Content uploaded by Serge Larivée
Author content
All content in this area was uploaded by Serge Larivée on Jun 05, 2017
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=pijp20
Download by: [Bibliothèques de l'Université de Montréal] Date: 08 June 2016, At: 09:28
International Journal of Psychology
ISSN: 0020-7594 (Print) 1464-066X (Online) Journal homepage: http://www.tandfonline.com/loi/pijp20
Six years in the cognitive and metacognitive life of
71 families
Serge Larivée , Sylvie Normandeau , Thérèse Bouffard-Bouchard , Thérèse
Peyrade , Sophie Parent & Richard-E. Tremblay
To cite this article: Serge Larivée , Sylvie Normandeau , Thérèse Bouffard-Bouchard ,
Thérèse Peyrade , Sophie Parent & Richard-E. Tremblay (1994) Six years in the cognitive and
metacognitive life of 71 families, International Journal of Psychology, 29:3, 367-391, DOI:
10.1080/00207599408246549
To link to this article: http://dx.doi.org/10.1080/00207599408246549
Published online: 24 Sep 2007.
Submit your article to this journal
Article views: 10
View related articles
Citing articles: 1 View citing articles
INTERNATIONAL JOURNAL OF PSYCHOLOGY,
1994,29
(3).
367-391
Six Years in the Cognitive
and
Metacognitive
Life
of
71
Families
Serge LarivCe, Sylvie Normandeau
School
of
Psycho- Education, University
of
Montreal, Montreal, Canada
ThCrese Bouffard-Bouchard
Psychology Department, University
of
Quebec, Montreal, Canada
ThCrese Peyrade, Sophie Parent, Richard-E. Tremblay
School
of
Psycho-Education, University
of
Montreal, Montreal, Canada
Seventy-one families from a low socio-economic background, each with a
son
at risk
of
developing psychosocial problems at the onset
of
the
study, were observed three times over a six-year period while solving a
combinatorial task presented
on
a microcomputer. The permutations task
was
used
to
gather information with regard to cognitive functioning and
cognitive self-regulation
of
the families, which were observed
as
a develop
mental unit. Cognitive functioning was assessed according to three variables:
cognitive strategies, performance, and completion
of
the task. Six
self-
regulatory activities were assessed: task definition, planning, supervision,
evaluation, parental support, and sharing
of
responsibilities. Although
inefficient strategies were mainly used across the six years, the cognitive
level
of functioning improved over the years. Cognitive self-regulatory activities
most often used across time were supervision and parental support.
Differences were observed in the three time periods for supervision, task
definition, parental support, and individual involvement. Results indicated a
Requests for reprints should be addressed to Serge LarivCe, Research Unit
on
Children’s
Psyco-social Maladjustment, School
of
Psycho-Education, GRIP, University of MontrCal,
750
boulevard Gouin est. MontrCal, Quebec H2C 1A6, Canada.
This research was supported by Conseil quCMcois pour la recherche sociale (CQRS), by
Fonds
pour
la Formation de Chercheurs et I’Aide
A
la Recherche (FCAR) du Quebec, and
by
Social Sciences and Humanities Research Council
of
Canada (SSHRC). The authors also
thank Mario Benoit, Frankie Berntche. Louise Bineau, Michtle Danis, Sylvie Foucault,
Estelle HCbert, Thertse Laporte, J&e Lehoux, Maryse Ravinelle for the data collection,
Christine Charron and Claire Gagnk for
scoring
the data, Lise Desmarais-Gervais and Nicole
Thkriault
for
computing the data, Htlbne Boileau
for
the statistical analyses, Minh Trinh
for
her help with the documentation, and HClbne Rossignol for her secretarial work.
Q
1994
International Union
of
Psychological Science
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
368
LARIVl!E
ET
AL.
link
between cognitive strategies used by the families to solve the task and
their self-regulatory activities. Results are interpreted
within
both the
Piagetian and Vygotskian perspectives. Families’ cognitive self-regulatory
activities observed
in
this study are an example of parental disengagement
and children’s growing involvement
in
a joint activity.
Key
words:
Cognitive development, metacognition, family, problem-solving.
Soixante et onze familles de milieu socio-tconomiquement faible et dont
un
des enfants Ctait
B
risque d’inadaptation psychosociale au debut de I’etude
ont ete observees
B
trois reprises sur une periode de six ans pendant la
resolution
d‘un
probltme de combinatoire
B
I’ordinateur. La tlche utilisee,
un
problkme de permutations
B
quatre elements, a permis de collecter des
donnkes
B
propos du fonctionnement cognitif et des activites d’autoregula-
tion des familles prises comme une unite developpementale. Trois elements
ont servi
B
evaluer le fonctionnement cognitif les strategies cognitives,
I’efficacitt de la demarche et le temps mis pour completer la tlche. Par
ailleurs, I’examen des activites d’autoregulation s’appuie sur six categories:
la definition de la tlche, la planification, la supervision, I’evaluation, le
soutien parental et le partage des responsabilites.
Au
plan du fonctionne-
ment cognitif, m&me si I’utilisation de strategies innefficaces demeure
majoritaire
aux
trois temps,
on
observe une amelioration generale du
fonctionnement cognitif des familles au
fil
des ans. Quant aux activites
d’autoregulation, les plus utilides aux trois temps sont la supervision
et le soutien parental. Par ailleurs, on observe des differences aux
trois temps quant
B
la supervision, la definition de la tiche, le soutien
parental et le partage des responsabilites. Enfin, I’analyse des relations
entre le fonctionnement cognitif et les activites d’autoregulation montre
que le type de strategie utilisee par les familles est associee
21
des
activitks d‘autoregulation differentes aux trois temps. Envisagee dans une
perspective developpementale
B
la fois piagetienne et vygotskienne, la
configuration des activites cognitives et
.
d’autoregulation fournit
un
be1
exemple de dksengagement graduel des parents devant I’implication
grandissante de leur enfant et ce, tout en lui conservant
un
support constant
et adapte.
Mots
cles:
Fonctionnement cognitif, metacognition, familles, resolution de
problhe.
Considering intelligence as an adaptative mechanism brings
us
to study
its manifestations
in
everyday
life
(Sternberg, 1984; 1985; Sternberg
&
Wagner, 1986). Until recently, researchers have preferred the social
cognitive perspective (Doise
&
Mugny, 1981; Selman, 1980) to the
study of everyday problem-solving requiring logical intelligence. Several
everyday events
can
be solved automatically whereas others require more
effort.
In
theory, taking charge
of
the problem-solving process explicitly
allows the use
of
heuristic cognitive strategies. It
also
allows self-regulation,
which is necessary
for
an effective supervision
of
the process (Lefebvre-
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
COGNITIVE AND METACOGNITIVE LIFE
369
Pinard
&
Pinard, 1985). However, individuals are not ready to make,
spontaneously, the efforts necessary to take charge of their cognitive
processes even when this is what is required
in
a particular situation
(Pinard, 1989).
During the past 20 years, theoretical (Brown, Bransford, Ferrara,
&
Campione, 1983; Lefebvre-Pinard
&
Pinard, 1985; Sternberg
&
Powell,
1983) and empirical (DeLoache, Sugarman,
&
Brown, 1985; Kluwe, 1987)
literature about self-regulation has been flourishing. Self-regulation refers
to the ability to use control processes such as planning how to solve a
problem, establishing strategies to reach certain goals, evaluating whether
these goals have been achieved,
or
having on-line supervision
to
control
the attainment of these goals. Self-regulation can be studied
in
children
solving problems by themselves, although they learn the most through
interaction with their parents and their peers (Azmitia
&
Perlmutter,
1989).
According to Piaget (1975), cognitive constructions essentially depend
on mechanisms of a psychological nature that are related to the subject.
By considering parents as social agents and mediators
in
the development
of
the child’s cognitive and metacognitive abilities, Vygotsky’s model
(1978; Wertsch, 1985a; 1985b) can be considered as complementary
to
Piaget’s model. The child’s cognitive development and cognitive func-
tioning may be understood better by the combination of these two theories
(Black, Ammon
&
Kroll, 1987; Forman
&
Cazden, 1985). However,
there has been little cognitive development research that considers
both Piagetian and metacognitive perspectives simultaneously (Byrd
&
Gholson, 1985; Demetriou
&
Efklides, 1985; Lynch, 1981; Normandeau,
1992).
Vygotsky’s central thesis is that individual psychological structures are
constructed through the child’s interactions with his social environment.
Internalisation is one
of
the central concepts of the Vygotskian model.
Through his interactions with a more competent individual, the child
gradually internalises cognitive functions that allow
him
to develop
greater intellectual autonomy. The competent partner contributes to the
development of self-regulatory abilities by guiding and structuring the
child’s activities during the problem-solving task (Gauvain
&
Rogoff, 1989;
Radzisweska
&
Rogoff, 1989; Rogoff, 1990). The child’s age and level
of competence
will
determine the type of intervention by the adult.
Throughout the developmental process, the child acquires new cognitive
tools. Parents therefore adjust themselves by gradually giving the child a
more significant part in the management
of
certain activities (Freund,
1990; Kontos, 1983; Kopp, 1982; Rogoff, Ellis,
&
Gardner, 1984; Weed,
1989). Parents also stimulate their child’s cognitive and metacognitive
functioning by inducing strategic behaviour that is more sophisticated than
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
370
LARIVEE
ET
AL.
what the child is capable
of
without help. Hence, the asymmetrical nature
of the adult-child relationship promotes intellectual development.
In everyday family life, many problems are solved by more than one
member
of
the family. Some of these problems are best solved when the
family unit adopts an active and reflexive attitude. Beyond the pass or fail
result of such a problem-solving process, problem-events are important for
children’s development. They not only allow children to learn various ways
of dealing with problems, but also progressively teach them the adaptive
value of a reflexive attitude.
These problem-events are particularly important because the boys
in
the current sample are at risk
of
psychosocial maladjustment due to
their disruptive behaviour. Patterson
(1986)
suggested that families with
disruptive children often have incoherent and coercive educational
practices, which could have an impact on behavioural development and on
the development of cognitive control abilities (Dodge,
1986).
Two reasons
explain why it seems suitable to study the way families undertake
a
problem-solving procedure. First, a reflexive approach to problem solving
has an adaptive value, which makes it a protective factor against maladjust-
ment (Kandel
et
al.,
1988).
Second, the important teaching role that
is normally attributed to families must be taken into consideration.
Moreover, parents play an important part in the problem-solving proce-
dure. They must have the cognitive ability and personal availability to
recognise their child’s cognitive limits and capacities in various demanding
situations.
The ability
to
solve social and interpersonal problems between family
members has previously been studied (Kendall
&
Fischer,
1984;
Patterson,
Dishion,
8i
Bank,
1985).
However, few studies have examined the
characteristics of the family as a developmental unit in managing and
solving logico-mathematical problems. This is especially true for families
of children with behaviour problems (Larivee et al.,
1989;
Normandeau et
al.,
1992).
Finally, a longitudinal research design should surpass the cross-
sectional designs that are usually used
in
showing the passage from an
interpsychological control to an intrapsychological control.
The object of this longitudinal study (every second year during
a six-year period) is to describe parents’ cognitive functioning and self-
regulatory activities with their sons during a Piagetian combinatorial
analysis task at the formal operational level. There are
five
main reasons
for using a combinatorial analysis task. First, the task is easy to perform:
It simply involves typing letters on a keyboard. Secondly, even though the
task may be achieved without formal strategies or specific knowledge,
optimal performance requires a systematic approach entailing certain
cognitive and metacognitive demands. Thirdly, despite laboratory observa-
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
COGNITIVE AND METACOGNITIVE LIFE
371
tions, the task remains enjoyable for children since
it
resembles a computer
game. Furthermore, among the various formal operational tasks, the
permutations task requires the same abilities as the ones required to play
other games of logic (i.e. MasterMind). Fourthly,
it
is possible to observe
the systematic problem-solving process. This problem-solving process can
also be observed
in
a number of everyday events by which individuals must
decide the most efficient strategies to solve certain problems. Finally,
instructions given to parents explicitly invite them to assist their child in
solving
the
problem, if they judge
it
is necessary. Further, this procedure
has the advantage of observing spontaneous interaction whereas this is not
possible
in
other studies.
According to the longitudinal design, there should be a progression
in
the strategies used to solve permutation problems. Consequently,
families’ cognitive performance should improve. Parallel to this cognitive
progression, there should be a modification in the way families take
charge of activities in the problem-solving process, with parents gradually
giving their child a more active role. Lastly, there should be a relation
between the families’ cognitive functioning and their self-regulatory
activities.
METHODS
Subjects
Subjects were 71 french-speaking Canadian families (42 with both parents
and 29 with one parent) living
in
the Montreal region. These families were
chosen according to three criteria: a
low
socio-economical status, a low
or
average level of education, and a disruptive son, as identified by
the “Preschool Social Behaviour Questionniare (PSBQ)” (Tremblay,
Desmarais-Gervais, Gagnon,
&
Charlebois, 1987).
Families were interviewed three times every
two
years during a six-year
period. At Time-1, the age of the boys varied between 6:9 and 7:9 years
(M
=
7:3 years; sd
=
0:4 years). The age of the fathers vaned between
24
and 47 years
(M
=
32:lO; sd
=
4:11 years) and
the
age of the mothers
at that time vaned between 24 and 43 years
(M
=
305; sd
=
3:9 years).
The fathers’ years of schooling varied between 3 and 14 years
(M
=
10:4
years; sd
=
2:2 years) and that of the mothers varied between 2 and 14
years
(M
=
1O:l years; sd
=
2:4
years). The families’ socio-economical
level was measured using the socio-economical index of Blishen, Carroll,
and Moore (1987). The index consists of a list of
480
occupations that are
classified according to 3 criteria: annual income, level of education, and
occupational prestige. The socio-economical level
of
the fathers vaned
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
372
IARIVkE
ET
AL.
between 25.0 and 57.3
(M
=
35.9; sd
=
10.2) and that
of
the mothers
vaned between 21.7 and 57.0 (M
=
35.9; sd
=
12.7); the national average
is 42.7 (sd
=
13.3).
Experimental Task
The task used to evaluate cognitive and metacognitive processes
of
the
families was transposed
into
a computer game. It consisted
of
an item
taken from the permutations task by LanvCe and Normandeau (1985),
which was inspired by the work
of
Piaget and Inhelder (1951) on the
development
of
permutations schema. The task consisted
of
discovering,
among the 24 possible codes,
4
that
will
enable the hidden object to
appear on screen. Each code must contain
four
letters
(R
J
B
V)
without
repeating any letter. From Time-1 to Time3 the task remains the
same but the situation changes every time. At Time-1 (Tl),
the
task
involves a martian, who returns to his planet bringing with him four objects
in
a chest. At Time-2 (T2) it is a clown who has lost four props and, at
Time3
(T3),
the goal is to find the four keys that
will
open the doors to
a castle.
Procedure
Parents’ socio-economical measures were taken at
T1.
During a six-year
period, families came to the laboratory three times (every two years)
to perform the permutations task. Each family was seen individually.
Families learned how
to
use the computer before this experiment as a part
of
a parent-child interaction in a learning situation (Tremblay, Charlebois,
et al., 1987). An experimenter presented the instructions and the goal
of
the task
to
all the members
of
the family. He invited parents to help their
child
if
they judged it was necessary. Afterward, he asked
the
child to type
a code that would cause the flying saucer to appear at T1; the clown to
appear at
T2;
and the castle to appear at
T3.
He then left the instructions,
a piece of paper, and a pencil. He specified that 10 minutes was allowed
to solve the problem and left the room. The entire session was video-
recorded.
Observation Coding Framework
of
Cognitive
Processes and Self-regulatory Activities
To
meet the needs of this study, a coding framework was created based
on the work of several authors (Bouffard-Bouchard
&
Pinard, 1988;
Flavell
&
Wellman, 1977; Larivte, Bouffard-Bouchard,
et
al., 1989;
Moss
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
COGNITIVE AND METACOGNITIVE LIFE
373
&
Blicharski, 1986; Normandeau et al., 1992; Sternberg, 1985). The coding
frameword consists
of
indicators of the families' cognitive functioning
and self-regulatory activities during the problem-solving task. Every
occurrence of a focal behaviour was noted
in
one of the categories
mentioned later.
All of the video recordings were scored by two trained observers. The
observers agreed on all of the cognitive functioning and the self-regulatory
activities. However, they did not agree on task definition or on parental
support categories. Agreement between the observers was calculated by
dividing the number of agreements by the sum
of
agreements and
disagreements, giving a total varying between
90%
and 100% depending
on the categories. Disagreements were settled by consensus.
Families' Cognitive Functioning
Cognitive functioning was evaluated by three items: (1) cognitive strategies
that were used;
(2)
the result of the cognitive problem-solving processes;
and
(3)
the time needed to complete the task.
Cognitive Strategies
Evaluation of cognitive strategies used
in
the permutations problem is
based on three parameters: the nature and frequency of use of the strategy,
the optimal strategy used by each family, and the within-session variability
of the strategies.
Nature
of
the Strategies.
Three types of problem-solving strategies in
the permutations task have been identified: non-systematic strategies
(NSS), partial strategies (PS), and systematic strategies
(SS).
These
strategies are associated respectively to the three developmental levels of
the combinatorial operations: pre-operational, concrete operational, and
formal operations (Henriques-Christofides
&
Moreau, 1974; LarivCe
&
Normandeau, 1985; Piaget
&
Inhelder, 1951; Vergnaud
&
Cohen, 1969).
Non-systematic strategies are characterised by their irregularity; the
subjects experiment by trial and error, by which they try to create as many
permutations as possible without paying attention to repetition. Families
using partial strategies search for regularities without complete success. A
sequence of three codes starting
with
the same letter is one criterion to
partial strategies. Among the systematic strategies, there is the constant
on the first column (Cl) strategy, which consists
in'
holding constant the
first
letter and permuting the other three.
In
this study, the criteria have
been lowered: only four (out of six) consecutive codes starting with the
same letter are required to identify Cl. The constant on two columns (C2)
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
374
LARIVtE
ET
AL.
strategy consists
in
(1) keeping the same letter
in
the first position six times
and
(2)
keeping the second letter
in
the second position twice, while
permuting the other two letters (LarivCe, Bouffard-Bouchard, et al.,
1989).
Thus,
in
a permutations sequence, several strategies can be
identified according to the regularity
in
the pattern.
Optimal Strategy.
This is determined as the most developed strategy
(in the operational development) used by families.
Within-session Variability.
During the problem-solving task, three
types of strategy variability profiles are possible: stability, progression, and
regression. Stability may be found in
two
particular circumstances: when
a
family uses only one strategy or when it uses several strategies but the
last one is the same as the first one (i.e.
NSS->SS->NSS).
There is
progression when a family uses several strategies and the last one is
qualitively superior to the first (i.e.
NSS->PS->SS).
Inversely, there
is
regression when the last strategy is qualitively inferior to the first
(SS->PS).
Cognitive Problem-solving Processes
Two criteria serve to evaluate
the
efficacy of cognitive problem-solving
processes: (1) the proportion of objects found (maximum four) about the
tested codes,
(2)
the time needed to complete the task. Although
all
of the families had an equal amount
of
time to complete the task,
some managed to complete
it
in five minutes or less. Consequently,
we have singled out the families that found objects
in
less than five
minutes.
Self-regulatory Activities
Evaluation of the families’ self-regulatory activities is based on six
categories. The first four categories (task definition, planning, supervision,
evaluation) resemble metacomponents whose function is to orchestrate
cognitive processes during problem-solving tasks (Sternberg, 1985; 1990).
The other two categories (sharing of responsibility and parental support)
are social components
in
the problem-solving process. The first one serves
to evaluate the respective role of each member of the family
in
taking
responsibility; this category is important since responsibility assumed by
parents and children are rarely measured separately (Freund,
1990).
The
second category serves to evaluate parents’ emotional involvement
with
their child.
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
COGNITIVE AND METACOGNITIVE LIFE
375
Task
Definition
This occurs when one member
of
the family reminds the other members
of the instructions, to ensure their understanding
of
the nature
of
the
problem. There may be task definition before undertaking the task or
when starting it. The scoring
of
this category is based on four types
of
behaviour:
(1)
determine the
goal
or
the general intention
of
the activity
or
ask another person to do it (“we must find four treasures”);
(2)
consult
the rules
by repeating instructions
or
part
of
them (“the codes must contain
all of the four letters”);
(3)
consult instructions;
(4)
use
analogies
by relating elements in the problem to prior knowledge (“it’s like
MasterMind”).
Planning
Occasionally, planning behaviours occur while the task is in process but
essentially, they occur during the beginning
of
the task. Planning occurs
when one
of
the family members specifies
or
clarifies the way in which they
must proceed to solve the problem. Planning may involve role distribution
or
selection of a particular problem-solving process. Evaluation
of
this
category is based on three types of behaviour:
(1)
planning the procedure
by suggesting a solving strategy (for example, “we must try all the ones
that start with the same colour”);
(2)
explicitly establishing roles
(“Mom
will write down the codes, you will try to find them”) or initiating turn
taking;
(3)
anticipating results
or the impact of an action not yet produced
(“if we always start with the same letter, we will get it”), or making a
reference to motives underlying the action (“it’s important to write it
down, this way we will not repeat the code twice”); this is what Caverni
(1988)
refers to as verbal justifications, which are related to making
decisions and formulating hypotheses.
Supervision
There is supervision (sometimes referred to as monitoring) when a
member of the family takes control
of
the cognitive problem-solving
process during the problem-solving procedure. Supervision is evaluated
according to four criteria:
(1)
taking note
of
tested codes;
(2)
controlling
repetition
by consulting the codes that have been written down or by
verbalising any doubt about the repetition of codes (“we have already tried
this one”);
(3)
repeating a code
(a high number
of
repetitions suggest a low
level of supervision);
(4)
resorting
to
tusk definition,
whereby one
of
the
members manifests one
of
the four types of behaviours in the “task
definition” category
during
the problem-solving procedure.
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
376
IARIVtE
ET
AL.
Evaluation
The evaluation of the self-regulatory activities includes two behaviours:
(1)
evaluating achieved progress
of the task, either simply (“we found two,
there are two left to be found”)
or
in more detail (“We did all of the
R’s”);
(2)
evaluating the method rhat
was
used,
that is, deciding to keep
or
to
change the procedure.
Parental Support
This category contains questions
or
statements by parents about the task
that are directed toward the child, encouraging
him
to proceed. This
category is especially scored when parents dictate the next step (“Erase
it
and find another one”)
or
when they encourage their son to persist
with
a
strategy that they think is more effective (“Continue
with
the colour
blue”).
Sharing
of
Responsibilities
In
evaluating sharing of responsibilities, three pieces of information are
noted for each participant: the number of decisions taken about the choice
of codes (DECIDE), the number of times they execute (EXECUTE), and
the number of times they write down a code (NOTE).
RESULTS
The results are divided into three parts. The first part deals with cognitive
functioning data, the second part deals with self-regulatory activities data,
and the third part consists
of
the relation between cognitive functioning
and self-regulatory activities.
There was no family structure (one parentboth parents) effect on the
ob-
served variables except with regard to sharing of responsibilities. Results
show
a
methodological artifact that has no theoretical meaning.
As
expected
when two rather than three people work together, children working alone
with
their mother will take proportionally more responsibilities than
will
children working
with
both parents. Consequently, analyses
will
be per-
formed on the entire sample without family structure distinction.
Cognitive Functioning Analysis
The evaluation of cognitive functioning was based on three elements:
cognitive strategies, cognitive problem-solving process outcome, and the
time needed to complete the task. Three parameters were taken into
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
COGNITIVE AND METACOGNlTlVE LIFE
377
TABLE 1
Mean Frequency
of
Strategy use
by
Families (n=71) at Times
1,
2,
and
3
Time
I
Time
2
Time
3
Strategy”
M
sd
M
sd
M
sd
NSS
1.23
0.59
0.97
0.48
0.92
0.63
PS
0.47 0.65 0.28 0.57 0.48 0.61
ss
0.27 0.53
0.31
0.62 0.52 0.67
a
NSS: Non-systematic strategy; PS: Partial strategy;
SS:
Systematic
strategy.
TABLE
2
Distribution
of
Families According to Their
Use
of
Optimal Strategy (N=71), at Times
1,
2,
and
3
Srrategy“ Time
I
Time
2
Time 3
NSS
PS
ss
36 41 24
19
13
17
16 17
30
a
NSS: Non-systematic strategy;
PS:
Partial strategy;
SS:
Systematic strategy.
consideration when evaluating cognitive strategies: nature and frequency
of strategies, the use of an optimal strategy by families, and within-session
variability of strategies. Table
1
summarizes the mean frequency of the
different types of strategies used by families (n
=
71) at every period.
A
multivariate analysis of variance (MANOVA) was performed, with time
as a repeated measure,
on
the frequency of use of the NSS,
PS,
and
SS
strategies. The analysis indicated a significant time effect (Pillais
=
0.253,
q6, 651
=
3.66,
P
<
0.003).
Further analyses of variance, after the Greenhouse-Geisser adjustment
showed a time effect on the frequency of use of the NSS(q2, 1051
=
5.87,
P
<
0.01) and the
SS
strategy (q2, 1051
=
4.16,
PC0.05).
Newman-Keuls
post-hoc comparisons,
P<
0.05,
indicated that families used the NSS more
frequently at T1 than at
T2
and
T3.
Inversely,
SS
was used more frequently
at
T3
than at
T1
and
T2.
Table 2 summarizes the distribution of families according to the optimal
strategy used at every period. Friedman anal sis indicated that the optimal
The
NSS
was mostly used at T1 (50.7%) and at
T2
(57.7%) whereas its
use diminished at T3 (33.8%). Otherwise, the use of the
C
strategy doubles
between T1,
T2,
and
T3.
At T3,
it
was used by 42.2%
of
the families.
Table 3 summarizes the distribution of families according to within-
session variability of the strategies used at every period. Friedman analysis
showed that within-session variability
of
the strategies used by families was
strategy used by families varied
with
time
(x
1-
[2,n
-
711
=
8.56,
P<O.Ol).
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
378
LARIVEE
ET AL.
TABLE
3
Distribution
of
Families According to Within-session
Variability
of
Strategies
Time
J
Time
2
Time
3
Variability
(n
=
71)
(n
=
70)
(n
=
70)
Stability
54
48
39
Regression
6
5
2
Progression
11
17
29
different from one session to the next (x2[2,n
=
691
=
6.18,
P
<
0.045).
At
T1
(n=54) and at
T2
(n
=
48),
the profile of the strategies used by a
majority of families was stable. However, at
T3,
39
families had stable
profiles whereas
29
families showed a progression profile.
Analysis of variance, with time as a reported measure, on the efficacy
of cognitive problem-solving process indicated
a
significant time effect
(q2,140]
=
20.01,
P
<
0.001).
Newman-Keuls post hoc comparisons,
P
<
0.05,
indicated that at
T2
(M
=
0.17)
and at
T3
(M
=
0.18)
families found
a higher proportion of objects than did families at
T1
(M
=
0.10).
There was a spectacular progression
in
the number
of
families that
completed the task
in
5
minutes. At
T1
and at
T2,
2
and
5
families
respectively completed the task
in
5
minutes, whereas at
T3, 31
families
reached the 5-minute criterion. Furthermore, at
T1
and at
T2
8
and
5
families respectively did not find any objects whereas at
T3,
all of the
families found at least
1
object.
Analysis
of
Self-regulatory Activities
Evaluation of the self-regulatory activities was based on six categories: task
definition, planning, supervision, evaluation, parental support, and sharing
of responsibilities. Table
4
summarizes the mean proportion of the first five
self-regulatory categories used at each period. Univariate analyses of
variance were performed, with time as a repeated measure, on the
mean proportion of the use of each category (except for the sharing of
responsibilities category). Considering Greenhouse-Geisser's adjustments,
analyses indicated a time effect for task definition
(q2,115]
=
7.50,
P
<
0.01),
for supervision
(q2,132]
=
8.16,
P
<
0.01),
and
for
parental support
(q2,136]
=
6.15,
P
<
0.01).
However, planning and evaluation categories
were not significantly different over time. Further mean comparisons
with
the Newman-Keuls procedure,
P
<
0.05,
showed that
in
proportion
to
all
self-regulatory behaviours families define the task more often at
T1
than
at
T2
or
at
T3.
They supervised problem-solving activities more often
at
'I2
than at
T1
or
T3.
Moreover, parents offered proportionally more
support to their child at
T1
and at
T3
than at
T2.
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
COGNITIVE AND METACOGNITIVE LIFE
379
TABLE
4
Mean Proportion
of
Self-regulatory Activities
by
Families (n=71) at
Times 1,2, and
3
Time
I
Time
2
Time
3
Self-reguhtory Activities
M
sd
M
sd
M
sd
Task definition
0.02
0.04
0.005
0.02
0.008
0.02
Planning
0.03
0.04
0.03
0.04
0.03 0.04
Supervision
0.80
0.10
0.86
0.08
0.81
0.09
Evaluation
0.02
0.03
0.02 0.03 0.03
0.04
Parental
support
0.13
0.08
0.09
0.06
0.12
0.07
TABLE
5
Mean Proportion of Task Responsibility (Decision, Execution, Note)
by
Children at Times
1,
2, and
3
Time
I
Time
2
Time
3
Responsibility
M
sd
M
sd
M
sd
Decision
0.30
0.21
0.37 0.26 0.43 0.25
Execution
0.55
0.33
0.77 0.32 0.82
0.28
Taking note
0.01
0.06
0.15
0.32 0.21
0.36
Table
5
summarizes the mean proportion of responsibilities taken
by the child at every period. Multivariate analyses
of
variance were
performed, with time as a repeated measure, on the proportion
of
the child’s responsibilities in decision making, execution, and note
taking. The analyses showed a significant time effect (Pillais
=
0.59,
46,591
=
14.23,
P
<
0.001. Univariate analyses of variance, after
the Greenhouse-Geisser correction, showed that these differences were
particular to decision-making (42, 1011
=
5.76,
P
<
0.01), to the
execution of permutation problems
(42,
1011
=
20.58,
P
<
0.01), and to
taking note of the permutations (42, 1011
=
10.08,
P
<
0.01). Newman-
Keuls post hoc comparisons,
P
<
0.05,
indicated that children made
proportionally more decisions at T3 than at T1. Also, at
T3
and at
T2
they executed and took note of a greater proportion
of
permutations than
at T1.
Analysis
of
the Relation Between Cognitive
Functioning and Self-regulatory Activities
The analysis
of
the relation between cognitive functioning and self-
regulatory activities was assessed through the relation between the optimal
strategy used at every period; the self-regulatory variables previously
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
380
IARIVtE
ET
AL.
described
in
the preceding section. Families’ cognitive level is best repre-
sented by the optimal strategy category since
it
describes their highest
level of cognitive functioning during
the
permutations task. Therefore,
according to the type of optimal strategy used by families, three indepen-
dent groups have been created
(SS,
PS,
NSS). Three multivariate analyses
of variance were computed,
with
the optimal strategy as an independent
variable, one for each period. The following self-regulatory activities were
compared: task definition, planning, supervision, evaluation, and parental
support.
At T1, the optimal strategy used by families influenced their self-
regulatory activities (Pillais
=
0.34;
t;I10,130]
=
2.63,
P
<
0.006).
Analyses
of variance indicated that planning (t;I2,68]
=
6.36,
P
<
0.003) and
supervision (1;12,68]
=
7.46,
P
<
0.001) were particularly influenced.
Newman-Keuls post hoc comparisons,
P<
0.05, indicated that families
using
SS
as an optimal strategy were also the ones planning and supervising
their problem-solving activities
the
most in the permutations task.
At
T2,
self-regulatory activities were also related
to
the optimal strategy
(Pillais
=
0.26, t;I10,130]
=
1.91,
P
<
0.049). This time, univariate
analyses of variance indicated differences in supervision, (1;12,68]
=
4.71,
P
<
0.012) and
in
parental support categories F(2,68)
=
3.56,
P
<
0.034).
Newman-Keuls post hoc comparisons,
P
<
0.05, indicated that families
who used
PS
or
SS
as optimal strategies also supervised problem-solving
activities more often. Furthermore, parents
of
families
who
used
PS
as
their optimal strategy supported their children more often than did parents
of families who used NSS as their optimal strategy.
At T3, optimal strategy and self-regulatory activities were interrelated
(Pillais
=
0.28, t;Il0,130]
=
2.09,
P
<
0.030).
Univariate analyses of
variance indicated that the type
of
strategy used by families only influenced
their supervision (1;12,68]
=
5.71,
P
<
0.005).
Newman-Keuls post hoc
comparisons,
P
<
0.05,
showed that families
who
used
PS
and SS as their
optimal strategy supervised more often than did families who continued
to use NSS.
Three other multivariate analyses
of
variance have been computed
with
optimal strategy as the independent variable. They were performed at
every period
on
the sharing of responsibility variables. The type of strategy
used at
T1
(Pillais
=
0.33, I;I6,130]
=
4.21,
P
<
0.001)
and at
T2
(Pillais
=
0.31,
q6,126]
=
3.92,
P
<
0.001) was.related to the child’s respon-
sibility
in
the problem-solving task. There were no significant differences
at T3. However, at
T1,
subsequent univariate analyses of variance showed
that the families’ optimal strategy influenced children’s participation
in
decision making (1;12,66]
=
4.14,
P
<
0.020) and
in
taking note of
permutations (q2,661
=
3.66,
P
<
0.031). Children of families who used
NSS as their optimal strategy were more involved
in
the decision-making
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
COGNITIVE AND METACOGNITIVE LIFE
381
process than were children of families whose optimal strategy was
SS.
Furthermore, children of families who used
SS
as their optimal strategy
more often took note of permutations than did children of families whose
optimal strategy was
NSS
or
SS.
At
T2,
only the decision-making process
was modified by the optimal strategy
(fl2,64]
=
8.88,
P
<
0.001).
Children
from families using
NSS
as their optimal strategy were proportionally
more involved
in
the decision-making process than were children from
families whose optimal strategy was
PS
or
SS.
DISCUSSION
The object
of
this six-year longitudinal study was to describe the develop-
ment of cognitive functioning and self-regulatory activities of families
during the problem-solving process of a logico-mathematical task
(permutations). We will discuss first the results related to cognitive
functioning and then those related to self-regulatory activities.
Cognitive Functioning
of
Families
Cognitive strategies used by families, problem-solving process efficacy, and
the time needed to complete the task were the three criteria taken into
consideration when evaluating cognitive functioning. When referring to
these criteria, there appears to be a noticeable improvement
in
families’
cognitive functioning. Even though a majority of families used ineffective
strategies
(NSS)
throughout the three periods, they nonetheless improved
their cognitive performance during this time. At
7-3,
three particular
areas showed improvement: problem-solving process efficacy, the use of
heuristic strategies
within
the same session, and the use
of
SS.
With a developmental perspective, two complementary explanations
may be offered
in
order to understand the relative improvement of
cognitive functioning and performance
of
families at
T3.
The first one is
related to the methodology and the other
to
the roles and characteristics
of the participants.
The Nature
of
the Task and its Relation to Operational
Development
This study is original
in
two ways. First, on a cognitive level, the family
was considered as a developmental unit. Second,
the
families’ problem-
solving process of a logico-mathematical task was observed over a six-year
period.
In
reference to operational development, the permutations task
requires the use of formal operational strategies (constant) to solve the
problem effectively (Inhelder
&
Piaget,
1955;
Piaget
&
Inhelder,
1951).
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
382
IARIVeE
ET
AL.
Members of a family forming a developmental
unit
differ from one
another regarding their level of cognitive development. Prior studies have
indicated that less than
50%
of adolescent and adult samples reach the
formal operational level of cognitive development (Gruber
&
Voneche,
1976; Larivee, 1986; Larivee
&
Normandeau, 1984). Further studies
suggested that their learning can be limited
in
scope (Larivee, Longeot,
&
Normandeau, 1989), especially
if
they live
in
lower socio-economic
environments (Higele, 1978). Therefore, since parents are the principal
decision makers, we could be tempted to consider the nature of their
strategies as an indicator
of
their level of cognitive development.
However, three reasons prevent us from ascribing to this point of
view. First, we do not have any independent measure of the parent’s
cognitive development. Secondly, studies concerning everyday cognition
(Guberman
&
Greenfield,
1991)
suggest that acquisition of formal
operational strategies is independent from the level of education for
individuals who must solve work-related problems that require formal
schemas (Carraher, Carraher,
&
Schliemann, 1985). Therefore,
it
is
possible that the game-like task had led families, particularly at
T1,
to
produce codes rapidly to find objects without searching for heuristic
strategies. Since the task was presented
on
computer, the objects could be
found without necessarily resorting to a formal operational strategy.
According to the law of parsimony,
why
make an effort when
it
is
only a game? Besides, certain parents reported that performance was not
important. Many considered the task as only a game (“it’s his turn to
play;” “let him do it, it’s only a game”). Therefore, the way families
represented the task may have influenced their choice of strategy,
Thirdly, instructions given to families enabled them to proceed as they
pleased. Since previously solved permutation problems would have been
available to them, the same task achieved by parents
in
a regular clinical
setting would have allowed them, for example, to receive immediate
feedback concerning their method of problem-solving process and the
heuristicity
of
their chosen strategies. However,
in
the present procedure,
only
the four last permutations were displayed
on
screen. The others were
immediately stored
in
memory for analysis. The fact that families only used
three strategies, whereas eight different strategies have already been
identified (for a description see LarivCe
&
Normandeau, 1985), may be
related to the experimental procedure. Nevertheless, the procedural
freedom given to families generated an additional self-regulatory element,
the option of taking note of the tested codes.
If
instructions specified the utility of defining a particular problem-
solving strategy, families might have modified their.objective and searched
for
or
used more effective strategies. Indeed, during the problem-solving
process, parents’ interpretation of the task influences the type
of
help
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
COGNITIVE AND METACOGNITIVE
LIFE
383
(supervision) they give their children (Renshaw
&
Gardner,
1990;
Rogoff
et al.,
1984).
However, problems that occur
in
everyday life do
not
come
with
an instruction booklet
or
an inventory of possible solutions. Intelli-
gent behaviour consists precisely of reflecting, selecting, and introducing
strategies that help reach certain objectives. This was the goal
of
this
study:
To
examine the way
in
which families spontaneously solved this
type of task three times during a six-year period.
On the basis
of
the present results, we must ascertain that parents are
far from preferring the use of formal level strategies even
if
these ensure
optimal task achievement.
In
spite of the methodological reasons pre-
viously mentioned, the non-systematic strategy did leave a lot
of
room
for improvement. For accommodation purposes, it is plausible
or
normal
to use the
NSS
as a first step. However,
it
could have been followed by
more effective strategies, especially since the operationalization criteria
of
the
SS
had been reduced. Yet, at
T1
and
T2,
only a small percentage
of
families chose and maintained, throughout the problem-solving process, a
strategy that was qualitively superior to the strategy they used to undertake
the task. Though the Piagetian paradigm predicts that parents should have
used cognitive strategies
at
a formal operational level, the fact that they
accomplish the task
in
conjunction with their child must be taken into
consideration when interpreting the results.
The
Role
of
the Participants
The role of the participants is the most plausible explanation for the
more frequent use of the
SS
by families at
T3.
Indeed, although at
T1
and
at
T2
the majority of
the
decisions were made by parents, families who
used the NSS had children who
took
the most active role
in
the decision-
making process. Comparing the task’s level of difficulty
with
children’s
cognitive abilities at
T1
and at
T2,
it
cannot be expected that these
children, given their age
(7
to
10
years), could master the combinatorial
schema. At
T3,
however, children were older
(11
to
12
years), more
educated, and therefore more familiar
with
this type of problem.
Nevertheless, at
T3
children were just as, or more, active
in
taking
responsibilities and making decisions. Videos showed that children paid
more attention to the experimenter’s explanations and they did not
hesitate to ask questions
in
order
to
understand
fully
the nature of
the
problem. A re-examination of the raw data showed that at
T1
and at
T2
there was not one child who used an
SS.
At
T3,
9
children used
it
independently and as their first strategy whereas
14
other children used
it
in
conjunction with their parents. Five out of the nine children who used
an
SS
independently were not introduced to
it
by their parents during
previous periods.
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
384
LARIVtE
ET
AL.
A
greater involvement by children remains an interesting explanation.
However, within the Piagetian paradigm,
it
does not
fully
explain
why
parents who favour pre-formal strategies at T1 and at
T2
would favour
formal strategies at T3. If the same question is reconsidered according to
Vygotsky’s developmental theory, the explanation would be that parents
simply adjust themselves to their child’s cognitive level. At first, these
parents would have adjusted their capacities to their child by adopting
problem-solving strategies that were adapted to their child’s cognitive
level of development at T1 and at
T2.
Then, at T3, solving a combina-
tonal problem was part
of
the
11-12
year-old child’s zone of proximal
development. Therefore, parents who mastered
the
combinatorial schema
would explain to their child the principles and the advantages of using an
SS.
This was the case with 13 parents at
T3,
2
at
T2,
and
1
at T1. This
process would illustrate the passage from an interpsychological control
(carried out by parents) to an intrapsychological control (internalized by
the child) (Wertsch, 1979).
Establishment of a genuine co-operation between an adult and a child
is one of the postulates of the Vygotskian model, which implies that
parents have a higher level
of
expertise than their child. They are thus able
to evaluate how their child is likely to perform,
so
allowing them to
transmit information adapted to the child’s level (Ignjatovic-Savic, Kovac-
Cerovic, Plut,
&
Pesikan, 1988). However, though
it
may
be
desirable,
parents’ expertise is not indispensable to the stimulation of development
(Tudge
&
Rogoff, 1989). In that respect, parents’ expertise
in
a
specific
domain does not
fully
explain supervision activities and conveyance of
self-regulatory abilities toward their children (Normandeau et al.,
1992). However, it does improve parents’ quality
of
supervision and
self-regulatory abilities, which,
in
turn, influence children’s individual
performance when they solve similar problems (Normandeau
&
Arsenault,
1992).
In
spite of the previous reservations concerning parents’ level of
expertise, the data show that parents adapt their self-regulatory activities
to their child’s level of development, especially
in
sharing responsibilities.
Self-regulatory
Activities
Most empirical research studying Vygotsky’s developmental theory of self-
regulation uses a cross-sectional design. This longitudinal research design
makes
it
possible to observe,
within
a
Vygotskian context, parents’ gradual
disengagement
in
a certain number of self-regulatory activities (task
definition, planning, supervision, evaluation, parental support). This could
particularly be observed
in
the modification of responsibilities.
At every period, families most often used supervision and parental
support. Task definition, planning, and evaluation represented only
a
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
COGNITIVE AND METACOGNITIVE LIFE
385
small proportion of self-regulatory activities (see Table
5).
Although the
proportion of planning and evaluation activities remained the same
over the three periods, differences were observed
in
task definition,
supervision, parental support, and sharing of responsibilities activities.
Task Definition
The fact that families produced a small proportion of task definition
behaviours suggests either that all members of the family understood the
task and the instructions
or
that only parents understood them but
they
did not consider
it
important to give further explanations to their child.
We must indicate that task definition depends on the task itself:
If
there
are many sub-levels, subjects
will
define
it
more often. Since the task
in
this study had neither sub-levels nor sub-goals,
it
is thus justified
to
have
a small proportion of task definition.
Reduction
in
the proportion of task definition behaviours between T1
and
T3
may be explained
in
two ways. First, at T1 the permutations task
may have presented an aspect of novelty for the families. Second, at
T2
and T3, neither parents nor children needed to redefine the task since the
child has a better understanding of the task and a growing part
in
taking
responsibilities.
Moreover, a re-examination of the data for the category
of
task
definition showed that, at T1, goal definition represented 67.6% of task
definition behaviours versus 41.1% at
T2
and at T3. However, 47.6% of
families used analogy at
T2
and at
T3,
compared to one family at T1.
It
could be suggested that families using an analogy, such as referring to the
MasterMind game
or
a similar game played
in
the past, define the task at
a higher level of abstraction than any other definition. Besides, Sigel(l982)
presents analogy as a distanciation strategy that incites children to have
a
general outlook on the task while keeping a solid base onto which other
self-regulatory activities may be added. Analogy, as operationalized
in
this study, is a cognitive procedure that includes several interrelated
representations of the same knowledge (Karmiloff-Smith, 1986).
Supervision and Parental Support
These two categories are discussed together since
they
are theoretically
and empirically related. Task supervision remains the most important self-
regulatory activity throughout the three periods. Despite the high pro-
portion
of
supervision behaviours at
T2
in
relation to
TI
and
T3,
families
whose optimal strategy was
NSS
supervised less than others. Inversely,
parental support was more frequent at T1 and at
T3
than at
T2.
It
seems as
if
these two activities compensate one another; the proportion
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
386
LARIVEE
ET
AL.
of family supervision is inversely related to the proportion
of
parental
support. It must be noted that the nature
of
parental support was different
at
T1
and at
T3.
More precisely, at T1 parental support consisted of
physical help in using the computer and intellectual help in choosing
strategies, whereas at
T3
it consisted mainly of encouraging the child to
pursue effective problem-solving process.
Moreover, two out of four indices
of
supervision are particularly
informative with regard
to
cognitive performance: repetition and taking
note of the tested codes. Besides preventing repetition, taking note of the
codes previously tested helps (according to Leskow
&
Smock, 1970)
detection
of
regularities and emergence of new strategies. In fact, results
show that at
T1
and at
T2,
families who used
SS
wrote down the codes
significantly more often than did other families (Tl:
t
=
-
3.87; dl
=
21;
P
=
0.001;
T2:
t
=
-
2.17; dl
=
25;
P
=
0.04).
At
T3,
more families wrote
down the codes but the difference between families using
SS
was no longer
significant. The increasing number
of
families who wrote down the codes
could be due to a learning effect by which families learn the utility
of
writing down the codes. Therefore,
it
could
be
assumed that, at T3, there
was a learning effect
or
that the task reactivated pragmatic knowledge
acquired at
T1
and
T2
about the utility of writing down the tested codes.
Sharing
of
Responsibilities
According to Rogoff and Gauvain (1989), an interaction between partici-
pants who share responsibilities is profitable to cognitive and meta-
cognitive development. Vygotsky’s hypotheses (Wertsch,l979) suggest
that progressive internalization of behaviour regulation (i.e. passing from
regulation by others to self-regulation) should be progressive (in four
periods). Adults first assign simple tasks to the child while keeping the
more complex tasks for themselves. Then, under adults’ supervision and
control, the child becomes more active and takes more initiatives. Finally,
the child should be able to solve the same task without help. Except for
the fourth step, the data of this longitudinal study demonstrated the
passage from interpsychological control to intrapsychological control.
From
T1 to
T3,
children became more active
in
executing and
in
taking
note of them. At T1, the children’s role almost exclusively involved testing
codes, whereas at
T3,
children had a greater part in the decision-making
process. They tested almost all of the codes and they more often took note
of the tested codes.
Beyond their greater involvement in taking responsibilities, children
were also more active in planning and evaluating processes. Nonetheless,
parents evidently remained
in
charge thoughout the three periods.
Although there were no differences over time
in
planning and evaluation
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
COGNITIVE AND METACOGNITIVE LIFE
387
of the families, differences were observed
in
children’s planning and
evaluation behaviours.
The
percentage of their planning behaviour went
from 12% at
T1
and
T2
to
30%
at
T3.
Also,
the percentage
of
evaluating
behaviour went from 13% at
T1
and
T2
to 32% at
T3.
In
short,
if
results
are interpreted
in
a developmental perspective, the outline
of
the self-
regulatory activities is a good example
of
parents’ gradual disengagement
from the activity. In spite
of
their child’s increasing involvement in
necessary problem-solving activities, parents continue to give their child
constant support. Thus, the importance
of
task-defining behaviours at
T1
is replaced by the importance
of
supervising behaviours at
T2
and by
parental support at
T3.
Research on mutual parent-child influence (Bell
&
Chapman, 1986) suggests that families’ cognitive improvement is
simultaneously related to children’s cognitive development, allowing them
to take more responsibility in the problem-solving process, and parents’
gradual disengagement from the problem-solving process. Both variables
mutually influence each other.
CONCLUSION
In this study, the cognitive and metacognitive development
of
71 families,
considered as developmental units, was observed during a six-year
period. The longitudinal perspective and the focus on the family as a
developmental unit are two original elements that characterize this study.
Results indicated how a certain number
of
parents gradually adjusted
themselves to their child’s cognitive capacities during a problem-solving
task on a computer. This was accomplished by a differentiated use
of
self-
regulatory activities according to the child’s age. Results pertaining to
family cognitive functioning confirm what was already known about
individuals taking charge
of
their own cognitive functioning. Taking charge
of
one’s cognitive functioning is not an automatic process: It is related
to the individual’s cognitive development. External validity is possibly
limited, since the observed families were from lower socio-economical
environments and since one
of
the family members was a boy at risk
of
developing psychosocial problems. However, we do know that families
who are not at risk have a higher level
of
cognitive functioning (Normandeau
&
Arsenault, 1992).
Intelligence may be considered as a factor which protects against
psychosocial problems. On the other hand, parents, along with the school,
are the principal mediating agents
of
their child’s cognitive development.
Our results suggest the importance of finding ways
to
promote children’s
actualization of cognitive and metacognitive potential. Therefore, inter-
vention
programs inspired by the Vygotskian model would allow parents
to improve their role as mediating agents and, thus, allow them to take
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
388
IARIVEE
ET
AL.
better charge
of
their child’s cognitive development. Parents hold a
primary role since they are the principal context multipliers and thus, the
most qualified persons
to
enable the transfer
of
knowledge from one
domain to another (Larivee,
et
al., 1989; Parent, LarivCe,
&
Bouffard-
Bouchard, 1991).
Manuscript first received November
1992
Revised manuscript accepted October
1993
REFERENCES
Azmitia,
M.,
&
Perlmutter,
M.
(1989).
Social influences on children’s cognition: State
of
the
art and future direction.
Advances in Child Development and Behaviour,
22,
89-143.
Bell, R.Q.,
&
Chapman,
M.
(1986).
Child effects in studies
of
socialization.
Psychological
Review,
75, 81-95.
Black, A., Ammon, P.,
&
Kroll, L.
(1987).
Development, literacy, and the social construction
of
knowledge.
The Genetic Epistemologist,
XV,
13-20.
Blishen, B.R., Carroll, W.K.,
&
Moore,
C.
(1987).
The
1981
socioeconomic index for
occupations in Canada.
Canadian Review
of
Sociology and Anthropology, 24,
465-488.
Bouffard-Bouchard,
T.
&
Pinard, A.
(1988).
Sentiment d’auto-efficacite et exercice des
processus d’autorbgulation chez des etudiants de niveau colltgial.
International Journal
of
Psychology, 23.
409-431.
Brown, A.L., Bransford, J.D., Ferrara, R.A.,
&
Campione, J.C.
(1983).
Learning,
remembering, and understanding. In J.H. Flavell
&
EM.
Markman (Eds.),
Handbook
of
childpsychology:
Vol.
III.
Cognitive development
(pp.
77-166).
New York: John Wiley.
Byrd,
D.M.,
&
Gholson, B.
(1985).
Reading, memory, and rnetacognition.
Journal
of
Educational Psychology,
77,
428436.
Capon,
N.,
&
Kuhn, D.
(1979).
Logical reasoning in the supermarket: Adult females’
use
of
a proportional reasoning strategy in an everyday Context.
Developmental Psychology,
15,
45M.52.
Capon, N.,
&
Kuhn, D.
(1982).
Can consumers calculate best buys?
Journal
of
Consumer
Research,
8,
449-452.
Carraher, T.N., Carraher, D.W.,
&
Schliemann, A.D.
(1985).
Mathematics in the streets
and in the schools.
British Journal
of
Development Psychology, 3,
21-30.
Cavemi, J.P.
(1988).
La verbalisation comme source d’observables
pour
I’Ctude du
fonctionnement cognitif. In J.P. Cavemi, C. Bastien, P. Mendelsohn,
&
G. Tiberghien
(Eds.),
Psychologie cognitive: Modkles er mkthodes
(pp.
253-274).
Grenoble: Presses
Universitaires de Grenoble.
DeLoache,
J.S.,
Sugarman,
S.,
&
Brown,
A.L.
(1985).
The development
of
error correction
strategies in
young
children’s manipulative play.
Child Development,
56,
928-939.
Demetriou, A.,
&
Efklides,
A.
(1985).
Structure and sequence of formal and postformal
thought: General patterns and individual differences.
Child Development, 56,
1062-1428.
Dodge, K.A.
(1986).
A social information-processing model
of
social competence in children.
In
M.
Perlmatter (Ed.),
Minnesota
Symposium
on
Child Psychology,
Vol.
18.
(pp.
77425).
Hillsdale, NJ: Lawrence Erlbaum Associates Inc.
Doise,
W.,
&
Mugny,
G.
(1981).
Le
developpemenr
social
de
l‘intelligence.
Pans: Inter Editions.
Flavell, J.H.,
&
Wellman, H.M.
(1977).
Metamemory. In R.V. Kall
&
J.W. Hagen (Eds.),
Perspectives
on
the development
of
memory and cognition
(pp.
3-33).
Hillsdale, NJ:
Lawrence Erlbaum Associates Inc.
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
COGNITIVE AND METACOGNITIVE LIFE
389
Forman, A.E.
&
Cazden, C.B. (1985). Exploring Vygotskian perspectives in education: The
cognitive value
of
peer interaction.
In
J.
Wertsch (Ed.),
Culture, communication and
cognition. Vygofskian perspectives
(pp. 323-347). Cambridge: Cambridge University
Press.
Freund, L.S.
(1990).
Maternal regulation
of
children’s problem-solving behaviour and its
impact
on
children’s performance.
Child Development, 61,
113-126.
Gauvain, M. (1989).
Influence
of
knowledge
of
a post-test and child age
on
adult-child
planning.
Paper presented at the Biennial Meeting
of
the Society for Research
in
Child
Development, Kansas City, Missouri, April.
Gauvain, M.,
&
Rogoff,
B.
(1989). Collaborative problem solving and children’s planning
skills.
Developmental Psychology,
25,
139-151.
Gruber, J.,
&
Voneche, J. (1976). Reflexion
sur
les operations formelles de la pende.
Archives de Psychologie,
XLIV,
45-55.
Guberman, S.R.,
&
Greenfield, P.M.
(1991).
Learning and transfer in everyday cognition.
Cognitive Development,
6,
233-260.
Henriques-Christofides, A.,
&
Moreau, A. (1974). Quelques donndes nouvelles
sur
les
opkrations combinatoires et la pensee formelle.
Cahiers de Psychologie, 17,
55-64,
Higele, P. (1978). Etude de
la
mesure du niveau opkratoire des adultes.
Psychologie
Franqaise,
32,
95-1
05.
Ignjatovic-Savic, N.. Kovac-Cerovic, T., Plut,
D.,
&
Pesikan, A. (1988). Social interaction
in
early childhood and
its
developmental effects.
In
J. Valsiner (Ed.),
Child development
within culturally structured environments: Vol.
I.
Parental cognition and adult-child
interaction
(pp. 89-159). New Jersey: Library
of
Congress.
Inhelder, B.,
&
Piaget,
J.
(1955).
De la logique de I’enfant
d
la logique de /‘adolescent.
Paris:
Presses Universitaires de France.
Kandel, E., Mednick, S.A., Kirke-Gaard-Sorensen,
L.,
Hutchings, B., Knop,
J.,
Roserberg,
R.,
&
Schulsinger,
F.
(1988).
IQ
as a protective factor
for
subjects at high risk for
antisocial behaviour.
Journal
of
Consulting and Clinical Psychology, 56,
224-226.
Karmiloff-Smith, A. (1986). From meta-processes
to
conscious access: Evidence from
children’s metalinguistic and repair date.
Cognition,
23,
95-147.
Kendall, P.C.,
&
Fischer, G.L. (1984). Behavioural and adjustment correlates of problem
solving: Validational analyses
of
interpersonal cognitive problem-solving measures.
Child
Development, 55,
8794392.
Kluwe, R.H. (1987). Executive decisions and regulation
of
problem solving behaviour.
In
F.E. Weinert
&
R.H. Kluwe (Eds.),
Metacopition, motivation, and understanding
(pp.
31-64). Hillsdale, NJ: Lawrence Erlbaum Associates Inc.
Kontos,
S.
(1983). Adult-child interaction and the origins
of
metacognition.
Journal
of
Educational Research,
77,
43-64.
Kopp, C.B. (1982). Antecedents
of
self-regulation: A developmental perspective.
Developmental Psychology,
18,
199-214.
Larivee,
S.
(1986).
Le
developpement du scheme des correlations chez des adolescents et les
jeunes adultes.
Revue des sciences de /’education,
XII,
233-250.
Larivke,
S.,
Bouffard-Bouchard,
T.,
Tremblay, R.E., Charlebois. P.,
&
Gagnon, C. (1989).
Strategies cognitives et processus d’autorkgulation chez des familles en situation de
resolution de problemes logiques.
International Journal
of
Psychology,
24,
465483.
Larivee,
S.,
Longeot, F.,
&
Normandeau,
S.
(1989). Apprentissage des op6rations fomelles:
Une recension des recherches.
L‘Annee psychologique,
89,
553-584.
Larivee
S.,
&
Normandeau,
S.
(1984). Performance cognitive de futurs professionnels de
I’intervention.
The Canadian Journal
of
Higher Education,
XIV,
33-48,
Larivee,
S.,
&
Normandeau,
S.
(1985). Maitrise du scheme de la combinatoire (permutations)
chez les adolescents en classes sp6ciales.
Canodian Journal
of
Education,
10,
34.5-361.
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
390
IARIVEE
ET
AL.
Lefebvre-Pinard, M.,
&
Pinard, A.
(1985).
Taking charge
of
one's cognitive activity: A
moderator
of
competence. In
E.
Neimark (Ed.),
Moderators
of
competence
(pp.
191-21
1).
Hillsdale, NJ: Lawrence Erlbaum Associates Inc.
Leskow,
S.,
&
Smock, C.D.
(1970).
Developmental changes in problem-solving strategies:
Permutation.
Developmental Psychology,
2,
412-422.
Longeot, F.
(1978).Les
stades operatoires de Piaget et les facteurs de l'intelligence.
Grenoble:
Presses Universitaires de Grenoble.
Lynch, H.K.
(1981).
Metacognition and itr relation to a Piagetian task.
Unpublished
doctoral dissertation. School
of
Education, Fordham University.
Mendelsohn, P.
(1981).
Analyse prddurale
et
analyse structurale des activitts de
permutations d'objets.
Archives de Psychologie, XLIX,
171-197.
Moss,
E.,
&
Blicharski,
T.
(1986).
The observation
of
teaching and learning strategies in
parent-child and teacherchild interaction.
Canadian Journal
of
Research in Early
Childhood Education,
I,
203-209.
Neisser, E.
(1968).
The multiplicity
of
thought. In P.C. Wason
&
P.N. Johnson-Laird (Eds.),
Thinking and reasoning
(pp.
307-323).
Baltimore: Penguin.
Normandeau,
S.
&
Arsenault, J. (Mai,
1992).
Effect
of
maternal
task's
familiarity on
scaffolding during mother-child interactions.
Conference presentee au Congrks annuel de
la Jean Piaget Society, Montrtal, Mai.
Normandeau,
S.
(1992).
Developpement des habiletts d'autocontrdle dans une tpreuve de
proportionnalitt chez des enfants de
8
h
14
ans.
Revue canadienne
de
psychologie,
46,
Normandeau,
S.,
Larivee,
S..
Bouffard-Bouchard,
T.,
Tremblay, R.E., Gagnon, C.,
&
Charlebois, P.
(1992).
Cognitive monitoring in a joint activity between children and their non-
expert parents on a formal operation
task.
Learning
and
Individunl Differences, 4,7340.
Parent,
S.,
Larivte,
S.,
&
Bouffard-Bouchard.
T.
(1991).
Comp&ence cognitive, capacites
d'apprentissage, et mttacognition.
Journal international
de
Psychologie,
26,
723-744.
Patterson, G.R.
(1986).
Performance models for aggressive
boys.
American Psychologist, 41,
432444.
Patterson. G.R., Dishion, T.J.,
&
Bank, L.
(1985).
Family interaction: A process model of
deviency training.
Aggressive Behaviour,
10,
253-267.
Piaget, J.
(1975).
L'Pquilibration des structures cognitives.
Probl2me
central du
developpement.
Etudes d'tpisttmologie gtnetique,
XXXIII.
Pans: Presses Universitaires
de France.
Piaget, J.,
&
Inhelder, B.
(1951).
La gen2se de ride
de
hasard chez l'enfant.
Pans: Presses
Universitaires de France.
Pinard, A.
(1989).
La conscience psychologique.
Monogaphies de psychologie,
9.
Quebec:
Presses de I'Universitt du Quebec.
Radziszewska, B.,
&
Rogoff, B.
(1988).
Influence
of
adult and peer collaborators on
children's planning skills.
Developmental Psychology, 24,
840-848.
Renshaw, P.D.,
&
Gardner, R.
(1990).
Process versus product task interpretation and
parental teaching practice.
International Journal
of
Behavioural Development,
13,
489-505.
Rogoff. B.
(1990).
Apprenticeship in thinking: Cognitive development in social context.
New
York: Oxford University Press.
Rogoff, B., Ellis,
S.,
&
Gardner. W.
(1984).
Adjustment
of
adult-child instruction according
to child's age and task.
Developmental Psychology, 20,
193-199.
Rogoff, B.,
&
Gauvain, M.
(1989).
Collaborative problem solving and children's planning
skills.
Developmental Psychology, 25,
139-151.
Selman, R.L.
(1980).
The growth
of
interpersonal understanding.
New York: Academic
Press.
117-137.
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
COGNITIVE AND METACOGNITIVE LIFE
391
Sigel,
J.
(1982).
The relationship between distancing strategies and child’s cognitive
behaviour. In L. Laosa
&
J. Sigel (Eds.),
Families
a~
learning environment for young
children
(pp.
47-86).
New York: Plenum.
Sternberg, R.J.
(1984).
Towards a triarchic theory
of
intelligence.
The Eehaviouraland Brain
Sciences,
7,
269-287.
Sternberg, R.J.
(1985).
Instrumental and componential approaches
to
the nature and training
of intelligence. In
S.F.
Chipman, J.W. Segal,
&
R. Glaser (Eds.),
Thinking and learning
skills:
Vol.
2.
Current
research and open questions
(pp.
215-243).
Hillsdale, NJ: Lawrence
Erlbaum Associates Inc.
Sternberg, R.J.
(1990).
Metaphors
of
mind.
Cambridge: Cambridge University Press.
Sternberg.
R.J.,
&
Powell, J.S.
(1983).
The development of intelligence. In J.H. Flavell
&
E.M. Markman (Eds.),
Handbook of child psychology:
Vol.
Ill.
Cognitive development
(pp.
341419).
New York: Wiley.
Sternberg, R.J.,
&
Wagner, R.K.
(19%).
Practical intelligence. Nature and origins
of
competence in the everyday world.
Cambridge: Cambridge University Press.
Tremblay, R.E., Charlebois, P., Gagnon, C., Larivee,
S.,
&
Laurent, D.
(1987).
Lesgarpns
aggressifs
d
I‘ecole
maternelle.
Montreal: Groupe de Recherche sur I’lnadaptation Psycho-
sociale chez I’enfant, Universite de Montreal.
Tremblay, R.E., Desmarais-Gervais, L., Gagnon,
C.,
&
Charlebois, P.
(1987).
The
preschool behaviour questionnaire: Stability of its factor structure between cultures,
sexes, ages, and socioeconomic classes.
International Journal
of
Eehavioural
Development,
10,
467-484.
Tudge,
J.,
&
Rogoff, B.
(1989).
Peer influences on cognitive development: Piagetian and
Vygotskian perspectives. In M.H. Bornstein
&
J.S.
Bruner (Eds.),
Interaction
in
human
development
(pp.
17-40).
Hillsdale, NJ: Lawrence Erlbaum Associates Inc.
Vergnaud, G.,
&
Cohen,
R.
(1969).
Sur
I’activite combinatoire des enfants de
8
ans.
Psychologie francake,
14,
321-332.
Vygotsky, L.S.
(1978).
Mind in society.
Cambridge, Mass.: Harvard University Press.
Wagner, R.K.,
&
Sternberg, R.J.
(1985).
Practical intelligence in real-world pursuits: The
role of tacit knowledge.
Journal of Personaliry and Social Psychology,
48,
436-458.
Weed,
K.
(1989).
Parental goals and child self-regulation.
Paper presented at the Biennial
Meeting of the Society for Research
in
Child Development, Kansas City, Missouri, April.
Wertsch, J.V.
(1979).
From social interaction
to
higher psychological processes. A clarifica-
tion and application of Vygotsky’s theory.
Human
Development,
22,
1-22.
Wertsch, J.V.
(1985a).
Adult-child interaction as a source
of
self-regulation
in
children. In
S.R. Yussen (Ed.),
The growth of reflection in children
(pp.
69-97).
Orlando, Florida:
Academic Press.
Wertsch, J.V.
(1985b).
Vygotsky and the social formation of mind.
Cambridge, Mass.:
Harvard University Press.
Downloaded by [Bibliothèques de l'Université de Montréal] at 09:28 08 June 2016
