War of Attrition: A Lab story

War of Attrition: A Lab story

By Shane Galvin

In psychology aggression is defined as the willingness to cause harm to another. This harm view is a highly contested definition of aggression and academics don’t seem to be able to really agree on what aggression is (for example, see back and forth between Christopher Ferguson and colleagues and Bushman and Anderson). Aggression is a social construct, in that it is a descriptive term for a wide variety of behaviours. As a consequence, it is largely the idea of harm caused that is the operationalised measure of aggression in lab tasks. There have been many heavy criticisms of lab research; One of which is the lack of a non-aggressive option for participants to interact with their opponents (Tedeschi & Quigly, 1996; Elson, Mophseni, Breuer, Scharkow & Quandt, 2014). This is exemplified in the most commonly used experimental task, the Taylor Aggression Paradigm (TAP) (Taylor, 1967); TAP conceives of aggression as the amount of harm that a person will deliver to another as the amount of electric shock. So when I sat down with my project supervisor, I considered what approaches may be best to look at phenomena that precede aggression. I had read a paper on using a game theory bargaining game to look at co-operative and uncooperative behaviours in relation to the Machiavellian construct and thought ‘maybe a similar strategy can be applied to aggression’. In my mind, competitive or conflict scenarios involve multiple factors, including persistence in the face of accumulation of costs, as well as infliction, so I decided to use the War of Attrition (WoA) (Maynard Smith, 1974).

WoA is basically a game of chicken – the first person to swerve (or forfeit, in this case) loses. In WoA, multiple rounds of the chicken game are played, and cost is accumulated over multiple rounds. Just imagine two animals fighting over a food resource; every moment they spend fighting costs them energy that they could be using to search for an alternative food source. The resource that the conflict is about should be worth the effort of fighting. When the resource value becomes less than the cost of the conflict, lesser aggressors should withdraw from the conflict, after accumulating their cost. The winner of the conflict exits with the value of the resource less the costs that they have accumulated.

In this lab task, participants compete for a potential prize value of €50 in a button pressing game. In each round a participant presses a button to fight and continue to the next round or to forfeit and exit the game. If their opponent also presses fight, the game continues. However, there is a cost of continuing; the €50 potential prize value depletes by €2 for every round both participants fight. Aggression in the WoA is classified based on whether the participant totally depleted their potential prize value, thus reaching a €0 prize value (Fig. 1).


Figure 1 WoA Flow Chart

I really think for a lab test to succeed there is a need for a non-aggressive interaction with opponents; rendering the procedure fully falsifiable. For example, if a co-operate option could be randomly presented to participants alongside other task options, the participant may offer co-operation, or accept or reject co-operation if offered. If some participants continue to fight, or refuse to offer co-operation, reaction time or bio-signal variation between and within groups can be analysed to capture a snapshot of the processes that occur to support aggressive behaviour. Although, I would bet there would be a lot less chickens in that kind of game.



Elson, M., Mohseni, M. R., Breuer, J., Scharkow, M., & Quandt, T. (2014). Press CRTT to measure aggressive behavior: The unstandardized use of the competitive reaction time task in aggression research. Psychological assessment, 26(2), 419-432.

Maynard Smith, J. (1974). The theory of games and the evolution of animal conflicts. The Journal of Theoretical Biology, 47(1), 209-221.

Taylor, S. (1967). Aggressive behavior and physiological arousal as a function of provocation and the tendency to inhibit aggression. Journal of Personality, 35(2), 297-310.

Tedeschi, J. T., & Quigley, B. M. (1996). Limitations of laboratory paradigms for studying aggression. Aggression and Violent Behavior, 1(2), 163-177.