How do bots influence beliefs on social media? Why do beliefs propagated by social bots spread far and wide, yet does their direct influence appear to be limited?

This model extends Axelrod’s model for the dissemination of culture (1997), with a social bot agent–an agent who only sends information and cannot be influenced themselves. The basic network is a ring network with N agents connected to k nearest neighbors. The agents have a cultural profile with F features and Q traits per feature. When two agents interact, the sending agent sends the trait of a randomly chosen feature to the receiving agent, who adopts this trait with a probability equal to their similarity. To this network, we add a bot agents who is given a unique trait on the first feature and is connected to a proportion of the agents in the model equal to ‘bot-connectedness’. At each timestep, the bot is chosen to spread one of its traits to its neighbors with a probility equal to ‘bot-activity’.

The main finding in this model is that, generally, bot activity and bot connectedness are both negatively related to the success of the bot in spreading its unique message, in equilibrium. The mechanism is that very active and well connected bots quickly influence their direct contacts, who then grow too dissimilar from the bot’s indirect contacts to quickly, preventing indirect influence. A less active and less connected bot leaves more space for indirect influence to occur, and is therefore more successful in the long run.

The model is based on Swann and Buhrmester’s Identity Fusion behavioural theory, which seeks to explain why an individual puts the group’s priorities above their personal expectations. In order to observe the theory and validate group behaviour, a case study was carried out focusing on scenarios of group violence in football stadiums in Brazil. For the modelling, each agent has a distribution of levels of identification with the group to which they belong, with their level of fusion varying between 1 and 5. According to behavioural theory, an individual’s degree of fusion with the group directly interferes with their behaviour of replicating actions and absorbing group beliefs.

The model presented here is extensively described in the paper ‘Talk less to strangers: How homophily can improve collective decision-making in diverse teams’ (forthcoming at JASSS). A full replication package reproducing all results presented in the paper is accessible at https://osf.io/76hfm/.

Narrative documentation includes a detailed description of the model, including a schematic figure and an extensive representation of the model in pseudocode.

The model develops a formal representation of a diverse work team facing a decision problem as implemented in the experimental setup of the hidden-profile paradigm. We implement a setup where a group seeks to identify the best out of a set of possible decision options. Individuals are equipped with different pieces of information that need to be combined to identify the best option. To this end, we assume a team of N agents. Each agent belongs to one of M groups where each group consists of agents who share a common identity.
The virtual teams in our model face a decision problem, in that the best option out of a set of J discrete options needs to be identified. Every team member forms her own belief about which decision option is best but is open to influence by other team members. Influence is implemented as a sequence of communication events. Agents choose an interaction partner according to homophily h and take turns in sharing an argument with an interaction partner. Every time an argument is emitted, the recipient updates her beliefs and tells her team what option she currently believes to be best. This influence process continues until all agents prefer the same option. This option is the team’s decision.

The my-side bias is a well-documented cognitive bias in the evaluation of arguments, in which reasoners in a discussion tend to overvalue arguments that confirm their prior beliefs, while undervaluing arguments that attack their prior beliefs. This agent-based model in Netlogo simulates a group discussion among myside-biased agents, within a Bayesian setting. This model is designed to investigate the effects of the myside bias on the ability of groups to reach a consensus or collectively track the correct answer to a given binary issue.

NeoCOOP is an iteration-based ABM that uses Reinforcement Learning and Artificial Evolution as adaptive-mechanisms to simulate the emergence of resource trading beliefs among Neolithic-inspired households.

This model simulates economic and epidemiological interaction between citrus production and the disease Huanglongbing (HLB), which is vectored by the Asian citrus psyllid. The model is used to evaluate area-wide coordinated spraying when free-riding is possible given individuals’ beliefs in other grower participation in area-wide spraying and in the information provided by extension on the threat as HLB spread.

The present model demonstrates that how two basic human features: (1) that in the human brain beliefs are interconnected, and (2) people strive to maintain a coherent belief system, gives rise to opinion polarization and the appearance of fake news.

Communication processes occur in complex dynamic systems impacted by person attitudes and beliefs, environmental affordances, interpersonal interactions and other variables that all change over time. Many of the current approaches utilized by Communication researchers are unable to consider the full complexity of communication systems or the over time nature of our data. We apply agent-based modeling to the Reinforcing Spirals Model and the Spiral of Silence to better elucidate the complex and dynamic nature of this process. Our preliminary results illustrate how environmental affordances (i.e. social media), closeness of the system and probability of outspokenness may impact how attitudes change over time. Additional analyses are also proposed.

The model is suitable to investigate the effects of different characteristics of apprenticeship programmes both in historical and contemporary societies. The model is built considering five societies, using an agent-based simulation model, we identified six main characteristics which impact the success of an apprenticeship programme in a society, which we measured by considering three parameters, namely the number of skilled agents produced by the apprenticeships, programme completion, and the contribution of programmes in the Gross Domestic Income (GDI) of the society. We investigate different definitions for success of an apprenticeship and some hypothetical societies to test some common beliefs about apprenticeships performance. The model also shows the number of unemployed agents given their work-based skills, wages, and the number of small and large companies who participate in training agents. The model enables exploring the impact of parameters, such as initial wages and the number of training years, along with the stated policies on the system.

Digital social networks facilitate the opinion dynamics and idea flow and also provide reliable data to understand these dynamics. Public opinion and cooperation behavior are the key factors to determine the capacity of a successful and effective public policy. In particular, during the crises, such as the Corona virus pandemic, it is necessary to understand the people’s opinion toward a policy and the performance of the governance institutions. The problem of the mathematical explanation of the human behaviors is to simplify and bypass some of the essential process. To tackle this problem, we adopted a data-driven strategy to extract opinion and behavioral patterns from social media content to reflect the dynamics of society’s average beliefs toward different topics. We extracted important subtopics from social media contents and analyze the sentiments of users at each subtopic. Subsequently, we structured a Bayesian belief network to demonstrate the macro patters of the beliefs, opinions, information and emotions which trigger the response toward a prospective policy. We aim to understand the factors and latent factors which influence the opinion formation in the society. Our goal is to enhance the reality of the simulations. To capture the dynamics of opinions at an artificial society we apply agent-based opinion dynamics modeling. We intended to investigate practical implementation scenarios of this framework for policy analysis during Corona Virus Pandemic Crisis. The implemented modular modeling approach could be used as a flexible data-driven policy making tools to investigate public opinion in social media. The core idea is to put the opinion dynamics in the wider contexts of the collective decision-making, data-driven policy-modeling and digital democracy. We intended to use data-driven agent-based modeling as a comprehensive analysis tools to understand the collective opinion dynamics and decision making process on the social networks and uses this knowledge to utilize network-enabled policy modeling and collective intelligence platforms.