:
Who else thinks this? what groups of people? how certain is the
theory? what use is it?
Is this an accepted theory, a hypothesis, a useful commonsense
classification, or what? What evidence is there or could there
be? If you don't know the status, where would you go in the
literature to establish its status i.e. find the published
evidence or arguments?
[This is associated with Perry, but is the "meta-information": not
which is right, but what the state of the debate is in others' opinion.]
- Deductive consequences: what implications does a theory have? especially
when combined with other facts and theories you know.
Maths is essentially formal deduction, and in sciences many non-obvious
conclusions can eventually be established.
- Working out what a new concept or fact means in multiple perspectives or
representations. This may only be obviously important in some disciplines not
others. But examples might include: coordinate geometry (relating algebraic
equations with their graphs); in chemistry linking the formula to the molecule
and its motions; in considering abnormal psychology, linking for every case
or condition, the "medical" view of an organic disturbance drug therapy and
behaviour viewed as the patient being "not themselves" and a social view of
this being part of the human experience like sleep or flu or being angry with
feelings that mean something.
- Concept to contradictions or inconsistencies, real or apparent, with any
other facts and theories (in other areas).
I.e. this is not about explicit rival theories (as Perry stresses), but
thinking about hidden contradictions or implications of the idea being
studied. In some ways this is similar to the previous point about working out
the deductive consequences, but with more emphasis here on searching one's
memory for possibly related items, and less on calculation (deduction). What
other cases, laws, commonsense can you think of that seem in conflict with the
concept? (For instance, when learning the Laurillard model, a good first
reaction of this type would be to think of something you learned by a one-off
exposure e.g. someone tells you in a pub. Surely you didn't need the other 11
activities to learn it? Or is there a reconciliation e.g. it wasn't
education, perhaps you DID do activity 2: you immediately discussed it with
friends, perhaps you did learn it but didn't fully understand it as you would
if you did practical activities involving it, ....)
- Understanding, insight, enlightenment (relevance, validity):
what pre-existing basic questions does this concept address? Why might it be
interesting or useful? Is this a neat academic game limited to one
experimental paradigm, or does it explain a feature of everyday
life, or solve a real problem e.g. make sick people better?
To put it crudely: so what? who cares? this is about judging a
theory, not in its own terms, not for whether it is true, but by
In the end, "understanding" and deep learning probably mean making all
the connections you could to other things you know or experience, as
opposed to simply being able to reproduce the idea in the way it was
given you. There are numerous kinds of such connection, and the more
links you make, the longer you are likely to retain this knowledge, and
the more ways you may be able to use it. But making these connections
takes time: time to think, and perhaps time to discuss it with others
to prompt your thinking.
Applying this to yourself as a study technique, you could apply the
above list to every concept and topic you learn: ask yourself
for examples and comments of each type (and write down as an aid to memory).
You would benefit yourself, according to this theory, in every
case: you would remember it longer, understand it better, feel greater
interest, and be able to discuss it in more ways, more originally, with
more original examples in exams. In the end, how do we judge whether
someone else has understood something? Basically, first by whether
they can restate it in their own words; secondly we are most struck if
they repeat it with an example of their own that is valid, but which we
ourselves had not thought of much less mentioned to them. (This works
in counselling; it works in exams. It's a universal truth (or
confidence trick).) Of course it takes some investment of time, which
is another reason why shallow learning is widely adopted, and still
more widely adopted when teachers overload students.
The issue of different types of deep learning links appears in several places
in my web pages:
If the three theories (Perry, Laurillard (particularly the two levels of
concepts vs. personal experience), deep and shallow learning) are
independent ideas (as I believe they are) we should be able to think of cases
of all possible combinations of them. Here are some examples:
For more references
see this link
- High Perry, shallow learning. A "businesslike" psychology honours student,
who knows that for many topics she must discuss rival theories, and has a
standard essay format for this, but never stops to consider how any of this
applies in real situations, nor whether one might really decide between the
- Low Perry, deep learning. A serious student of basic physics or
chemistry, who reasonably enough does not expect to discover anything false
about the official theories, but does work hard at understanding their
application to everyday situations and to unusual cases not described in the
- Low Perry applied to Laurillard's conceptual level: assumes that theories
are true or false, and only one theory can be true.
- Low Perry applied to Laurillard's level of personal experience:
assumes that there can only be one right way of doing things (such as
swimming backstroke, doing an arm sling in first aid, boiling an egg, that
there is only one correct colour name for a given colour and any other opinion
is just wrong). This is often justified as "commonsense", meaning the speaker
only knows one way, and cannot think of any reason for it, other than they
have never considered any alternative.
- High Perry applied to Laurillard's conceptual level: discusses alternative
theories and their strengths and weaknesses (but not how they apply to
- High Perry applied to Laurillard's level of personal experience: can tie
multiple bandages and discuss the pros and cons of each, knows more than one
way to boil an egg, etc.
- Shallow learning applied to Laurillard's level of personal experience.
Can do first aid bandages in class, but not in the cold on a mountainside
without bandages of the size they trained on. Can only fry an egg in
the pan and cooker they are used to.
a diagram on Perry's view.
Kuhn, D. (1991) The skills of argument
(Cambridge University Press: Cambridge)
Laurillard, D. (1993) Rethinking university teaching: A framework
for the effective use of educational technology (Routledge: London).
Marton,F., D.Hounsell & N.Entwistle (1984) (eds.) The experience of
learning (Edinburgh: Scottish academic press)
Perry, W.G. (1968/70) Forms of intellectual and ethical development in the
college years (New York: Holt, Rhinehart and Winston)