Territorial defense involves fighting conspecifics to protect a given resource. In some social systems, organisms form dominant-subordinate relationships that enable them to reduce the amount of fighting while maintaining the integrity of their territorial boundaries. In an attempt to determine how fighting affects dominance relationships in fish, several researchers have staged fights between individuals to investigate if previous winning or losing experiences affect whether an individual wins or loses subsequent encounters. Most of these experiments have found that winners tend to win again and losers tend to lose again. Interestingly, some studies have found that the effect of losing lasts longer than the effect of winning. Bakker et al. (1989) found that male threespine stickleback were more likely to win a second encounter with an unfamiliar rival after a winning experience if the second encounter occurred immediately after the winning experience but not if it occurred three hours later. But, males with a prior losing experience had a zero probability of winning a second encounter when that encounter occurred immediately after or three hours after the losing experience, and they won less than 25% of second encounters if these second encounters occurred six hours after the losing experience. Bakker et al. suggest two reasons for the asymmetry in the effects of fighting experience. The first reason is that winners and losers may undergo different changes in physiology as a result of the fight. The second reason is that winners and losers receive different conditioning experiences because losers operantly control the end of the fight (e.g. by fleeing) while winners do not.

I performed an experiment in which males were given a winning or losing experience outside of their territory with an unfamiliar rival. Four hours later, each male was placed into one side of a plexiglass box and his neighbor, who was housed adjacent to the test male, was placed into the other side of the box. The box, which did not allow physical contact between the males but did permit visual, auditory, and chemical communication, was placed at various locations within each male's territory. I found that males behaved similarly during the interactions with their neighbor regardless of whether they had won or lost the previous encounter. Given this finding, it seems unlikely that physiological changes can explain the loser effect. If physiological changes in the mechanisms that underlie fighting behavior had occurred, then I would have expected them to prevent the expression of aggressive behavior (attacks, biting). With respect to the learning experience of losing, the conditioning does not seem strong enough to prevent the losing fish from exhibiting aggressive behavior when faced with a familiar rival.

Even though fish with prior losing experiences are capable of expressing aggression at the same level as winners, they tend to lose future aggressive encounters. At least two different explanations could account for this phenomenon: (1) pre-existing relationships are not affected by winning or losing a fight (Beacham and Newman 1987), (2) losers are choosing to suppress their aggression when faced with the possibility of physical harm or (3) aggressive behavior does not determine who wins or loses a fight. The first and second explanations would involve cognitive mechanisms and imply that males recognize individuals and that they are able to discern dangerous and non-dangerous situations. These interpretations are supported by the fact that in Bakker et al.'s study, only 29 losers but 60 winners initiated the second agonistic encounter. The third explanation might depend on some rule-of-thumb such as "fight for five minutes and then flee." In this case, persistence during a fight, and not the degree of overt attack, might be the most important factor in deciding who wins or who loses (Francis 1988, Stamps 1994). Thus, losing (i.e., escaping to end a fight) might be a strategy that individuals employ to avoid costly interactions but that is unrelated to a male's ability to obtain a territory. Fitzgerald and Kedney (1987) found that neither aggression by territorial male threespine stickleback towards an unknown rival, in a bottle, nor a winning or losing experience with an unknown rival correlated with ability to establish a territory when in a group context. Hence, instead of resulting from particular changes in physiological mechanisms or conditioning experiences, the loser effect may reflect the fact that losers choose when to attack and that they employ a low persistence strategy. As such, losing would be a specific fighting tactic rather than a consequence of previous experiences.

Literature Cited:

Bakker, T.C.M., Feuth-De Bruijn, E., and P. Sevenster. 1989.Asymmetrical effects of prior winning and losing on dominance in sticklebacks (Gasterosteus aculeatus). Ethology. 82: 224-229.

Beacham, J.L. and J.A. Newman. 1987. Social experience and the formation of dominance relationships in the pumpkinseed sunfish, Lepomis gibbosus. Animal Behavior. 35: 1560-1562.

Fitzgerald, G.J. and G.I. Kedney. 1987. Aggression, fighting, and territoriality in sticklebacks: three different phenomena? Biology of Behaviour. 12:186-195.

Francis, R.C. 1988. On the relationship between aggression and dominance. Ethology. 78: 223-237.

Stamps, J. 1994. Territorial behavior: Testing the assumptions. Advances in the Study of Behavior. 23:173-232.