iTech Seminar

1
iTech Seminar

• David W. Brooks
• January 24, 2005

2
Todays Goals

1. Describe ICML.
2. Indicate what is known.
3. Relate what is know (tenuously) to neural
   structures.
4. Focus on what is NOT known.
5. Speculate.

3
Punch Line

• From an instructional design perspective, we
  dont know as much about motivation as we know
  about working memory.
• WM issues are becoming formulaic motivation is
  still quite hit-or-miss.

4
ICMLInteractive Compensatory Model of Learning
5
ICML Reference

• http//dwb.unl.edu/Chau/CompMod.html

6
Interactive, Compensatory

• Perhaps the biggest notion of the ICML is that we
  humans are not cast in stone or otherwise trapped
  by some genetic or early-life-event factors. Most
  folks can work out of a learning issue effort
  matters.

7
ICMLInteractive Compensatory Model of Learning
8
(No Transcript)
9
Best Predictor

• The best predictor of what one can learn is their
  current knowledge. This generally accounts for
  no less than one third of the variance, and
  often accounts for gt80 of variance.

10
Shapiro Paper

• A recent paper by Shapiro suggests that prior
  knowledge often is underestimated as a source of
  learning success.
• Shapiro, A. (2004). How including prior knowledge
  as a subject variable may change outcomes of
  learning research. American Educational Research
  Journal, 41(1), 159-189.

11
Ability (Intelligence)

• An enormous amount of hand-waving goes on related
  to intelligence. First, it is clear that
  intelligence does exist and does matter. Next,
  it also is clear that differences in intelligence
  do not impact on performance much among
  professionals suggesting that all can achieve
  appropriate levels of knowledge.

12
g and Working Memory

• A very recent paper suggests an essentially
  complete overlap between g and working memory.
  That is, measuring g gives one essentially all
  one needs to know about working memory. While
  certainly not a wonderful paper, this report does
  make sense.

13
Colom et al.

• Colom, R., Rebollo, I., Palacios, A.,
  Juan-Espinosa, M., Kyllonen, P. C. (2004).
  Working memory is (almost) perfectly predicted by
  g. Intelligence, 32, 277-296.

14
Design Overview -- SME

• One can summarize nearly ALL of design work along
  these lines.
• Content folks work in the knowledge box and
  figure out the sequence of steps needed to
  achieve understanding.
• Occasionally they add something about strategy
  instruction.

15
Design Overview -- IBE

• The designer pretty much works on lowering
  cognitive load to make material accessible. This
  is important when the material is difficult
  (possibly just new).

16
Conflicts -- SME vs IBE

• Conflicts often arise between SMEs and IBEs.
  These often come from issues where like the IBE
  takes out something that the SME thinks of as
  very interesting or important (even if a bit off
  topic).

17
The BIG Puzzles

• While this certainly is not meant to trivialize
  what is not known about knowledge structures and
  working memory, we know still less about --
• Motivation

18
Study 1--Zoo

• Working with Zoo animals, children seemed to give
  special respect not afforded to animals brought
  to classrooms by teachers.
• Trainin et al. showed this to be related to the
  uniqueness of the animal rather than some aura
  connected with the Zoo.
• Trainin, G., Wilson, K., Wickless, M., Brooks,
  D. (2005). Extraordinary Animals and Expository
  Writing Zoo in the Classroom. J. Sci. Educ.
  Technology, accepted.

19
Study 2--EE

• Advanced electrical engineering students found
  computer simulations of signals to be somehow
  fake and those of hardware simulators to be
  real.
• Reality versus Simulation, Srilekha Srinivasan,
  Lance C. Pérez, Robert D. Palmer, David W.
  Brooks, David Fowler (under revision)

20
Anecdote 1

• Enormous effort is expended in biology education
  to deal with the teaching of evolution.
• The point is that sometimes motivational issues
  become all-consuming.

21
TEAC vs EDPS

• I would assert that much of the difference in
  approach between TEAC and EDPS is in how one
  values feel good. Indeed, feel good is often a
  primary goal in TEAC.
• I think of qualitative research as feel-good
  research.

22
Knowledge

• Knowledge changes probably are based in changes
  in the cortex. These involve -- one way or
  another -- lowering energy transmission through
  the cortex. The better nuanced and articulated a
  neural net, the more nuanced and articulated can
  be a neural output.

23
Ericsson

• Ericssons work can be reduced to explanation in
  terms of articulating complex neural networks.
• Performance understood in terms of automation --
  which involves reducing the amount of energy
  required to support a performance.

24
Ericsson et al.

• Ericsson, K. A., Krampe, R. T., Tesch-Romer, C.
  (1993). The Role of Deliberate Practice in the
  Acquisition of Expert Performance. Psychological
  Rev., 100(3), 363-406.

25
Hippocampus

• The working memory is probably connected to the
  hippocampus.
• Involvement of the cortex is probably a confusion
  based upon use of the (frontal) cortex to store
  sequencing information.

26
Genetic / in Utero

• Working memory is probably determined by some
  combination of genetic and/or in utero factors.

27
Hypothalamus

• The tiny hypothalamus serves as the Health
  Maintenance Organization of the body, regulating
  its homeostasis, or stable state of equilibrium.
  The hypothalamus also generates behaviors
  involved in eating, drinking, general arousal,
  rage, aggression, embarrassment, escape from
  danger, pleasure and copulation.
• http//www.geocities.com/hhugs2001/roleofhyp.htm

28
(No Transcript)
29
Locus of Reward

• McClure, S. M., Laibson, D. I., Lowenstein, G.,
  Cohen, J. D. (2004). Separate Neural Systems
  Value Immediate and Delayed Monetary Rewards.
  Science, 306, 503-507.

30
Locus

• Immediate -- limbic system
• Delayed -- cortex

31
Two Approaches

• Ive found two approaches in the literature from
  which to try to tackle this issue.
• Self-efficacy (Bandura)
• Emotional Design (Norman)

32
Self-efficacy

• Bandura introduces a concept, goal challenge.

33
Self-efficacy

• While it might be possible to make a fit here,
  there is not a large enough base of work for that
  to seem efficient.

34
Emotional Design

• Norman suggests that we perceive designs on three
  levels
• Visceral
• Behavioral
• Reflective

35
Interesting, but

• While Normans discussions are interesting and
  bring up many issues, I think they beg a major
  point.
• Learning is cyclic todays struggle is
  tomorrow's automaticity.

36
Best way

• I think that the most efficient way to deal with
  this is that, when effects such as these (Zoo,
  EE) are encountered, address them by working
  explicitly to include knowledge about them as
  part of the desired learning outcomes. That is,
  establish cortex-based learning goals.
• In other words, take this off the table.

37
Zoo, EE

• In the case of the Zoo, perhaps teachers should
  indicate that scientists often are engaged with
  novel and interesting cases, but that ordinary
  cases require that same kind of study and
  observation.
• In the case of the EE, it probably is important
  to stress that being able to simulate is a virtue
  because of both ease and cost, and to develop
  some activities that make this point.

38
Evolution

• In the case of teaching content with deeper
  emotional connections, the problem may be MUCH
  larger.

39
Cognitive Load Revisited

• The notion of cognitive load sometimes is
  illustrated as follows.

40
Cooper Sketches
41
Three Suggestions

• I make three suggestions of how to think about
  these issues.

42
Emotional Load

• Add a factor called emotional load that takes up
  mental resources.
• This would compete for working memory.

43
Change the Mental Resources

• Perhaps some mental resources are reserved to
  deal with safety and security.
• Perhaps the amount of resources thrown at a
  problem can be changed by emotional factors.

44
Systemic Changes??
45
Testable

• It probably is possible to test among these
  possibilities.

46
Stress

• For example, learning is known to be most
  effective when the learner experiences a moderate
  amount of stress. No stress or too much stress
  are not as effective.

47
Cortex Learning

• Practice
• Practice
• Practice
• Of course, thats deliberate practice.

48
Ability

• Lower cognitive load. Adjust to available
  working memory. The less WM you need, the more
  effective your are likely to be.

49
Just Maybe

• Maybe the BEST way to address this is to say that
  motivational components compete for working
  memory slots.

50
Are We Ready?

• Are we ready to begin making explicit
  consideration of emotional factors in
  instructional design?
• Should we design and undertake experiments akin
  to those undertaken by Sweller (and Mayer)?