Cours # 3

Cours # 3

Introduction sur lʼinné-lʼacquis

Intelligence

Génétique comportementale

Etudes corrélationnelles de familles
et de jumeaux
1
Inné/Acquis ...
Rigidité/Flexibilité ...
Déterminisme/Libre arbitre

Une question clé est de savoir si l’inné disparaît au profit de
l’acquis dans l’échelle de l’évolution.
Exemple : Le crapeau, tel un automate (l’animal machine de
Descartes), ne fait preuve d’aucune plasticité cérébrale. Au
contraire, nous ne sommes pas préprogrammés pour jouer aux
échecs et le sentiment d’avoir un libre arbitre ne nous invite pas à
nous percevoir comme étant pré-cablés.
2

Si tel est le cas (l’inné disparaît au profit de l’acquis, si
bien que le cerveau se structure essentiellement par
l’expérience) c’est l’environnement qui prime... ce qui
signifie que la famille et la société sont entièrement
responsables de l’éducation et du devenir de ses
enfants. Dans cette vision, l’environnement doit être
enrichi et l’éducation fonctionne (dans la vision
extrême empiriste) sur le simple mode de l’instruction.
Watson : “Give me a dozen healthy infants… and my own
specified world to bring them up in and I’ll guarantee to take
any one at random and train him to become any type of
specialist I might select – doctor, lawyer, (…) thief etc.”

Dans le cas contraire, le progrès de l’enfant dépend d’un
processus de maturation et c’est l’individu qui est au premier
plan de son propre développement.
3
Une icône du débat inné-acquis

L’expérience de la falaise avec de nombreuses espèces a
longtemps été évoquée pour dire qu’il n’est pas évident que plus
un animal est évolué, plus les fonctions innées disparaissent.
Gibson et Walk (1960) : Is experience the teacher ? No... La
réponse a une connotation nativiste.
5
Le chevreau âgé d’un jour ne s’aventure pas, et il
est pétrifié si on le place sur la partie vitrée.
6
Adolph et al. (2014) : Le comportement de l’enfant
dépend de son expérience.
Rappel sur les périodes critiques

Les nativistes considèrent que l’expérience est nécessaire. Ils pensent
seulement qu’elle n’est pas suffisante. En privant d’expérience un
organisme vivant, ses gènes ne sont pas activés et la fonction dépérit, ce
qui est conçu par le modèle nativiste.
Ex: vision perturbée par la cataracte congénitale (Pawan Sinha, MIT);
langage non acquis chez certains enfants sauvages (e.g., Genie, a feral
child*).
* enfant sauvage en anglais
Figure 2. Some of the tests included in the Prakash assessment battery
Sinha et Held (2012) -
Besides basic measurements of acuity, contrast sensitivity and motion coherence thresholds, the tests for examining visual
function and
its development
Project
Prakash
following late sight-onset also assess children’s performance on ‘higher-order’ visual tasks such as image segmentation, shape matching and face localization.
(Image credits: Landolt C chart from Precision Vision Inc. and the Worth Four dot test from Optimetrics Inc.)
http://www.ted.com/talks/pawan_sinha_on_how_brains_learn_to_see
L’intelligence... qq avertissements
- Gould (1997), analyse les nombreuses mal mesures
historiques de l’homme... Par exemple, Broca remplit les
crânes de sable afin d’en mesurer la capacité, mais il ne prend
pas en compte la pondération des corps correspondant aux
crânes.
- L’intelligence sert à la sélection à l’université (surtout
étrangères) et dans le monde du travail, ce qui par circularité
favorise la corrélation entre QI et réussite.
- Etre intelligent ne signifie pas qu’on ne peut pas agir de
façon stupide, par arrogance par exemple (Sternberg, 2013).
- Le QI est une notion limitée qui ne prend en compte ni la
10
créativité, ni la sagesse, etc. (Sternberg, 2013).
Effet du milieu
Premiers résultats classiques, chez le rongeur, ...
... qui dénotent aussi une interaction entre génétique et
environnement.
11
Effet de l’adoption
Répercussion
13
Risques du milieu
Risques du milieu
Facteurs avantageant le QI
How to Make a Young Child Smarter:
Evidence From the Database of
Raising Intelligence
Perspectives on Psychological Science
8(1) 25–40
© The Author(s) 2013
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1745691612462585
http://pps.sagepub.com
John Protzko, Joshua Aronson, and Clancy Blair
New York University
Abstract
Can interventions meaningfully increase intelligence? If so, how? The Database of Raising Intelligence is a continuously updated
compendium of randomized controlled trials that were designed to increase intelligence. In this article, the authors examine
nearly every available intervention involving children from birth to kindergarten, using meta-analytic procedures when more
than 3 studies tested similar methods and reviewing interventions when too few were available for meta-analysis. This yielded
4 meta-analyses on the effects of dietary supplementation to pregnant mothers and neonates, early educational interventions,
interactive reading, and sending a child to preschool. All 4 meta-analyses yielded significant results: Supplementing infants with
long-chain polyunsaturated fatty acids, enrolling children in early educational interventions, reading to children in an interactive
manner, and sending children to preschool all raise the intelligence of young children.
Keywords
intelligence, developmental psychology, early childhood, interventions, preschool
A good deal of research confirms what most people consider
self-evident: Intelligence matters for academic and life success (Herrnstein & Murray, 1994; Nisbett et al., 2012). Accordingly, many researchers and educators have attempted to
increase the intelligence of children—particularly children
from disadvantaged backgrounds. The question of the modifiability of intelligence remains a contentious one, and findings
from these famous interventions are nuanced enough to sup-
study has to include each of the following components: a
sample drawn from a general, nonclinical population; a pure
randomized controlled experimental design; a sustained intervention; and a widely accepted, standardized18
measure of intelligence as an outcome variable.
Our first criterion is that the participants must have been
drawn from the general population. Although data from clinical populations can be informative, generalizing the effects of
ssful intervention
nitoring to ensure
ncrease a young
ons and word use
ievement (Hart &
ng such intervendings reflect the
s. However, with
umptions, we can
ons are promising
of LC-PUFA, specifically docosahexaenoic acid (DHA), per child’s IQ. Th
day; infants who were supplemented received formula that plementation
ranged from 0.2% to 0.5% LC-PUFA.
acids because
We found that supplementing either a pregnant mother development
or supplementing infant formula with LC-PUFA raises a & Stephenso
young child’s IQ by more than 3.5 points2 (g = 0.236, 95% class) make u
Protzko et al.
CIWeighted Variance = .043 to .429); the benefits to LC-PUFA sup- ing periods o
plementation does not appear when the children are tested in from all of th
infancy
= −0.064). includes eight studies, providing 10 Yergey, 1986
on.
This(gmeta-analysis
Other rese
effect sizes across 499 participants (see Appendix).
of children di
LC-PUFA plus arachidonic acid supplements
order with LC
Reading
to
a
child
in
an
interactive
style
raises
his
or
her
and intelligence
problems, an
IQ by over 6 points (g = 0.404, 95% CI = .153 to .654).
Two studies from the LC-PUFA supplementation literature 2003), all of w
With one exception, interventions that begin after the child is
merit closer attention, as they included a condition in which kley, 1997). T
42
months old do not raise the IQ; however, in the random
infants’ diets were supplemented with not only DHA but also mentation ra
effects
model,
age(ARA),
is not3 aa significant
moderator.fatty
In each
arachidonic
acid
second, nonessential
acid.of synaptogenes
these
interventions,
children and their
parentssupplements
engage with to
sto- development
Although
the two interventions
introduced
rybook
reading
in an interactive
way.period
The child
is an
active this hypothes
neonates’
diets during
the same time
(within
5 days
participant
thediffered
reading,inwith
the adult
encouraging
child weeks of L
after birth),in
they
duration.
Whereas
neonatesthe
in one
19
to
be asreceived
elaborate
as possible.forWith
exception,
these
study
supplements
onlyone
3 months,
those
in interthe received thes
ventions
do not on
appear
to and
raise
the IQ if the childdiets
is more
than vation in the
other continued
DHAARA-supplemented
for 12
Effet Flynn
Risque associé à cet effet dans la pratique du psychologue :
omission de l’identification du retard mental par l’utilisation
de normes dépassées.
Test de l’hypothèse de Greenfield
Le score de référence de 1965 est 5 pour tous les sous-tests,
à l’Echelle Collective de Niveau Intellectuel
Bradmetz & Mathy (2006)
Intelligence fluide vs cristallisée
- L’intelligence fluide est censée facturer le facteur g (pour
facteur général). Elle est utilisée de façon prédominante pour
refléter le facteur génétique.
- L’étude de Kan et al. (2013) montre pourtant que plus les
sous-tests sont culturellement dépendants (faisant référence à
une intelligence cristallisée), plus l’héritabilité est importante.
La conclusion préférée des auteurs invoque l’effet de G*E.
Les individus acquièrent des connaissances sur la base de leur
capacité de traitement reflétée par le facteur g. De plus, les
individus favorisés par g ont tendance à solliciter des
environnements dans lesquels le développement des
connaissances encyclopédiques est encouragé. Au final,
l’intelligence fluide finit par corréler avec l’intelligence
22
cristallisée.
Research Article
On the Nature and Nurture of Intelligence
and Specific Cognitive Abilities: The More
Heritable, the More Culture Dependent
Psychological Science
24(12) 2420–2428
© The Author(s) 2013
Reprints and permissions:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0956797613493292
pss.sagepub.com
Kees-Jan Kan1,3, Jelte M. Wicherts2,3, Conor V. Dolan1,3, and
Han L. J. van der Maas3
1
Department of Biological Psychology, VU University; 2Department of Methodology and Statistics,
Tilburg University; and 3Department of Psychological Methods, University of Amsterdam
Nature and Nurture of Intelligence
Table 1.! Cultural Load of the Wechsler Intelligence Test
4. We computed the Pearson correlation
AbstractSubtests on the Verbal IQ (VIQ) and Performance IQ (PIQ)
these corrected proportions and the subt
To further
knowledge concerning the nature and nurture of intelligence, we scrutinized how heritability coefficients
Subscales
and obtained
vary across specific cognitive abilities both theoretically and empirically. Data from tural
23 twinloadings
studies (combined
N = the corres
(one-tailed)
p
value.
7,852) showed that (a) in adult samples, culture-loaded
Cultural load subtests tend to demonstrate greater heritability coefficients
5. a subtest’s
We divided,
withinofeach
of the studies fro
than do culture-reduced subtests; and (b) in
samples of both adults and children,
proportion
variance
(percentage
shared with
general intelligence is a function
of its cultural
explanation
because they
Subtest
of adapted
items) load. These
Scale findings require
weanobtained
heritability
coefficients, ea
do not follow from mainstream theories of intelligence. The findings are consistent withtest’s
our hypothesis
that
heritability
heritability coefficient by the corres
Vocabulary
VIQin the contribution of genotype-environment
coefficients
differ across cognitive abilities as a35result of differences
mean reliability coefficient, which resulted
Information
VIQ are the most culture-dependent abilities sheds
covariance.
The counterintuitive finding that the22
most heritable abilities
tability coefficients corrected for attenuat
Comprehension
15
VIQ
a new light on the long-standing nature-nurture debate of intelligence.
6. We computed, within each of these stu
Similarities
9
VIQ
Pearson correlation between the subte
Arithmetic
8
VIQ
Keywords
rected heritability coefficients (h2) and c
Picture
Completion
3
PIQ individual differences
intelligence,
behavior
genetics, cognitive ability, environmental
effects,
proportions of variance shared with gene
Picture Arrangement
2
PIQ
ligence (gl 2).
Received 5/7/12;
BlockRevision
Designaccepted 5/17/13
1
PIQ
7. We pooled (weighted averaged) these
Coding
0
PIQ
tions, whereby the square root of the stud
Digit Span
0
VIQ
Whether intelligence depends more on nature or on nurnot covary, whereas genotype-environment covariance is
sizes
constituted
the weights.
Assembly
0
PIQ present and ple
ture is a Object
long-standing
issue dating back to 17th-century
presumably
might
account
for as much
as
8. in We
a combined
rationalism and empiricism and with roots in the ancient
30% of the variance
adultcalculated
IQ ( Johnson,
Penke, & p value u
Note: Subtest cultural load was obtained from Georgas, van de Vijver,
Stouffer
method
(Rosenthal,
1991; i.e., w
Greek philosophies
of Plato and
Aristotle
(Fancher,
1996).
Spinath,
2011).
Weiss, and Saklofske
(2003,
Table 18.1),
except
for the Coding
subtest
With thecultural
emergence
of psychometrics
and F.behavioral
aim in the presentformed
researcheach
was toone-tailed
further knowlp value into a
load, which
was obtained from
J. R. van de VijverOur
(personal
genetics in
the
20th
century,
it
became
possible
to
address
edge
concerning
the
nature
and
nurture
of
intelligence
communication, November 30, 2011).
multiplied each z value by the square ro
this issue empirically—in terms of individual differby scrutinizing how heritability
coefficients
vary sample
across size, sum
corresponding
study
ences—through the decomposition of variance in psyspecific cognitive abilities, both theoretically and empirioutcomes, divided this sum by the nu
chometric& intelligence
genetic
and &environmental
cally. We
evaluated the implications of the empirical findThompson,into
1994;
Owen
Sines, 1970; Segal,
1985;
studies, back-transformed the outcome in
variance components (Plomin, DeFries, McClearn, &
ings for theories of intelligence, notably, the mainstream
Nature and Nurture of Intelligence
Table 1.! Cultural Load of the Wechsler Intelligence Test
Subtests on the Verbal IQ (VIQ) and Performance IQ (PIQ)
Subscales
Subtest
Vocabulary
Information
Comprehension
Similarities
Arithmetic
Picture Completion
Picture Arrangement
Block Design
Coding
Digit Span
Object Assembly
Cultural load
(percentage
of adapted items)
Scale
35
22
15
9
8
3
2
1
0
0
0
VIQ
VIQ
VIQ
VIQ
VIQ
PIQ
PIQ
PIQ
PIQ
VIQ
PIQ
4.
5.
Note: Subtest cultural load was obtained from Georgas, van de Vijver,
Weiss, and Saklofske (2003, Table 18.1), except for the Coding subtest
cultural load, which was obtained from F. J. R. van de Vijver (personal
communication, November 30, 2011).
6.
7.
8.
We co
these c
tural l
(one-ta
We div
we ob
test’s h
mean r
tability
We co
Pearso
rected
propor
ligence
We po
tions, w
ple siz
We ca
Stouffe
formed
multip
corresp
outcom
zoom sur la figure précédente
Stabilité du QI :
Corrélations avec soi-même

r QI_5ans-QI_15ans = .67 (Humphreys, 89); environ
une dizaine de points de QI.

r 4ans-5ans = .80; r 6ans-7ans = .87;
r 8ans-9ans = .90 (Brody, 92)

Matrice fictive ...
4;6
5;6
6;6
7;6
5;6
0.9
6;6
0.89
0.91
7;6
0.88
0.9
0.92
C’est le trait psychologique le plus stable (Brody 92).
Difficile donc d’attendre mieux des jumeaux MZ !
8;6
0.87
0.89
0.91
0.93
28
Héritabilité (≠ hérédité)
var(génotype)
H =
var( phénotype)
2
29
Attention, l’héritabilité concerne une population, pas un individu !
Wiki ...
Wiki : In some combinations of environments and genotypic ranges, heritability can be 100% even while group
differences are completely environmental. For heritability to be 100%, random variation in expression must not occur.
30
Héritabilité dans wiki ...
31
Génétique comportementale

La génétique comportementale est l’étude des variations
des traits entre individus. Ces variations résultent de la
combinaison des facteurs génétiques et environnementaux.

Dans le domaine de l’intelligence, elle se fonde sur les études
familiales.

Elle utilise les études corrélationnelles plutôt que les plans
expérimentaux, malheureusement ...

Elle met en oeuvre de nouvelles techniques utilisant une
comparaison directe de l’ADN entre individus de familles
différentes.
32
Degré de parenté en ratio
% de gènes communs:
- jumeaux MZ : 100%
- jumeaux DZ : 50% (en moyenne)
- frères et soeurs : 50% (en moyenne)
- parents/enfants : 50%
- demi -frères ou -soeurs 25% (en moyenne)
- cousins :12.5% (en moyenne)
- non apparentés biologiquement ~ 0%
33
Rappel
Transmission d’une paire de
chromosomes:
Les allèles paternels sont a et b
Les allèles maternels sont c et d
ac
Enf. 2 ad
bc
bd
ac
2
1
1
0
ad
1
2
0
1
Enfant 1
bc bd moy
1
0
1
0
1
1
2
1
1
1
2
134
Rappel sur les corrélations
Negative correlation
No correlation
B
B
A
A
Corrélation typique entre QI de
jumeaux MZ
Positive correlation
B
QI J2
100
A
100
35
QI J1
Echantillons hétérogènes
Corrélation positive globale ...
b
... qui masque un effet de cohorte
b
a
a
36
Hypthèses
 Environment partagé : génération,
famille, grandir ensemble, statut socioéconomique, etc.

Environment non partagé: expériences
hors de la famille, loisirs, rang de
naissance, experiences uniques

Problème : les gènes déterminent autant le
QI qu’ils sollicitent l’environnement. Les
stimulations des enfants ayant un QI élevé
sont élicitées par leur QI.
37
Etudes familiales
(fondées sur les mesures de degré de parenté)
Etudes de familles
 Etudes de jumeaux
 Etudes des adoptions

38
Hypothèse et résultats
Les corrélations entre QI de membres de la même famille dépend
directement du % de gènes communs, selon l’hypothèse nativiste.
39
Interprétation …
Les corrélations augmentent en fonction
du % de gènes en commun (r = .96).
 L’environnement étant modifié pour des
gènes différents, et l’environnement
étant uniformisé pour des gènes
identiques, la portion de variance
attribuable au facteur environnement
est difficilement estimable.

40
Adoptions
41
Interprétation

Le facteur génétique l’emporte

La part du facteur environnemental est sousestimée, compte tenu des procédures de
placement familial en cas d’adoption : des
familles similaires étaient choisies pour le
placement des enfants (d’autant plus qu’ils
étaient jumeaux).
42
Adoption studies
43
Interprétation

La dernière catégorie d’études fait penser que
l’environnement influence le développement de
l’intelligence...

... mis à part que .... lorsque ces mêmes études sont
conduites chez l’adulte, la corrélation pour les non-biological
relatives parent-offspring tombe à 0 !
Raison: Les influences respectives des facteurs
environnementaux et génétiques ne sont pas constantes au
long de la vie. Cette dernière donnée montre plutôt que la
génétique opère un programme à long terme (e.g., le
développement du cerveau à l’adolescence est crucial), et
que le facteur environnemental ne peut avoir qu’une
influence temporaire.
44
485087
research-article2013
Variation de l’héritabilité avec
l’âge et l’environnement
CDPXXX10.1177/0963721413485087Tucker-Drob et al.Genetics of Cognition
Genetic and Environmental Influences on
Cognition Across Development and Context
Current Directions in Psychological
Science
22(5) 349–355
© The Author(s) 2013
Reprints and permissions:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0963721413485087
cdps.sagepub.com
Elliot M. Tucker-Drob, Daniel A. Briley, and K. Paige Harden
Department of Psychology and Population Research Center, University of Texas at Austin
Abstract
Genes account for between approximately 50% and 70% of the variation in cognition at the population level.
However, population-level estimates of heritability potentially mask marked subgroup differences. We review the
body of empirical evidence indicating that (a) genetic influences on cognition increase from infancy to adulthood,
and (b) genetic influences on cognition are maximized in more advantaged socioeconomic contexts (i.e., a Gene ×
Socioeconomic Status interaction). We discuss potential mechanisms underlying these effects, particularly transactional
models of cognitive development. Transactional models predict that people in high-opportunity contexts actively
evoke and select positive learning experiences on the basis of their genetic predispositions; these learning experiences,
in turn, reciprocally influence cognition. The net result of this transactional process is increasing genetic influence with
increasing age and increasing environmental opportunity.
Keywords
cognitive ability, intelligence, Gene × Environment interaction, behavior genetics, cognitive development
Intelligence is mostly a matter of heredity, as we
know from studies of identical twins reared apart.
. . . Social programs that seek to raise I.Q. are bound
to be futile. Cognitive inequalities, being written in
genetic studies of unrelated persons have converged on
similar heritability estimates (Chabris et al., 2012; Davies
et al., 2011). Despite the vociferous objections of critics
of behavioral genetic research (e.g. Charney, 2012),
351
ers: age/
Below, we
f research
ood within
ility
hey experiuts as they
ic variation
ychological
his intuitive
n actually
ults from 11
s of cogni-
Fig. 1. Proportion of variance in cognition as a function of age. Shad-
ate infancy (age 2 years) (A). Data come from a nationally representative sample of Ameri-
Pour aller plus loin...
Intelligence 42 (2014) 83–88
Contents lists available at ScienceDirect
Intelligence
Genetic influence on family socioeconomic status and
children's intelligence☆
Maciej Trzaskowski a, Nicole Harlaar b, Rosalind Arden a, Eva Krapohl a, Kaili Rimfeld a,
Andrew McMillan a, Philip S. Dale c, Robert Plomin a,⁎
a
b
c
King's College London, MRC Social, Genetic and Developmental Psychiatry Centre, Institute of Psychiatry, De Crespigny Park, London, SE5 8AF, United Kingdom
Department of Psychology and Neuroscience, University of Colorado Boulder, 345 UCB, Boulder, CO 80309, United States
Department of Speech and Hearing Sciences, University of New Mexico, 1700 Lomas Blvd, NE Suite 1300, Albuquerque, NM 87131, United States
a r t i c l e
i n f o
Article history:
Received 19 September 2013
Received in revised form 22 October 2013
Accepted 1 November 2013
Available online 6 December 2013
Keywords:
SES
Socioeconomic status
Intelligence
Cognitive abilities
GCTA
1. Introduction
a b s t r a c t
Environmental measures used widely in the behavioral sciences show nearly as much genetic
influence as behavioral measures, a critical finding for interpreting associations between
environmental factors and children's development. This research depends on the twin method
that compares monozygotic and dizygotic twins, but key aspects of children's environment such as
socioeconomic status (SES) cannot be investigated in twin studies because they are the same for
children growing up together in a family. Here, using a new technique applied to DNA from 3000
unrelated children, we show significant genetic influence on family SES, and on its association
with children's IQ at ages 7 and 12. In addition to demonstrating the ability to investigate genetic
influence on between-family environmental measures, our results emphasize the need to consider
genetics in research and policy on family SES and its association with children's IQ.
© 2013 The Authors. Published by Elsevier Inc. All rights reserved.