Agent-based simulation of ethnic prejudice transmission using Entorama tool

1
Agent-based simulation of ethnic prejudice
transmission using Entorama tool
Armano Srbljinovic Ministry of Defence
Institute for Research and Development of Defence
Systems, Zagreb, Croatia Vjekoslav Afric Faculty
of Humanities and Social Sciences University of
Zagreb, Croatia Branimir Cace Faculty of
Mechanical Engineering and Naval Architecture
University of Zagreb, Croatia
2
Outline

• Theoretical background
• Research hypotheses
• High-level description of the model
• Expected support from the model to the hypotheses
• Brief demonstration of the model using Entorama

Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
3
Contact theory

• Gordon W. Allport The Nature of Prejudice, 1954
• Thomas F. Pettigrew Racially Separate or
  Together, 1971

• Relates contact between ethnic groups with
  prejudice within groups against members of other
  groups

Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
4
The role of contact

• Contact amplifies or reduces effects of other
  factors present in situations in which contact
  occurs

Contact
Other factors
Ethnic prejudice
Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
5
Unresolved issues of contact theory

• Does not specify
• Factors which might be influenced by contact
• Mechanisms by which contact influences other
  factors

Contact
?
?
Other factors
Ethnic prejudice
Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
6
Theoretical hypotheses

• Economic context as an additional factor of the
  contact theory
• Economic context impacts ethnic prejudice
• Contact moderates the influence of economic
• context

Contact
Economic context
Ethnic prejudice
Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
7
Support to the hypotheses

• Empirical research (1)
• Higher levels of economic development reduce the
  likelihood of ethnic wars. (Sambanis, 2001)
• The occurrence of a recession increases the
  probability of internal conflict. (Blomberg and
  Hess, 2002)
• Contact theory itself (2)
• Prejudice is lessened when the two groups
• possess equal status,
• seek common goals,
• are cooperatively dependent upon each other,
• interact with the positive support of
  authorities, laws, or customs. (Pettigrew, 1971)

Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
8
Support to the hypotheses

• Can we get additional support using an
  agent-based simulation model ?

Agent-based model
Contact
Economic context
Ethnic prejudice
Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
9
Reasons for using AB modelling

• Allows playing with theories without the need
  to interfere with real human subjects
• Particularly adequate for investigating
  micro-foundations of macro-phenomena

Micro-sociological foundations of group
movements and conflicts have been neglected in
political science, mainly because the proper
theoretical tools to investigate the motives and
thoughts of individuals, as well as the
mechanisms and conditions of group mobilization,
are lacking. (Bollerup and Christensen, 1997)
Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
10
Basic experimental design

• Dependent variable ethnic prejudice
• Independent variables level of contact, presence
  of economic context

Contact
low
high
not present
not present
Economic context
low
high
present
present
Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
11
Operationalization the model

• 2 model settings
• Basic
• Pure prejudice transmission model
• Corresponds to the absence of additional factors
  in contact theory
• Extended
• Prejudice transmission embedded in economic
  context
• Economic context as additional factor
• 2 ethnic segregation patterns
• High segregation corresponds to low contact
• Low segregation corresponds to high contact

Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
12
Basic setting prejudice transmission

• Agents of 2 ethnic groups blue and red
• Main attribute prejudice level
• Low, medium, or high
• Updated in each step by taking into account
  prejudice levels in an agents socializing
  circle
• Socializing circle of an agent
• A (fixed) group of other agents that an agent
  meets and has an opportunity to interact with
• Interaction observing the prejudice level of
  the other agent

Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
13
Basic setting prejudice transmission

• Probability of interaction with a given agent
  from the socializing circle depends on
• Ethnic membership of the given and the other
  agent
• Strength of the norm of exclusion within the
  given agents ethnic group
• Prejudice level of the given and the other agent
• Strength of the norm of exclusion
• Group parameter indicating the inclination of
  group members to ostracize
• Members of other groups
• Members of the own group who are not willing to
  ostracize members of other groups
• Operationalized as a proportion of agents with
  high prejudice level within an ethnic group

Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
14
Extended setting economic transactions

• Basic setting extended with a simple economy of
  exchange with unlimited resources (EoE-UR)
• Additional attribute wealth
• Each interaction accompanied with a simple
  economic transaction which increases the wealth
  of the interacting agents by random amount
  between 0 and 1

Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
15
Extended setting economic transactions

• If the wealth of the given agent is approximately
  equal to the wealth of the agents from the
  socializing circle
• Prejudice level update is the same as in the
  basic setting
• If the wealth of the given agent is significantly
  less than the wealth of the agents from the
  socializing circle
• The given agent is dissatisfied and has an
  incentive to try a new form of behaviour
• The new prejudice level of the given agent is
  chosen randomly

Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
16
Ethnic segregation patterns

• High segregation/low contact
• Blue agents spatially segregated from red agents
• Low segregation/high contact
• Blue agents spatially interspersed with red agents

Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
17
Operational hypotheses

• (H0) No significant effects of contact on ethnic
  prejudice in the absence of additional factors
  (basic setting)
• (H1a) Negative effect of the EoE-UR on ethnic
  prejudice
• (H1b) Amplifying effect of contact
• Stronger negative effect of the EoE-UR on ethnic
  prejudice when the contact is high, than when the
  contact is low

Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
18
Micro-explanation (H0)

• In the segregated setting agents only meet more
  agents of the other colour, all other things
  being equal.
• Prejudice level depends mainly on the amount of
  prejudice in the surroundings
• If the proportions of prejudiced agents in the
  socializing circles are approximately equal
  (irrespective of the prevailing colour),
  prejudice levels of agents will stay
  approximately equal.
• ? Increased contact between groups will have no
  significant effect on prejudice.

Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
19
Micro-explanation (H1a)

• Wealth increases with interaction
• Ideal-case assumption unlimited resources, no
  competition
• Less prejudiced agents are more open - have
  more opportunities for interaction
• There is a clear incentive to become less
  prejudiced

Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
20
Micro-explanation (H1b)

• In the non-segregated setting, prejudiced agents
  interact less than in the segregated setting
  (less brethren in the surrounding)
• Prejudiced agents become less wealthy in the
  non-segregated setting than in the segregated
  setting
• There are more incentives to become less
  prejudiced in the non-segregated setting than in
  the segregated setting
• Stronger negative effect of the EoE-UR on ethnic
  prejudice in the non-segregated setting (when the
  contact is high)

Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
21
Control variables

• Initial proportion of prejudiced agents in the
  population
• Radius of the socializing circle
• Population density

Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
22
Implementation in Entorama
Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool
23
References

• Allport, Gordon W., 1954, The Nature of
  Prejudice, Addison-Wesley.
• Axelrod, Robert, 1997, The Complexity of
  Cooperation Agent-Based Models of Competition
  and Collaboration, Princeton University Press.
• Blomberg, S. Brock and Gregory D. Hess, 2002, The
  Temporal Links between Conflict and Economic
  Activity, Journal of Conflict Resolution, 46(1),
  74-90.
• Bollerup, Søren Rinder and Christian Dons
  Christensen, 1997, Nationalism in Eastern Europe
  Causes and Consequences of the National Revivals
  and Conflicts in Late-Twentieth-Century Eastern
  Europe. Macmillan.
• Cace, Branimir, 2008, Entorama,
  http//www.entorama.com
• Forbes, Hugh D., 1997, Ethnic Conflict Commerce,
  Culture, and the Contact Hypothesis, Yale
  University Press.
• Grim, Patrick, Evan Selinger, William Braynen,
  Robert Rosenberger, Randy Au, Nancy Louie and
  John Connolly, 2005, Modeling Prejudice
  Reduction Spatialized Game Theory and the
  Contact Hypothesis, Public Affairs Quarterly,
  19(2), 95-125.
• Hardin, Russell, 1997, One for All The Logic of
  Group Conflict, 2nd edition, Princeton University
  Press.
• Pettigrew, Thomas F., 1971, Racially Separate or
  Together, McGraw-Hill.
• Sambanis, Nicholas, 2001, Do Ethnic and
  Non-Ethnic Civil Wars Have the Same Causes? A
  Theoretical and Empirical Inquiry (Part I),
  Journal of Conflict Resolution, 45(3), 259-282.
• Srbljinovic, Armano, 2008, The Transition of a
  Multiethnic Community from Segregation to
  Cultural Pluralism, Doctoral thesis (in
  Croatian), University of Zagreb.

Srbljinovic, Afric, Cace Agent-based simulation
of ethnic prejudice transmission using Entorama
tool