الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 32 |  |  |
| | | from | 32 | | |
Abstract
Altruism has been a puzzling behavior studied by many scholars. Understanding the
behavior is important to better predict decision making process outcomes, gain better
insights into evolution and human nature. Work has been conducted to analyze altruism,
how it is a↵ected by reputation, empathic agents and spatial structure. The purpose of
this thesis is to study altruism di↵usion in social networks. The main contribution of the
work is answering two questions, how social networks a↵ect the di↵usion of the altruistic
behavior and how the structural characteristics of the network alters the di↵usion process.
Evolutionary game theory, particularly the Ultimatum Game and complex networks
analysis for a Barabasi-Albert network form the foundations of this thesis.
In the first part of the work, a comparison between altruism di↵usion in a social
network Vs. its di↵usion in a well mixed population was conducted. A model simulating
the Ultimatum Game on a Barabasi-Albert network was developed to understand how the
di↵usion di↵ers between the two environments. The work demonstrated that starting from
an entirely selfish population, the structure of the social network limits the interactions
between agents and thus limits the di↵usion of the altruistic behavior.
The second part of this thesis is concerned with analyzing the centrality e↵ect on
the di↵usion. A model was developed to simulate the altruistic behavior di↵usion in a
Barabasi-Albert network comparing between starting from central and non-central agent.
Centrality was studied based on two main measures betweenness and degree centrality.
The work demonstrated that agents with high betweenness centrality have a higher ability
to spread the behavior while high degree centrality agents have an opposite e↵ect. The
ability of agents with high degree centrality to spread altruistic behavior is limited by
their selfish neighborhood e↵ect which indicates the importance of neighborhood nature
and size in the di↵usion process.