What can we do?
moving debates about genetic
determinism in new directions
peter.taylor@umb.edu
Science, Technology & Values
UMass Boston
Ask: What can we do?
(in light of what scientists are
claiming about genetics and
environment in the development of
individual lives)
Caspi, A., et al. (2002)
"Role of Genotype in the Cycle of
Violence in Maltreated Children."
Science 297: 851-854.
1.75
1.5
Composite
ComIndex
posite Index
behavior
of antisocial
of antisocial
behavior
1.25
Lo w M AOA
1
0.75
0.5
High M AOA
0.25
0
Childho od M altre atm e nt
(No -Pr obable-Seve re )
1.25
1
Se ver e CM
ComIndex
posite Index
Composite
behavior
of antisocial
of antisocial
behavior
0.75
0.5
Pr ob able CM
0.25
0
No CM
-0.25
-0.5
M AOA
(Lo w-High)
Childhood
Maltreatment
No or Probable
Severe
MAOA
low
low antisocial
high
low antisocial
HIGH ANTI- low antiSOCIAL
social
1.75
1.5
Com posite
Index Ind ex
Composite
of antisocial
behavio r
behavior
of antisocial
1.25
Lo w M AOA
1
0.75
0.5
High M AOA
0.25
0
Childho od M altre atm e nt
(No -Pr obable-Seve re )
ComIndex
posite Index
Composite
behavior
of antisocial
of antisocial
behavior
1.75
Se ver e CM
1.5
1.25
1
0.75
0.25
Pr ob able CM
0.5
No CM
0
M AOA
(Lo w-High)
Childhood
Maltreatment
No or Probable
Severe
MAOA
low
low antisocial
high
low antisocial
HIGH ANTI- low antiSOCIAL
social
misclassification
stereotyping
individuals treated according to
group membership
(remember: information to
individualize has a cost)
Ask: What can we do?
(in light of what scientists are claiming about genes and
environment in the development of individual lives)
Challenge actions based on
differences between group
means
(variation vs. cost of individualizing)
Ask: What can we do?
(in light of what scientists are claiming about genes and
environment in the development of individual lives)
Challenge actions based on differences
between group means
(variation vs. cost of individualizing)
Investigate development
(expose multiple points of engagement)
Brown, G. W. and T. Harris (1978)
Social Origins of Depression
New York, The Free Press.
continuing working class
working class
*
class
*
institution
family
loss of
mother
*
*
*
chronically
difficult
living
conditions
severe
event
*
[premarital
pregnancy]
unsupportive
partner
lack of
care
***
anxious
attachment
psychology
genetic
predisposition
hypothetical
genetics/
biochemistry
*
childhood helplessness & low self-esteem
biochemical
predisposition
*
*
hopeless
depression
current
helplessness
*
*
* = point of
possible
intervention
time
continuing working class
working class
*
class
*
institution
family
loss of
mother
*
*
*
chronically
difficult
living
conditions
severe
event
*
[premarital
pregnancy]
unsupportive
partner
lack of
care
***
anxious
attachment
psychology
genetic
predisposition
hypothetical
genetics/
biochemistry
*
childhood helplessness & low self-esteem
biochemical
predisposition
*
*
hopeless
depression
current
helplessness
*
*
* = point of
possible
intervention
time
continuing working class
working class
*
class
*
institution
family
loss of
mother
*
*
*
chronically
difficult
living
conditions
severe
event
*
[premarital
pregnancy]
unsupportive
partner
lack of
care
***
anxious
attachment
psychology
genetic
predisposition
hypothetical
genetics/
biochemistry
*
childhood helplessness & low self-esteem
biochemical
predisposition
*
*
hopeless
depression
current
helplessness
*
*
* = point of
possible
intervention
time
Kendler, K. S., et al. (2002)
"Towards a comprehensive
developmental model for major
depression in women"
American Journal of Psychiatry
159: 1133-1145
[link]
Paul, D. (1997)
The history of newborn
phenylketonuria screening in the U.S.
http://biotech.law.lsu.edu/research/
fed/tfgt/appendix5.htm
newborn screening for PKU
-> special diet -> severe
retardation prevented
multiple points of engagement:
reduction in false positives; diagnosis of
variability in PKU; appropriate diet
personal motivation & understanding of
people with some mental deficits
family, peer, cultural support for diet
insurance coverage
prevention of “maternal PKU”
Ask: What can we do?
(in light of what scientists are claiming about genes and environment in the
development of individual lives)
Challenge actions based on differences
between group means
(variation vs. cost of individualizing)
Investigate development
(expose multiple points of engagement)
Avoid discussion of “genes”
“environment” “interaction”
(unhelpful or questionable)
Jensen, A. R. (1969)
"How much can we boost IQ and
scholastic achievement?"
Harvard Educational Review 39: 1-123
within group variation
high heritability
gap between group means
within group => between group
social policy -> sustained IQ increase
sociological factors can’t explain all of gap
plausible: significant genetic component (w/in & b/w)
=> do not dismiss innate differences
(e.g., abstract vs. rote) & educate accordingly
25 April 2005
Black-White-East Asian IQ differences at least
50% genetic, scientists conclude in major law
journal
A 60-page review of the scientific evidence, some based on stateof-the-art magnetic resonance imaging (MRI) of brain size, has
concluded that race differences in average IQ are largely genetic.
The lead article in the June 2005 issue of Psychology, Public Policy
and Law… examined 10 categories of research evidence from
around the world to contrast "a hereditarian model (50% genetic50% cultural) and a culture-only model (0% genetic-100%
cultural)."
The paper, "Thirty Years of Research on Race Differences in
Cognitive Ability," by J. Philippe Rushton of the University of
Western Ontario and Arthur R. Jensen of the University of
California at Berkeley…
"Neither the existence nor the size of race differences in IQ are a
matter of dispute, only their cause," write the authors…
Dickens, W. T. and J. R. Flynn
(2001)
"Heritability estimates versus large
environmental effects:
The IQ paradox resolved."
Psychological Review 108(2): 346369.
within group variation
high heritability
generational
gap between group means
^
within group => between group
social policy -> sustained IQ increase
sociological factors can’t explain all of gap
plausible: significant genetic component (w/in & b/w)
logic must be wrong for racial mean gap as well
reciprocal
causation models
adulthood
(or when trait
is measured)
birth
small
variation
Matching and
Reciprocal
causation
Y0
Y1
Y2
growing up in environments
whose differences
(Y 0 ...Y 1 ...Y 2 ...) in part match
differences in the trait (or
in underlying traits) at each
stage of life & in part are
given by transient
non-matching influences.
In addition, every
individual's environment
(Ys) follows society-wide
trends that result from
average of all individuals'
changes.
ranking among adults
correlates with ranking
at birth, yet generation
to generation trends
can occur
matching of
environments
+
social
multiplier
Ask: What can we do?
(in light of what scientists are claiming about genes and environment in the
development of individual lives)
Challenge actions based on differences
between group means
(variation vs. cost of individualizing)
Investigate development
(expose multiple points of engagement)
Headstart
Improved scholastic
ability
continued or declining?
?
Socially desirable youth
& adult outcomes
Taylor, P. J. (2005)
“The ‘IQ paradox’ reconceived:
Visualizing the limited relevance
of human heritability estimates in
explaining differences between
means across groups or across
generations”
Working paper.
Association
if you see X, you see Y
?
Intervention/ Engagement
if you do X, then you see Y
agricultural research
difference in
statistical
“effects”
heritability
for b/w group
means
“genetic”
not separable
from
“environmental”
because twin
replication is w/in
group/location
vs.
human behavioral genetic research
measurable factors
association
(regression
analysis)
envtl./social/
cultural
factors
(Caspi et al.
genetic)
experimentally varied
developmental
causes
School A
School B
Teacher:
Jones
algebra scores
of 15 students
from school A
randomly
assigned to J.
Teacher:
Mahoud
algebra scores
of 15 students
from school B
randomly
assigned to M.
Kenton
scores of a
different 15
randomly
assigned to K.
Nykov
scores of a
different 15
randomly
assigned to N.
Lee
etc.
O’Keefe
etc.
School A
average
School B
average
Teacher:
Jones
algebra scores
of 15 students
from school A
randomly
assigned to J.
86
Teacher:
Mahoud
algebra scores
of 15 students
from school B
randomly
assigned to M.
32
Kenton
scores of a
different 15
randomly
assigned to K.
42
Nykov
scores of a
different 15
randomly
assigned to N.
44
Lee
etc.
64
O’Keefe
etc.
56
School A average 64
School B average 44
Interpretations?
• school A superior to B in facilities, admin, etc.
• teachers in school A superior (on av.) to
teachers in B
• school A superior students (on av.) in algebra
School A
average
School B
average
Teacher:
Jones
algebra scores
of 15 students
from school A
randomly
assigned to J.
86
Teacher:
Mahoud
algebra scores
of 15 students
from school B
randomly
assigned to M.
32
Kenton
scores of a
different 15
randomly
assigned to K.
42
Nykov
scores of a
different 15
randomly
assigned to N.
44
Lee
etc.
64
O’Keefe
etc.
56
School A average 64
School B average 44
Justified interpretation: for nested situation
algebra scores for students in school A taught
by teachers in school A are superior (on av.)
to algebra scores for students in school B
taught by teachers in school B
School A
average
Teacher:
Jones
algebra scores
of 15 students
from school A
randomly
assigned to J.
86
Kenton
scores of a
different 15
randomly
assigned to K.
42
Lee
etc.
64
School B
average
Teacher:
Mahoud
algebra scores
of 15 students
from school B
randomly
assigned to M.
32
Nykov
scores of a
different 15
randomly
assigned to N.
44
O’Keefe
etc.
56
Location A
average
Twin:
J---
IQ scores of 2 J--twins growing up in
separate families in
location A
126
K---
IQ scores of 2 K--twins growing up in
separate families
etc.
82
L---
etc.
104
Location B
average
Twin:
M---
IQ scores of 2 M--twins growing up in
separate families in
location B
72
N---
IQ scores of 2 N--twins growing up in
separate families
etc.
84
O---
etc.
56
Location A
Euro-Am.
Twin:
J---
average
IQ scores of 2 J--twins growing up in
separate families in
location A
126
K---
IQ scores of 2 K--twins growing up in
separate families
etc.
82
L---
etc.
104
Justified interpretation:
IQ scores for Euro-Am’s
growing up with experiences
associated w/ Euro-Am. group
membership are superior (on
av.) than IQ scores for AfrAm’s growing up with
experiences associated w/ AfrAm. group membership
includes
experiences
associated w/
Euro-Am.
racial group
membership
includes
experiences
associated w/
Afr-Am. racial
group
membership
Location B
average
Afr-Am.
Twin:
M---
IQ scores of 2 M--twins growing up in
separate families in
location B
72
N---
IQ scores of 2 N--twins growing up in
separate families
etc.
84
O---
etc.
56
agricultural research
difference in
statistical
“effects”
heritability
for b/w group
means
“genetic”
not separable
from
“environmental”
because twin
replication is w/in
group/location
vs.
human behavioral genetic research
measurable factors
association
(regression
analysis)
envtl./social/
cultural
factors
(Caspi et al.
genetic)
experimentally varied
developmental
causes
25 April 2005
Black-White-East Asian IQ differences at least
50% genetic, scientists conclude in major law
journal
A 60-page review of the scientific evidence, some based on stateof-the-art magnetic resonance imaging (MRI) of brain size, has
concluded that race differences in average IQ are largely genetic.
The lead article in the June 2005 issue of Psychology, Public Policy
and Law… examined 10 categories of research evidence from
around the world to contrast "a hereditarian model (50% genetic50% cultural) and a culture-only model (0% genetic-100%
cultural)."
The paper, "Thirty Years of Research on Race Differences in
Cognitive Ability," by J. Philippe Rushton of the University of
Western Ontario and Arthur R. Jensen of the University of
California at Berkeley…
"Neither the existence nor the size of race differences in IQ are a
matter of dispute, only their cause," write the authors…
4 themes to move debates about genetic
determinism in new directions
Ask: What can we do?
(in light of what scientists are claiming about genes and environment in the
development of individual lives)
Challenge actions based on differences
between group means
(variation vs. cost of individualizing)
Investigate development
(expose multiple points of engagement)
Avoid discussion of “genes” “environment”
“interaction”
(unhelpful or questionable)