Framing Public Policy and Prevention of Chronic Violence in Youth Kenneth A. Dodge Presentation to the Norwegian Center for Child Behavioral Development October 26, 2010 Support is appreciated from from NIMH, NIDA, and DoE.
Framing Public Policy and Prevention of Chronic Violence in Youth
Kenneth A. Dodge
Presentation to the Norwegian Center for Child Behavioral Development
October 26, 2010
Support is appreciated from from NIMH, NIDA, and DoE.
Framing the Problem of Chronic Violence in Youth
“On the horizon… are tens of thousands of severely morally impoverished juvenile super-predators …doing homicidal violence in ‘wolf packs’” (John Dilulio, 1996)
“Much of public policy toward the disadvantaged starts from the premise that interventions can make up for genetic…disadvantages, and that premise is overly optimistic.” (Herrnstein & Murray, The Bell Curve, 1994)
Framing the Solution to theProblem of Chronic Violence
Dramatic public response and change in policy away from prevention and toward restraint:
The Violent Youth Predator Act of 1996 required certain juveniles to be tried in adult court
Zero-tolerance policies in schools (Expulsion)
Metal detectors and School Resource Officers (SROs)
"most massive prison expansion in American history" (Hayden, 1995, p. B13)
Theses of this Talk
1. Metaphors and frames influence how we conceptualize a problem, its cause, and its solution.
2. The metaphors used to characterize adolescent offenders have been incite-ful but not insightful.
3. Science can inform the accuracy of a frame.
4. Prevention efforts that are based in a scientifically accurate frame can be effective.
Framing Problems in Public PolicyStudy of metaphors in human communication:
Anthropology (Bateson)Linguistics (Lakoff)Cognitive psychology (Schank; Tversky & Kahneman )Public policy (Goss)
Metaphors succeed when they lead to:
1. Assimilation recognizable to listener2. Accommodation reframe to broader schema3. Action evoke response4. Accuracy science supports
The “Super Predator” Frame
1. AssimilationAnimal-like, inevitable, vicious, no bounds
2. Accommodation Biological/genetic defect, unchangeable
3. ActionConstrain and “track” foreverMurray & Herrnstein: Advocate policy to “manipulate
fertility” of poor families
4. Accuracy?
The Genomic Revolution
Framing of Genetic Causes
Francis Collins (June, 2000): “We have caught our first glimpse of our own instruction bookpreviously known only to God.”
James Watson: “Our fate is in our genes.”
Nelkin (2001, Science): "The meaning of genes for human development and even for complex diseases requires an understanding of their social and developmental context."
Types of Gene x Parenting Interactions
I. Parenting context buffers or mutes the causal effect of genes on behavior
II. Genetic context alters the causal effect of parenting on behavior
III. Dynamic cascades
Parenting Buffers theCausal Impact of Genes
GABRA2 (Gamma-aminobutyric acid receptor α2)
-- A major inhibitory neurotransmitter
-- Single nucleotide polymorphisms (# copies risk allele) in larger haplotype block have been associated with alcohol dependence, antisocial personality, and conduct disorder (Dick et al., 2006, Behavior Genetics)
Hypothesize that high parental supervision could buffer effect of risk allele frequency
Child Development Project Research Design
(with Jack Bates, Greg Pettit, Jennifer Lansford; NICHD, NIMH, NIDA)
Community sample of 585 preschoolers-- 51% male; 19% African American; 3 sites-- diverse income; 50% live with single parents
Start in 1987 and follow through age 26, with 90% providing data in last three waves
Annual assessments by parents, teachers, peers, partners, self, tests, archival records, observations in home and school, lab tasks
GABRA2 Exerts Main Effect on Externalizing Disorder at Ages 11-16
(Dick, Latendresse, Lansford, Buddle, Goate, Dodge, Pettit, & Bates, 2009, Archives of General Psychiatry)
0
5
10
15
20
25
30
0 copies 1 copy 2 copies
Per
cent
E
xter
naliz
ing
Dis
orde
r
Parental Supervision Buffers Effect of GABRA2 on Externalizing Disorder
(Dick, Latendresse, Lansford, Buddle, Goate, Dodge, Pettit, & Bates, 2009,Archives of General Psychiatry)
Types of Gene X Parenting Interactions
I. Environmental context buffers or mutes the causal effect of genes on behavior
II. Genetic context alters the causal effect of parenting on behavior
III. Dynamic cascades
Transposing the Framing of Genes and Environment
Genes can be the context, not just the text.
Environment can be the more direct cause of externalizing disorder.
Figure and Ground in Genes and the Environment.
MotherGenes
FatherGenes
Figure and Ground in Genes and the Environment
.
Drink from the Environmental Goblet
The Effect of Child Maltreatment on Adolescent Behavior Problems
(Lansford et al., Archives of Pediatrics and Adolescent Medicine, 2002)
0
10
20
30
40
50
60
Maltreated
NotMaltreated
Per
cent
Arrest GradeRetained
SpecialEd
Expulsion GangMember
Pregnancy
Control Variables To Account for Maltreatment Effects
1. Ethnicity2. Family poverty and education3. Single-parent headed household4. Family stressors5. Maternal social support6. Family values endorsing violence7. Child’s exposure to domestic violence8. Child’s exposure to neighborhood violence9. Difficult infant temperament10.Medical problems at birth11.Gender12.Parents’ use of harsh discipline in adolescence
MAOA in Disruptive Behavior Problems
MAOA is an enzyme that selectively degrades serotonin, norepinephrine, and dopamine.
When faced with threat or provocation, humans naturally experience fear and an impulse to react aggressively with rage, but the MAOA enzyme (presumably following from the MAOA gene) enables us to titrate and regulate that response.
MAO x Maltreatment in Conduct Disorder
So we hypothesize…
1. Severe and chronic maltreatment leads to conduct disorder,
2. But only in a genetic context in which serotonin, norepinephrine, and dopamine are not degraded by high levels of MAOA activity.
3. Hypotheses: main effect of maltreatment, plus G x E interaction
Main Effect of Early Maltreatment on Adult Arrest in CDP
(Edwards, Dodge, et al., 2010, Journal of Child Psychology & Psychiatry)
01020304050607080
No Yes
Early Physical Abuse
% Ar
rested
by A
ge 22
MAOA Buffers Maltreatment Effect on Arrest in CDP
(Edwards, Dodge, et al., 2010, Journal of Child Psychology & Psychiatry)
01020304050607080
No Yes
Early Physical Abuse
% Ar
rested
by A
ge 22
Low MAOAHigh MAOA
III. Dynamic Cascade Models
Genes (and the individual) operate through symbiotic transactions with environment:
1. Infant elicits parent response.2. Parent response elicits child adaptation,3. Which elicits new parent response.4. Pattern continues across development.
A Cascade Model of the Development of Disruptive Behavior Disorder
Behavior Problems
Early Difficultness
Genetic Context
Disruptive Behavior Disorder
A Cascade Model of the Development of Disruptive Behavior Disorder
Adverse Environment
Behavior Problems
Early Difficultness
Maltreatment
Genetic Context
Disruptive Behavior Disorder
Poor Monitoring & Supervision
Harshness
A Cascade Model of the Development of Disruptive Behavior Disorder
Adverse Environment
Behavior Problems
Early Difficultness
Maltreatment
Genetic Context
Disruptive Behavior Disorder
Poor Monitoring & Supervision
Harshness
A Cascade Model of the Development of Disruptive Behavior Disorder
Adverse Environment
Behavior Problems
Early Difficultness
Maltreatment
Genetic Context
Disruptive Behavior Disorder
Poor Monitoring & Supervision
Harshness
Parenting Disruptions in the Cascade
Enriched Environment
No Behavior Problems
Early Difficultness
MoreConsistent
Reinforcement
Genetic Context
NoDisruptive Behavior Disorder
Better Monitor & Supervision
Frames in Youth ViolenceFrame Cause Policy Accuracy
Super predator genetic eugenics & nodestiny tracking
The “Moral Defect” Frame
William Bennett wrote: Body Count: Moral Poverty… (1996)
John Dilulio (Weekly Standard, 11-27-95):“moral poverty…creates super-predators”
Solution?“My one big idea is….called religion.”
Scientific Status of Moral Defects as Cause for Youth Violent Behavior
Situational Stimulus
AggressiveBehavior
Processing Step DysregulatoryProcess
1. Encoding Selective attention2. Interpretation Attributional bias3. Affect and goal setting Anger & self defense4. Response generation Access to aggression5. Response evaluation Endorse aggression6. Enactment
Procedure for Assessing Social Information Processing
Social Information-Processing Patterns inAggressive and Non-Aggressive Children
(Dodge et al., 1986, SRCD Monographs; Dodge et al., 1996, JAP)
-0.2-0.1
00.10.20.30.40.50.60.70.8
Encoding Interpretation ResponseGeneration
ResponseEvaluation
Non-Aggressive
Aggressive
Social-Information-Processing Patterns inMaltreated and Non-Maltreated Children
(Dodge et al., SRCD, 2001
-0.10
0.10.20.30.40.50.60.70.8
Encoding Interpretation ResponseGeneration
ResponseEvaluation
Not MaltreatedMaltreated
Frames in Youth ViolenceFrame Cause Policy Accuracy
Super Predator biology incarceration no
Moral Defect sin religion no
Frames in Youth ViolenceFrame Cause Policy Accuracy
Super predator biology incarceration no
Moral defect sin religion no
Contagious disease illness quarantine partial
Permanently broken defect of death in prison nocharacter
Frames in Prevention: Vaccine
September 2005: Robert L. Johnson, M.D., a Rutgers University pediatrician and chair of the Panel for the NIH State-of-the-Science Conference on Preventing Violence in Adolescence, interrupted the proceedings to blurt out his exasperation with the prevention models currently in practice by asking:
“Why can’t there be a ‘vaccine’ for violence prevention?”
Frames in Prevention of ViolenceFrame Policy Action Accuracy
Vaccine Early inoculation noCorrective Surgery Home removal inadequateChronic disease Harm reduction maybeCardiovascular disease Multi prevention perhapsPreventive dentistry Ongoing perhaps
Illiteracy Education promising
Framing Prevention of Violence as a Literacy Issue
Universal schooling for at least 13 years (K-12)
Growing appreciation for importance of early development
Extra support for children with difficulty (special education)
Family shares responsibility; Community support necessary
Role of national government and media
Consequences of failure are not incarceration
Framing an InterventionModel for Conduct Disorder
Early Starter
Conduct Disorder
ParentingPeer relationsSIPAcademics
Intervention
Birth to Age 5 Age 5-14 Adolescence
Empirical Questions in Prevention of Violence
1. Can a comprehensive program alter proximal mechanisms (i.e., parenting, peer relations, social-cognitive skills, academics)?
2. Can a comprehensive program alter important life outcomes?
A. Lifetime prevalence?B. Equal efficacy across severity of risk levels?C. Do returns to investment cumulate across development?
The Fast Track Study1. 1991-3: screen 9,341 kindergarteners at 4 sites, 3 cohorts
-- Identify early starters (n=891)-- 69% male; 45% African American; low SES
2. Randomly assign, by school cluster, to intervention
3. Implement ten-year program to address:* Parenting skills training* Peer relations skills training* Social Information Processing (skills training)* Academic skills training
4. Follow up through annual data collection (grade 12, 85%)
School-Entry TransitionContinuing SupportMiddle School TransitionContinuing Support
FAST TRACK TIMELINEYEAR (1991-
1993)
ScreeningImplementationOutcome/Mediators
INTERVENTION
ASSESSMENTGrade K 1-3 4-5 6-7 8-10 11 12 19 20
Fast Track Intervention ComponentsParenting (weekly groups, biweekly home visits)
* Behavior management, warmth, monitoring
Peer Relations (universal, friendship groups, coaching)* Behavioral and social skills, prosocial groups
Social Information Processing (skills training)* Problem solving and attributions
Academics (tutoring, after-school groups)* Reading and organization skills
Continuously over 10 Years
Question 1: Intervention Efficacy in Parenting
-0.2
-0.1
0
0.1
0.2
Observer-RatedAppropriateDiscipline
Use of PhysicalPunishment
Intervention
Control
Intervention Effects in Peer Relations
0
0.2
0.4
0.6
Peer-Nominated Rejection Deviant Peer Association
Intervention
Control
Intervention Effects in Social Cognition
-0.3
-0.2
-0.1
0
0.1
0.2
0.3
EmotionRecognition
HostileAttibutional Bias
AggressiveResponseGeneration
AggressiveRetaliation
Intervention
Control
Intervention Effects in Academic Skills
-0.1
0
0.1
0.2
0.3
0.4
0.5
Gr 1 Standard Spache Reading Gr 3 Special Education Diagnosis
Intervention
Control
Lifetime Prevalence of Conduct Disorder: Significant Interaction of Intervention x Initial Risk Level
(CPPRG, Child Development, in press)
0
0.1
0.2
0.3
0.4
0.5
Life
time
Prev
alen
ce
Risk Group
NMod
CMod
IHiC I
HiI
Intervention Effects on Violent Arrests (Murder, Rape, Kidnapping, Arson, Assault w/ Weapon):
Sig Interaction of Intervention x Initial Risk Level, p<.04 (CPPRG, Journal of Experimental Criminology, 2010)
0
0.1
0.2
Prop
ortio
n A
rres
ted
Risk Group
HiC
HiI
Intervention Effect on Cumulative Juvenile Arrest Index (weighted Severity), Age 14-18
Significant Main Effect of Intervention (p < . 05)(CPPRG, Journal of Experimental Criminology, 2010)
1
2
3
4
Wei
ghte
d In
dex
Risk Group
Con IntNorm
Cumulative Prevalence of Conduct Disorder
(Highest-Risk Group)
0
0.1
0.2
0.3
0.4
0.5
Grade 3 Grade 6 Grade 9 Grade 12
ControlIntervention
Economic Analysis of Fast Track
Each chronic criminal costs society 1.25 million dollars (Cohen, 1998)
Fast Track costs about $58,000/child.
If Fast Track reduces prevalence of chronic criminals by > 5% points, benefit-cost ratio is +.
Assignment to intervention reduces lifetime prevalence of conduct disorder by 21% points.
Cost Savings Due to Intervention, Per Child
-100,000
-50,000
0
50,000
100,000
150,000C
ost S
avin
gs P
er C
hild
, in
Dol
lars
Low 90% Top 5-10% Top 5%
Public Policy Conclusions
1. Educational framing of development and prevention of chronic violence has support
2. Early screening is crucial, and assignment to intervention reaps new benefits each year.
3. Intervention may be a wise investment if administered to the highest-risk children and continued over time.
4. Early starters are not destined for lives of crime.
Components of a National Strategy to Prevent Interpersonal Violence
1. Implement top-down strategy
Framing for mediaFraming for funders/policy makers
2. Implement evidence-based programs
Build local infrastructureBuild scientific basis