Limits...
Violating social norms when choosing friends: how rule-breakers affect social networks.

Hock K, Fefferman NH - PLoS ONE (2011)

Bottom Line: Using dynamic, self-organizing social network models we demonstrate that defying conventions in a social system can affect multiple levels of social and organizational success independently.Such actions primarily affect actors' own positions within the network, but individuals can also affect the overall structure of a network even without immediately affecting themselves or others.These results indicate that defying the established social norms can help individuals to change the properties of a social system via seemingly neutral behaviors, highlighting the power of rule-breaking behavior to transform convention-based societies, even before direct impacts on individuals can be measured.

View Article: PubMed Central - PubMed

Affiliation: Department of Ecology, Evolution & Natural Resources, Rutgers, The State University of New Jersey, New Brunswick, New Jersey, United States of America. hock@aesop.rutgers.edu

ABSTRACT
Social networks rely on basic rules of conduct to yield functioning societies in both human and animal populations. As individuals follow established rules, their behavioral decisions shape the social network and give it structure. Using dynamic, self-organizing social network models we demonstrate that defying conventions in a social system can affect multiple levels of social and organizational success independently. Such actions primarily affect actors' own positions within the network, but individuals can also affect the overall structure of a network even without immediately affecting themselves or others. These results indicate that defying the established social norms can help individuals to change the properties of a social system via seemingly neutral behaviors, highlighting the power of rule-breaking behavior to transform convention-based societies, even before direct impacts on individuals can be measured.

Show MeSH
A 3-node network examples demonstrating how individual centrality metrics were measured.A) Betweenness centrality: individual in the middle is the necessary intermediary between the left and right individual, as it lies on the shortest path between those individuals; the middle individual therefore has higher individual betweenness centrality than the other two individuals; B) In-degree centrality: both left and right individual have connections towards the middle individual, making the middle individual ‘popular’ as a partner; middle individual therefore has higher individual in-degree centrality than the other two individuals.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3198795&req=5

pone-0026652-g001: A 3-node network examples demonstrating how individual centrality metrics were measured.A) Betweenness centrality: individual in the middle is the necessary intermediary between the left and right individual, as it lies on the shortest path between those individuals; the middle individual therefore has higher individual betweenness centrality than the other two individuals; B) In-degree centrality: both left and right individual have connections towards the middle individual, making the middle individual ‘popular’ as a partner; middle individual therefore has higher individual in-degree centrality than the other two individuals.

Mentions: Individuals in our networks chose ‘friends’ using two different measures of social centrality (Fig. 1; see also [6], [40], [43]), either according to the partner's quality as a necessary intermediary between others (also called betweenness) or according to the partner's popularity (also called in-degree), to determine which affiliations to maintain. While these measures may be considered proxies for any evaluative metric by which a self-organizing social system yields differences in centrality among individuals [6], populations that employ such centrality measures or their proxies, and their potential implications for individual fitness, have been documented in real-world networks [13], [14], [22]. Individuals that did not follow these rules of conduct instead changed their social partners at random, thus breaking the social conventions used by the rest of the group. Initially, all affiliations were assigned randomly and all individuals had the same probability of being chosen as partners. This random initial structure then dictated each individual's desirability as a partner according to the assigned affiliation criteria: for individuals that followed conventions, the more prominent the individual's social position, the more likely others were to remain affiliated with it. The criteria for affiliation were treated as simple social conventions [44], [45]: while they may or may not have specific fitness analogues in particular systems, in our simulations they served only to drive the self-organizing behaviors in non-deterministic social systems.


Violating social norms when choosing friends: how rule-breakers affect social networks.

Hock K, Fefferman NH - PLoS ONE (2011)

A 3-node network examples demonstrating how individual centrality metrics were measured.A) Betweenness centrality: individual in the middle is the necessary intermediary between the left and right individual, as it lies on the shortest path between those individuals; the middle individual therefore has higher individual betweenness centrality than the other two individuals; B) In-degree centrality: both left and right individual have connections towards the middle individual, making the middle individual ‘popular’ as a partner; middle individual therefore has higher individual in-degree centrality than the other two individuals.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3198795&req=5

pone-0026652-g001: A 3-node network examples demonstrating how individual centrality metrics were measured.A) Betweenness centrality: individual in the middle is the necessary intermediary between the left and right individual, as it lies on the shortest path between those individuals; the middle individual therefore has higher individual betweenness centrality than the other two individuals; B) In-degree centrality: both left and right individual have connections towards the middle individual, making the middle individual ‘popular’ as a partner; middle individual therefore has higher individual in-degree centrality than the other two individuals.
Mentions: Individuals in our networks chose ‘friends’ using two different measures of social centrality (Fig. 1; see also [6], [40], [43]), either according to the partner's quality as a necessary intermediary between others (also called betweenness) or according to the partner's popularity (also called in-degree), to determine which affiliations to maintain. While these measures may be considered proxies for any evaluative metric by which a self-organizing social system yields differences in centrality among individuals [6], populations that employ such centrality measures or their proxies, and their potential implications for individual fitness, have been documented in real-world networks [13], [14], [22]. Individuals that did not follow these rules of conduct instead changed their social partners at random, thus breaking the social conventions used by the rest of the group. Initially, all affiliations were assigned randomly and all individuals had the same probability of being chosen as partners. This random initial structure then dictated each individual's desirability as a partner according to the assigned affiliation criteria: for individuals that followed conventions, the more prominent the individual's social position, the more likely others were to remain affiliated with it. The criteria for affiliation were treated as simple social conventions [44], [45]: while they may or may not have specific fitness analogues in particular systems, in our simulations they served only to drive the self-organizing behaviors in non-deterministic social systems.

Bottom Line: Using dynamic, self-organizing social network models we demonstrate that defying conventions in a social system can affect multiple levels of social and organizational success independently.Such actions primarily affect actors' own positions within the network, but individuals can also affect the overall structure of a network even without immediately affecting themselves or others.These results indicate that defying the established social norms can help individuals to change the properties of a social system via seemingly neutral behaviors, highlighting the power of rule-breaking behavior to transform convention-based societies, even before direct impacts on individuals can be measured.

View Article: PubMed Central - PubMed

Affiliation: Department of Ecology, Evolution & Natural Resources, Rutgers, The State University of New Jersey, New Brunswick, New Jersey, United States of America. hock@aesop.rutgers.edu

ABSTRACT
Social networks rely on basic rules of conduct to yield functioning societies in both human and animal populations. As individuals follow established rules, their behavioral decisions shape the social network and give it structure. Using dynamic, self-organizing social network models we demonstrate that defying conventions in a social system can affect multiple levels of social and organizational success independently. Such actions primarily affect actors' own positions within the network, but individuals can also affect the overall structure of a network even without immediately affecting themselves or others. These results indicate that defying the established social norms can help individuals to change the properties of a social system via seemingly neutral behaviors, highlighting the power of rule-breaking behavior to transform convention-based societies, even before direct impacts on individuals can be measured.

Show MeSH