Skip to ContentSkip to Navigation
founded in 1614  -  top 100 university

Hemelrijk group

Mechanisms underlying dominance interactions


Dominance is of major importance to group life, since it often determines access to resources. How dominance hierarchies develop, whether and how they are determined by the self- reinforcing effect of winning and losing fights is under debate.

Required education
Additional subjects
Starting date
Specifics Theoretical, computer model DomWorld

Project description

Lindquist and Chase [1] compare several mathematical models of this to empirical data. This includes our obligate attack model based on self-reinforcing competitive interactions [2] . They describe hierarchical development with new methods, such as bursts of attack, number of counter-attacks (pair flips), details of rank development, rank stability, state occurrences using a directed-graph and network motifs based on interactions between 2-6 individuals, etcetera. While ignoring in our model the spatial effects of interactions and the risk-sensitivity of attacking [3] , they reject all these models. They conclude that the observed hens needed more sophisticated cognitive levels to generate their dominance interactions than is usually assumed.
In their subsequent paper on hierarchical development in cichlids and hens Chase and Lindquist
describe the formation of hierarchies in terms of dynamic network motifs of three individuals. They assume two rules, namely that every individual interacts with all others and that interactions are most frequently transitive rather than intransitive. Transitive interactions imply that when A wins over B and B over C, A will win from C also.
These rules ignore that in reality not all individuals interact with each other and that transitivity is not a mechanism but a description.

The aim of this master thesis is to investigate whether grouping, and competing with winning and losing having self-reinforcing effects and the risk-sensitive way of attacking cause all these patterns to emerge automatically
[3] . The study is done using a computer model of individuals that group and compete, called DomWorld. The self-reinforcing effects in the model will be represented in line with empirical studies with the Elo-rating method [5] . Students will need to program the extensive analysis methods as described by Chase and Lindquist themselves.


[1] Lindquist, W. B. & Chase, I. D. 2009 Data-Based Analysis of Winner-Loser Models of Hierarchy Formation in Animals. Bull. Math. Biol. 71, 556-584.

[2] Hemelrijk, C. K. & Wantia, J. 2005 Individual variation by self-organisation: a model. Neuroscience & Biobehavioral Reviews. 29, 125-136.

[3] Hemelrijk, C. K. 1999 An individual-oriented model on the emergence of despotic and egalitarian societies. P Roy Soc Lond B Bio. 266, 361-369.

[4] Chase, I. & Lindquist, W. 2016
The Fragility of Individual-Based Explanations of Social Hierarchies: A Test Using Animal Pecking Orders. PLoS ONE. 11.

[5] Franz, M., McLean, E., Tung, J., Altmann, J. & Alberts, S. C. 2015 Self-organizing dominance hierarchies in a wild primate population. P Roy Soc B-Biol Sci. 282.

Contact information

For more information, contact Charlotte Hemelrijk

More information

Behavioural & Physiological Ecology, Hemelrijk group - Self-organisation of social systems

Laatst gewijzigd:07 april 2017 12:36