Fish select their leaders by consensus, according to a study published in a recent issue of the journal Current Biology.
David Sumpter and coauthors studied how groups of Three-spined sticklebacks (Gasterosteus aculeatus) made decisions about which fish to follow when presented with a choice.
The authors presented groups of Three-spined sticklebacks with two fish images differing in characteristics, including size, corpulence (fatness), shade, and spottiness, that convey information about the fitness of the individual (eg. abdomen profile can indicate success in foraging, while small black spots may indicate a parasite infestation).
Eleven different images were created, depicting small, medium, or large; fat, medium, or thin; light, medium, or dark; and spotted or plain fish. After carrying out the experiments with varying numbers of focal fish (one, two, four and eight), the authors found that the sticklebacks preferred to follow large fish to small fish, fat to thin, and healthy to ill (ie. dark to light and plain to spotted).
This in itself was not surprising, but the fact that a positive social feedback mechanism seemed to increase the preferences as group size increased, was.
The authors then applied mathematical models to three hypotheses of decision-making (individual, aggregate and majority), and compared the results of their experiments with these models.
It was found that the sticklebacks employed what is known as a quorum-response rule when deciding which fish to follow: "Individuals watch the decisions of others before committing themselves to a decision.
"In the model, one or two individuals sometimes take the less attractive option initially, but usually a larger number of individuals have taken the more attractive option.
"Undecided individuals are biased toward also taking the option that is more popular, and this choice becomes amplified."
The unfortunate side effect of implementing consensus is that in a few of the trials, the fish tended the follow the less attractive model. The authors attribute "...submission to peers and occasional cascades of incorrect decisions... a by-product of what is usually accurate consensus decision-making."
For more information, see the paper: Sumpter, DJT, J Krause, R James, ID Couzin and AJW Ward (2008) Consensus decision making by fish. Current Biology 18, pp. 1773"1777.