Ford Denison, in his book, Darwinian Agriculture: How Understanding Evolution Can Improve Agriculture (2012) discusses some interesting research on breeding for cooperation among chickens which is reported below. Richard Dawkins recommended this study to him which was originally reported in William M. Muir, 1996, Group Selection for adaptation to multiple-hen cages: selection program and direct responses. Poultry Science, 75, 447-445.

Over much of their evolutionary history, hens have competed among themselves. Those who were more aggressive tended to get more than their far share of food, allowing them to lay more eggs and have more descendants. Although success in fighting increased the relative egg production of the most-aggressive hens, it decreased egg production by the losers even more, thereby reducing total egg production.

So Muir took a novel approach to selecting for increased egg production. Rather than breeding from the individual hens that laid the most eggs, he bred from the groups - four hens per group, raised together - that laid the most eggs. The most-productive groups turned out to be those where hens had less genetic propensity to peck each other. In only six generations of selection, there was a major decrease in fighting-related injuries and a 30 percent increase in the percent of hens laying an egg each day, from 52 percent of hens to 68 percent.

Selection based on group-level traits, like egg production per group, are know as group selection. Group selection deliberately imposed by humans (as in the chicken example just discussed) can have major evolutionary effects. p. 140

The study offers us a nice parable regarding the virtues of cooperation. It also provides a good example of what group selection refers to in the context of evolutionary theory. Group selection is not believed to be a very big factor overall in the evolution of plants and animals as most biologists view evolution as operating at the individual or gene level rather than the group level; however, group selection can be carried on by humans over the plant and animal world as exemplified by this study. The study provides some insight into how production can benefit (e.g, total eggs) when management selects for group-level attributes in addition to, or instead of, individual-level attributes.