w4 Flashcards
why should we create a formal model for some theory
Formalization in the form of mathematical equations or computer code
- Forces us to be precise (assumptions, scales of variables, relations etc.)
- Demonstrates that predictions really follow from the model
- Detect by-effects (unwanted predictions)
- Prevents misunderstanding
Agent-Based Models:
These simulate interactions between individuals (or agents) to explore how simple rules governing individual behaviors can result in complex societal patterns.
Ising Model & Voter Models:
These models show how attitudes can align under social pressure and how interactions between agents can lead to societal polarization.
Hysteresis in Attitudes:
Hysteresis effect is when more energy is needed to convert to the original state fro the new stable state
. In the context of polarization, it means that once individuals adopt a strong opinion, it’s difficult for them to change back to a more moderate stance, even when presented with counter-information.
Role in Polarization:
Resistance to Change: When someone is deeply involved in a belief, they become resistant to changing their opinion, even if new evidence contradicts it. This resistance makes it hard for individuals to reconsider their views.
Polarization Increase: Because of this resistance, individuals may become more extreme in their beliefs over time, leading to a wider divide between opposing viewpoints. This can create a cycle where groups become increasingly polarized as they reinforce each other’s extreme positions.
In summary, the hysteresis effect contributes to stability in extreme views and difficulty in reaching consensus, thus enhancing polarization within society.The presentation explores how attitudes can be stable over time and resistant to new information, leading to polarization.
what is the ising model
Originally model of magnets
- Magnets consists of elements (‘needles’) with spin left or right
- They tend to align with their neighbors and with an external field (tau) if it exists
- This all depends on the temperature,
- Low temperature ‘listen to neighbors’
- High temperature ‘change randomly
in what 2 ways is the Ising model used
- As a model of an individual attitude
- As a model of polarization
explain how is the ising model used as a model of polarisation
- Spins are people (pro/contra)
- Neighborhood is social environment
- External field = external social field
- Temperature = randomness of agents
how is the Ising model used as a model of individual attitude
the more involved u get with something the more polarized u are
(attitude changes as cusp catastrophies)
(Formalization: random field Ising model
Inverse temperature = attention)
voter model
population of voters (only to positions, yes/no or 1/-1)
- u randomly meet people and you always copy peoples opinion
Hegselmann-Krause model
Opinion continuously valued
Talk to a random person
Average opinions - everyonen than agrees and end up at the average oppinion
But homophily (bounded confidence) - Reject interactions with agents that are too dissimilar, then we get
people only interact if their difference in opinions is smaller than the bound, the smaller the bound is the more subgroups in the stimulation
what are complex systems
- open systems consisting of many sub-systems that interact nonlinearly
- Various models, tools and techniques have been developed to study such systems
- Applied in physics, chemistry, biology etc.
- Laser, weather, spiral waves, chemical patterns, swarms, ant nests, brains, traffic jams, internet
A key observation about complex systems
Behavior in all kinds of complex systems alternates between stable regimes, demarcated by phase transitions
cusp
A cusp model shows how small changes can lead to sudden, drastic shifts in behavior or opinions. Attitudes may stay stable for a while but can suddenly flip when external factors like attention or pressure reach a critical point.
Interaction assumptions of Networks of networks: simplification using the cusp
- P(agent selection) depends on involvement (asynchronous update)
2a. Involvement decays
2b. Involvement increases when neighbors interact
3. Averaging information weighted by involvement
4. Opinion updated according to (stochastic) cusp equation
Persuasion paradox:
activism may lead to opposition (polarization)
Depends on speed of change in attention (involvement) versus information