Tel-Aviv University | School of Education | satec | ktl
Early development of
technological/engineering stance by
Kindergarten children
Children perceptions of artifacts with adaptive
behaviour
David Mioduser & Asi Kuperman
1
Rationale: perspectives and definitions
☞On studying the encounter between
children and the artificial world:
‘curricular perspective’: technological
[literacy, content knowledge, PCK,
curricular differentiation/integration,
STEM, ...] ‘with the child in mind’
cognitive perspective:
technological/engineering/design ‘stance’
☞‘Design Stance’ [Dennett, 1987]
☞‘Human intelligence and Technology
[Sternberg; Cole and Derry, 2002]
☞‘Intuitive engineering’ [Pinker, 2002]
2
Rationale: perspectives and definitions
‘Intuitive engineering’ - [Pinker, 2002]
“the world is an heterogeneous place, and we are equipped with different
kinds of intuitions and logics, each appropriate to one department of
reality ... intuitive physics ... intuitive biology ... spatial sense ... number
sense ... mental database and logics ... language ... [and]... intuitive
engineering - which we use to understand artifacts - [objects] with a
purpose, designed by a person to achieve a goal”
design stance [Dennett, 1987]
“an abstract explanatory schema that captures the relationship between
features of an entity (e.g. its material, shape and activities) in terms of a
coherent organizing notion: the purpose for which its designer created it”
what is innate ?
what develops?
when "it" develops?
what conditions support "its" development [education question]
3
not apples but artificial minds
4
artificial minds
☞
A robot as many microprocessor-based artifacts in our
everyday environment, is a unique artifact: it is
characterized by purposeful functioning and
autonomous decision-making (it 'behaves'?),
programmability and knowledge accumulation
capabilities (it 'learns'?), and adaptive behavior (it
'makes decisions'?)
☞
This new category of creatures affects the traditional
and intuitive distinctions between the alive and notalive, animate and inanimate, human-operated and
autonomous.
☞
5
natural & artificial minds
research questions
☞ Question 1:
What are kindergarten children's perceptions of
programmable adaptive artifacts in terms of the
stance adopted (i.e., engineering vs.
psychological)?
☞ Question 2:
Do these perceptions vary as a function of the
complexity of the task and involvement in
programming the artifact's behaviour?
6
method
☞ Participants: 10 children, 5 boys and 5 girls, age ranging
from 5:4 years to 6:3 years, arbitrarily chosen from a
group of 25 children attending a kindergarten of average
socio-economic status in the central region in Israel.
☞ Instruments: the robotic environment (programming
interface and physical robot) and a progression of tasks
of increasing complexity.
☞ Procedure: Data collection lasted two months. All
sessions and interviews took place in the kindergarten's
robotics corner and were videotaped.
7
setting
8
Question 1: explanatory stance
Explanatory language
Explanations
Definition
Examples of children's
explanations
Use of
anthropomorphic
language
Robot's behaviour is
explained in terms of
intentions, volition,
feelings and humanlike actions
"… He sees that it is the
sea and decides to
turn…"
"… If he sees a person
then he has to tell
him…"
Use of technological
language
Robot's behaviour is
explained in terms of
its components'
functions,
mechanisms, and
formal decision-making
rules
"… we simply wrote
[programmed], when he
gets to the black area he
stops and when in the
white area turns back…"
"… and if one [sensor]
sees white and the other
sees black then [turn]
left…"
9
Question 1: explanatory stance
Statements
N=684
Anthropomorphic language
107 (16 %)
Technological language
577 (84 %)
☞ Predominance of the technological/engineering stance
☞ Example of typical (expected in age level) use of
anthropomorphic language:
"He's walking only on the white area because it feels warm … he
wears a hat and he knows that he is wearing the hat"
☞ Functional use of anthropomorphic language when children felt
that is perfectly natural to use human-related terms for an
explanation, even if they are explicitly aware that they report
about an artifact's behaviour
10
Question 2: explanatory stance by activity or task
Activity
Anthropomorphic language
Technological language
Observation (N=107)
54 (50 %)
53 (50 %)
Construction (N=577)
194 (34 %)
383 (66 %)
Task complexity
Anthropomorphic language
Technological language
One rule (N=197)
76 (39 %)
121 (61 %)
Rule+routine (N=204)
79 (39 %)
125 (61 %)
Two rules (N=283)
93 (33 %)
190 (67 %)
11
Question 2: explanatory stance by activity and task
Task
complexity
One rule
(N=197)
Rule+routine
(N=204)
Two rules
(N=283)
12
Activity
Anthropomorphic
language
Technological
language
Observer (N=44)
21 (48 %)
23 (52 %)
Constructor (N=153)
55 (36 %)
98 (64 %)
Observer (N=41)
19 (46 %)
22 (54 %)
Constructor (N=163)
60 (37 %)
103 (63 %)
Observer (N=22)
14 (64 %)
8 (36 %)
Constructor (N=261)
79 (30 %)
182 (70 %)
Question 2: explanatory stance by activity and task
A first glance on the data unveils several facts:
☞ The number of statements increased with the complexity of the
tasks
☞ In all tasks, about two thirds of the statements were phrased using
technological language
☞ Most of these statements were generated by the constructors,
who generated five times more statements than the observers
☞
In all tasks, the percentage of observers' statements using
anthropomorphic or technological language was similar (~50%),
while two thirds of the constructors' statements were phrased
using technological language
☞
With the increase in tasks' complexity, the use of anthropomorphic
language by the observers increased and by the constructors
decreased. At the same time the use of technological language by
the constructors remained at constant level - about two thirds of
the statements
13
[preliminary] Discussion
☞ Technological language is needed for addressing tasks of
increasing complexity, both for understanding and explaining
the artifacts behaviour and more evidently for programming it
☞ While approaching the "breed" of behaving and adaptive
artifacts children very rapidly adopt appropriate (even if not
accurate or correct) language and explanatory approach.
☞ In contrast with previous findings, which reported on
kindergarten-age children's tendency to adopt animistic and
psychological perspectives, we have observed that the
engagement in constructing the anthropomorphic artifacts'
behaviour promoted the use of technological language and
indicated the early development of the engineering stance
☞ Children's involvement in tasks integrating symbolic (i.e.,
14
working with the iconic interface) and physical (i.e.,
manipulating and observing a real artifact) activities supports
their thinking and acting beyond the expected at this age
level ("concrete-abstractions").
Implications and future work
Research-based:
☞ Definition of contents: foci, scope and pace
☞ Pedagogical design of developmentally appropriate learning
opportunities, learning materials and learning environments
☞ Formalization of developmentally appropriate pedagogical
solutions
☞ Design of teacher formation plans and contents
☞ Design of comprehensive implementation plans:
sustainability, transferability and scalability prospects
15
Tel-Aviv University | School of Education | satec | ktl
Thank you !!!
David Mioduser & Asi Kuperman
16