topics: |  psychology | cognitive | learning | expertise | education

Review

For a book written by a committee, this is an amazingly cogent coherent and up-to-date account of the human learning process, focusing on how new ideas from cognitive science are changing how we understand human learning in general and how it impacts processes in education.

The title of the first chapter, "From speculations to science" recalls Quine's From Stimulus to Science, perhaps a deliberate reference mapping the of cognitive processes to the Carnnapian enterprise of constructing high-level concepts ultimately from many sensory relatum, with the notion of similarity as the guiding principle.

Here are some detailed excerpts from the fascinating chapter 2, dealing with the nature of expertise.

Excerpts

In a classic behaviorist study by Edward L. Thorndike (1913), hungry cats had
to learn to pull a string hanging in a "puzzle box" in order for a door to
open... Thorndike concluded that the cats did not think about how to escape
and then do it; instead, they engaged in trial-and- error behavior...
Sometimes a cat in the puzzle box accidentally pulled the strings while
playing and the door opened, allowing the cat to escape.  But this event did
not appear to produce an insight on the part of the cat because, when placed
in the puzzle box again, the cat did not immediately pull the string to
escape. Instead, it took a number of trials for the cats to learn through
trial and error.  p.6-8

   When put into the box, the cat would show evident signs of discomfort and
   impulse to escape from confinement. It tries to squeeze through any
   opening; it claws and bites at the wire; it thrusts its paws out through
   any opening and claws at everything it reaches. . . . It does not pay very
   much attention to the food outside but seems simply to strive
   instinctively to escape from confinement. . . .  

   The cat that is clawing all over the box in her impulsive struggle
   will probably claw the string or loop or button so as to open the
   door. And gradually all the other unsuccessful impulses will be
   stamped out and the particular impulse leading to the successful act
   will be stamped in by the resulting pleasure, until, after many
   trials, the cat will, when put in the box, immediately claw the
   button or loop in a definite way" (Thorndike, 1913:13).

	
	initially, the cat claws all over the box trying
	to get out, until it accidentally pulls the loop.
	after many trials, it learns to immediately pull
	the loop.
	image: teorije-ucenja.zesoi.fer.hr

* link: Chapter 2 from Thorndike's Animal Intelligence (1911)
	http://psychclassics.yorku.ca/Thorndike/Animal/chap2.htm

Thorndike's main conclusion: learning is incremental and not
insightful. 

Turning conscious into subconscious

Of course, how cats learn to escape is not very different 
from much of human experience.
Here is a video of a baby trying to turn over: 


For most tasks where we have gained expertise - bicycling, typing,
talking, playing chess -  initially there is a period when we are
flailing around, looking for the right moves.  Or we are thinking hard,
deliberating - and acting very very slowly.  Eventually, it all becomes
effortless and instinctive - we just know what to do without thinking. 

Gaining expertise means turning the conscious into subconscious.

Knowledge and problem solving

Research on expertise in areas such as chess, history, science, and
mathematics demonstrate that experts’ abilities to think and solve problems
depend strongly on a rich body of knowledge about subject matter (e.g., Chase
and Simon, 1973; Chi et al., 1981; deGroot, 1965).

 [Argues for processes that lead to deeper understanding]

Even young infants are active learners who bring a point of view to the
learning setting. The world they enter is not a "booming, buzzing
confusion" (James, 1890), where every stimulus is equally salient. Instead,
an infant's brain gives precedence to certain kinds of information:
language, basic concepts of number, physical properties, and the movement
of animate and inanimate objects. In the most general sense, the
contemporary view of learning is that people construct new knowledge and
understandings based on what they already know and believe (e.g., Cobb,
1994; Piaget, 1952, 1973a,b, 1977, 1978; Vygotsky, 1962, 1978). 

A classic children's book illustrates this point:


   Fish Is Fish (Lionni, 1970) describes a fish who is keenly interested
   in learning about what happens on land, but the fish cannot explore land
   because it can only breathe in water. It has a friend, a tadpole who grows
   into a frog and eventually goes out onto the land. The frog returns to
   the pond a few weeks later and 
   describes all kinds of things like birds, cows, and people. The book
   shows pictures of the fish's representations of each of these
   descriptions: each is a fish-like form that is slightly adapted to
   accommodate the frog's descriptions— people are imagined to be fish who
   walk on their tailfins, birds are fish with wings, cows are fish with
   udders. This tale illustrates both the creative opportunities and
   dangers inherent in the fact that people construct new knowledge based
   on their current knowledge.


---from Fish is a Fish-- 

    “I have been about the world — hopping here and there,” said the frog,
    “and I have seen extraordinary things.”

    “Like what?” asked the fish.

    “Birds,” said the frog mysteriously. “Birds!” 

    And he told the fish about the birds, who had wings, and two legs,
    and many, many colors. 

    As the frog talked, his friend saw the birds fly through his mind
    like large feathered fish.



the fish imagines a bird from the frogs description - 

    "What else?" asked the fish impatiently. 

    "Cows!" said the frog, "Cows have horns, four legs, they eat grass
    and carry pink bags of milk."


fish imagines a cow. 


links: Fish is a Fish animated film: 
https://vimeo.com/39374062

Constructivist models of learning

A common misconception regarding "constructivist" theories of knowing
(that existing knowledge is used to build new knowledge) is that teachers
should never tell students anything directly but, instead, should always
allow them to construct knowledge for themselves....
Constructivists assume that all knowledge is constructed from previous
knowledge, irrespective of how one is taught (e.g., Cobb, 1994)
— even listening to
a lecture involves active attempts to construct new knowledge.
...
there are times, usually after people have first grappled with issues on
their own, that "teaching by telling" can work extremely well (e.g.,
Schwartz and Bransford, 1998).

sixth graders in a suburban school who were given inquiry-based
physics instruction were shown to do better on conceptual physics problems
than eleventh and twelfth grade physics students taught by conventional
methods in the same school system. A second study comparing seventh-
ninth grade urban students with the eleventh and twelfth grade suburban
physics students again showed that the younger students, taught by the
inquiry-based approach, had a better grasp of the fundamental principles of
physics (White and Frederickson, 1997, 1998).

New curricula for young children have also demonstrated results that are
extremely promising: for example, a new approach to teaching geometry helped
second-grade children learn to represent and visualize three-dimensional
forms in ways that exceeded the skills of a comparison group of undergraduate
students at a leading university (Lehrer and Chazan, 1998). Similarly, young
children have been taught to demonstrate powerful forms of early geometry
generalizations (Lehrer and Chazan, 1998) and generalizations about science
(Schauble et al., 1995; Warren and Rosebery, 1996).

(Scardamalia and Bereiter, 1991):
Teacher's A and B give projects - which could be anything from old-style
workbook activities to the trendiest of space-age projects, and
supervise the work (A) and also monitor the learning (B).

Teacher C does this as well, but continually turns more of the learning
process over to the students.  In one striking example of a Teacher C
classroom, the students had been studying cockroaches and had learned so
much from their reading and observation that they wanted to share it with
the rest of the school; the production of a video came about to achieve
that purpose (Lamon et al., 1997).

Pre-existing concepts: Initial understandings

Research on early learning suggests that the process of making sense of the
world begins at a very young age. Children begin in preschool years to
develop sophisticated understandings (whether accurate or not) of the
phenomena around them (Wellman, 1990). Those initial understandings can have
a powerful effect on the integration of new concepts and information.
Sometimes those understandings are accurate, providing a foundation for
building new knowledge. But sometimes they are inaccurate (Carey and Gelman,
1991). In science, students often have misconceptions of physical properties
that cannot be easily observed. In humanities, their preconceptions often
include stereotypes or simplifications, as when history is understood as a
struggle between good guys and bad guys (Gardner, 1991). A critical feature
of effective teaching is that it elicits from students their pre-existing
understanding of the subject matter to be taught and provides opportunities
to build on—or challenge—the initial understanding. James Minstrell, a high
school physics teacher, describes the process as follows (Minstrell, 1989:
130-131):

       Students’ initial ideas about mechanics are like strands of yarn, some
       unconnected, some loosely interwoven. The act of instruction can be
       viewed as helping the students unravel individual strands of belief,
       label them, and then weave them into a fabric of more complete
       understanding. Rather than denying the relevancy of a belief, teachers
       might do better by helping students differentiate their present ideas
       from and integrate them into conceptual beliefs more like those of
       scientists.

The understandings that children bring to the classroom can already be quite
powerful in the early grades. For example, some children have been found to
hold onto their preconception of a flat earth by imagining a round earth to
be shaped like a pancake (Vosniadou and Brewer, 1989). This construction of a
new understanding is guided by a model of the earth that helps the child
explain how people can stand or walk on its surface. Many young children have
trouble giving up the notion that one-eighth is greater than one-fourth,
because 8 is more than 4 (Gelman and Gallistel, 1978). If children were blank
slates, telling them that the earth is round or that one-fourth is greater
than one-eighth would be adequate. But since they already have ideas about
the earth and about numbers, those ideas must be directly addressed in order
to transform or expand them.

Chapter 2: How experts differ from novices

experts in an area : able to think efectively about problems in areas.
not general abilities like memory or intelligence, nor general strategies.
Extensive knowledge that affects what they notice and how they organize,
represent, and interpret information --> affects how they remember and
reason abt problems.

experts in chess, physics, maths, electronics, history:

    1. Notice features and meaningful patterns not noticed by others

    2. have acquired a great deal of content knowledge org'd in ways that
	reflect a deep understanding of the subject

    3. this knowledge cannot be reduced to some isolated facts or
	propositions but reflect contexts of applicability -
	"conditionalized" on the circumstances

    4. able to flexibly retrieve important aspects of their knowledge with
	little attentional effort

    5. may be poor at teaching, though they know the discipline thoroughly

    6. approach to new situations requiring flexibility - varies widely p.31

Meaningful patterns of information : Chess expertise

DeGroot 65 - hypothesized that chess experts would consider all possibilities
- greater breadth of search, and consider more countermoves (greater
depth). But found that masters did not exhibit greater depth or breadth than
lesser players.  Both lesser and masters did not cover all possibilities.
But somehow masters  considered possibilities for moves of higher quality
than those considered by weaker players.  Masters were more likely to recog
meaningful chess configurations and realize the strategic implications of
these situations; this recognition allowed them to consider superior
moves

	We know that increeasing experience and knowl in a sp field (chess,
	for instance) has the effect that things (properties, etc) which, at
	earlier stages, had to be abstracted, or even inferrred, are apt to
	be immediately perceived... abstraction is replaced by perception -
	but we do not know much abt how this works, nor where the borderline
	lies.  as a result, a so called "given" problem situation is not
	really given since it is seen differently by an expert than by an
	inexperienced person.

based on think-aloud protocols, cross-validated w other methodologies. (see
Ericsson and Simon 1993).   p.32


[from Chabris/Simons: theinvisiblegorilla.com:

	Some of the details of de Groot’s claims, which he made before the
	appropriate statistical tests were in widespread use, did not hold
	up to later scrutiny—experts do consider somewhat more options,
	look a bit deeper, and process positions faster than less expert
	players (Holding, 1992). But de Groot was right about the limited
	nature of expert search and the importance of knowledge and
	pattern recognition in expert performance.

---

people perceive chunks of information which affects their memory for
what they see. 
Chess masters are able to chunk together several chess pieces in a
configuration based on some strategic component of the game.  10 and 11 year
olds more exp in chess are able to remember more chess pieces than coll
students who are not chess players, though coll students may be better at
other stimuli, like remembering strings of numbers.  (Chi 1978, Schneider
1993 - see fig 2.3).

[BOX: expt: 5 seconds viewing of chess position from mid-game; then subj
attempts
to recreate the position on separate board.  Trials repeated until full
board.  On 1st trial, master player put 16 pcs correct, class A player 8, and
novice 4.  But if chess posn shown was a random jumble of pcs, the recall of
master was same as novice/class A.  from Chase and Simon 73

expts on children w chess experience vs adults - Chi:78]

Other domains of expertise

similar skills demo'd for experts in electronic circuitry (Egan/Schwartz:
1979), radiology (LEsgold 1988), computer programming (Ehrlich and Soloway
1984).

expertise = sensitivity to patterns of meaningful information - not
avlbl to novices.

Expert circuit designers chunked several circuit elements (e.g
resistors and capacitors) after only a few seconds of viewing; novices could
not.  e.g. "amplifier chunk".

Math experts: recognize patterns of information, such as particular
problem types that involve specific classes of math solutions (Hinsley
etal 1977, Robinson and Hayes 78).

e.g.
   Board sawed into two pieces.  One pc 2/3ds as long as the whole board, and
   was exceeded in length by the 2nd piece by 4 ft. how long was board
   before it was cut?

experts: quickly realize prob is impossible
novices: apply eqns and get -ve length (some students catch it also)

There are 26 sheep and 10 goats on a ship.  How old is the captain?

	5th grade child: after giving an answer of 36, said: 
		Well you need to add or subtract or multiply in problems
		like this, and this one seemed to work best if I add"
		[Bransford and Stein 1993:196]

	More than 3/4ths of school children in a study attempted to
	provide a num solution.

Teaching scenario

Expert and novice teachers shown videotaped classroom lesson
(Sabers etal 1991)
three screens show sim events in classroom (left, center, right) - experts
and novices talk aloud about what they see.  Later, asked q's abt the
scene.

expert 8:                                 | novice 1:
On the left monitor, the students         | can't tell what they're doing
note taking indicates they've seen        | they're getting ready for class,
sheets like this and presentations        | but I can't tell what they're doing
like this before.  its fairly efficient   |
at this point because they're used        |
to the format they are using              |
                                          |
expert 7: I don't understand why          | novice 3:
the students cant be finding out this     |        she's trying to communicate
info on their own rather than listen      | w them here abt somthing
to someone tell them - if you watch       | but i sure couldn't tell what
the faces of most of them they start      | it was.
out for abt the first 2-3 mins sort of    |
paying attention then drift off           |        novice N: it's a lot to watch
                                          |
expert 2: i haven't heard a bell, but     |
the students are already at their desks   |
doing some purposeful activity - and      |
this is abt the time that I decide they   |
must be an accelerated group because they |
came into the room and started sth rather |
than just sitting down and socializing    |

Learning: when viewing informational texts, slides and video, info noticed by
novices can be quite diff from that of experts [Sabers etal 1991, Bransford
etal 1988]

One dimension of acquiring competence = incr ability to segment the
perceptual field (learning how to see).
Pedagogy: students need to enhance ability to recog meaningful patterns of
info [Simon 80]

Organization of expert knowledge

not just a list of facts and formulae relevant to the domain, but around core
concepts or "big ideas" that guide their thinking of these domains.

Physics problem solving

Physicsts recog problems of river currents and headwinds and airplanes as
involving similar mathem principles, such as relative vels.  The expert
knowledge that underlies the ability to recog problem types has been
characterized as involving the devlpmt of org'd conceptual structures, or
schemas, that guide how problems are repr'd and understood (e.g. Glasser and
Chi, 1988).

experts: mention major principles or laws that were applicable and a
	 rationale for why those laws applied to the problem and how to apply
	 the laws. [Chi etal 1981]
competent beginners:  which equations they would use and how these eqns would
	 be manipulated.

also experts often pause to draw simple qualitative diagrams - don't attempt
to just plug numbers into a formula.
diagram is then often elaborated as the expert seeks to find a workable
solution path

Sorting physics problems given in index cards

experts: based on principles (e.g. cons of energy)
novices: problems w inclined planes [more surface characteristics; not v
	useful since may involve completely diff principles]

nature of schema in physics [Chi 82] - inclined plane:
novice - surface features ("block on incl plane", "coeff of friction")
expert - links w laws of physics and conditions when these laws apply
       ("conservn of energy", "either you shd know cons of e, or work is lost
       somewhere.")

Timing of pauses in responses:
experts - one eqn quickly activates related equations
novices - retrieve eqns more equally spaced in time - suggesting sequential
	access.

experts - more efficient organization of knowledge, clusters of meaningful
relations (chunks)
"knowing more" = more conceptual chunks, more relations or features for each
	chunk and its interrelations. [rather than more facts]

Experts in History

[Wineburg 1991]:
History experts vs
high-achieving high school seniors (from AP history course):

given a list of facts from US history.  Some historians knew most of the
items; others specializing in other areas knew only 1/3d of the facts.
Several students scored > some historians on knowledge of the facts.

The study then compared how the historians and students made sense of
historical documents - results revealed dramatic diffs on almost all
criteria.

when asked to choose which one of three paintings was an image of the batttle
of lexington, students generally looked at the pics and then made a choice
(as if in a multi-choice test).
Experts went back and forth between the documents, carefully analyzing
details, to make a choice.

Task 2 [Wineburg 1991]
historians and future history teachers asked to read and interpret a set of
documents about lincoln's view of slavery.  his writings reflect many
conflicts and contradictions and is a complex situation.  One historian was
an expert and interpreted them readily.  Another historian (Asia expert) and
teacher trainees were confused and both appealed to knowledge that is more
current - speech writers, press conferences, "spin doctors" etc.  But 2nd
historian took this as working hypothesis and went on reading the documents,
trying to understand the 19th c view - and eventually emerged with an
interpretive structure that brought him at tasks end, close to where his more
knowledgeable colleague had started.  (p. 47)

Summary: though students scored well on facts abt history, they had no
systematic way of making sense of contradictory claims or formulating
reasoned interpretnations.

Retrieval

retrieval may be "effortful", "fluent", or "automatic"
[Schneider-Shiffrin-1977] - latter two assoc w expertise

despite fluency, experts may take more time in solving problems - because
they want to understand it first.  [Getzels - Csikszentmihalyi 1976].
But fluency is important because it places little demand on conscious
attention.
Ease of processing one aspect relieves capacity for other tasks. 44

e.g. driving a car... novices can't drive and converse.
     novice readers : so focused on what they are reading, that they can't
	     worry abt understanding

developing fluency v imp in instructional environments

Third Intl Math and Sci Society (TIMSS [Schmidt etal 1977]) criticized
curricula "a mile wide and an inch deep" - much more of a problem in the US
than elsewhere.

experts: what part of knowledge would be relevant to a problem?  search
efficiently based on high-level concepts, not burdened by details that may
overwhelm working memory [Miller 1956]

expert knowledge is "conditionalized" - includes spec of contexts where it is
useful [Simon 80; Glaser 92]
conditionalized knowledge is often "inert" - not activated even though it may
be relevant [Whitehead 1929]

Adaptive experts: aware of own limitations (metacognition - q. own knowledge)
	- expert is not one who "knows all the answers". Adaptive experts
	keep learning through lifetime. [Hatano and Inagaki 1986]

Chapter 4: How children learn

long-held theories that infants lack the ability to form complex ideas
(tabula rasa) have been abandoned (Piaget's sensorimotor infant onwards)
with mounting evidence, using carefully designed methods, that showed
infants to as competent active learning agents in charge of their own
conceptual development [Bruner 1972, Bruner-1981, Carey and Gelman 1991,
Gardner 1991, Gelman-Brown 1986, Wellman-Gelman-1992]
  [Bruner 1972] Toward a sense of community Sat Rev (review of Gentner etal
  [Bruner 1971], "Children teach  children"
  [Bruner-1981] Intention in the structure of action and interaction, in book

A major move away from the tabula ras view of the infant mind was Piaget's
theory (beginning 1920) of cog devpmtl stages, w initial repr of objects,
space, time, causality, and self constructed gradually during first 2
years.

patterns in the perceptual space [Gibson 1969] - though differing consid from
Piaget, shares notion of children as active learners.  early concepts incl
biological hierarchies, number sense, basic physics; also cogn strategies for
remembering understanding, and solving problems.

Active role of learners emphasized by [Vygotsky 1978] - pointed to role of
environmt incl tools and cult objects - "zone of proximal development" -
bandwidth of competence that learners can navigate w help from a supportive
context, which incl caregivers. (see [Newman etal 1989] for a modern formulation)

Zone of Proximal Development

distance betn actual developmental level (ability to solve problems
independently ) and level of potential development (problem solving under
adult supervision).  What children can do with assistance is more indicative
of their mental development.  [Vygotsky 1978:85]

ZPD = concept of readiness to learn - emphasizing upper levvels of competence
- these boundaries constantly change w learners increasing indiv competence.

Native Biases

- Native Biases: Predisposition to learn abt some things and not others -
	privileged domains: e.g. physical and biological concepts,
	causality, number, and lg [Carey and Gelman 1991]
- Strategies and Metacognition: effort to learn, awareness of degree of
	learning. [Brown-deLoache 1978, deLoache-etal-1998]
- Theories of Mind: what it means to learn and understand - how to situate
	themselves in an intentional learning scenario
	[bereiter-scardamalia-1989], various t-o-m's [dweck-legget-1988].
- community: incl adults, but also other children - community of learners

to contribute some excerpts from your favourite book to book excerptise. send us a plain text file with page-numbered extracts from your favourite book. You can preface your extracts with a short review.
email to (bookexcerptise [at] gmail [dot] com).

We reply to all feedback!