A basic assumption of human nature is that we all want the greatest reward for the least amount of effort. Take for example the students in my class. Often they apply just enough effort to earn a satisfactory grade (e.g., a low B or high C), while avoiding extra-credit assignments like the plague.
But in some rare instances, people will ignore the reward and put forth more effort than is necessary to achieve a goal. Take for example the (arguably) best team in the American football league, the New England Patriots. The Patriots are notorious for going after what is known as the two-point conversion, even in instances when the extra point is not necessary. A two-point conversion occurs after a team earns a touchdown (6 points) and instead of kicking a field goal for 1 additional point (total of 7 points), they score essentially another touchdown (from a very short distance from the goal line) for 2 additional points (total of 8 points).
An example of a two-point conversion can be seen during last year’s Super Bowl:
According to “rational” economic models, such as a reward maximizing model, it would seem “irrational” for a team to exert additional effort for the little reward gained from the two-point conversion. But what if the team were operating with multiple goals in mind? What if they not only wanted to beat their opponent, but had a second goal of creating as large of a gap in score as possible? In other words, how does the motivation of multiple goals result in a departure from objective reward maximizing behavior (a.k.a. doing as little as possible to gain the greatest amount of reward)?
This question was posed by Timothy Ballard and colleagues Simon Farrell and Andrew Neal in a recent article published in the Psychonomic Society journal, Psychonomic Bulletin and Review, where they sought to quantify the subjective value of goal achievement.
More specifically, Ballard and colleagues assessed whether people discount the value of achieving one goal (e.g., first touchdown) by measuring the extent to which they attempt to achieve both goals (e.g., touchdown and two-point conversion), especially when doing so decreased the expected reward overall.
Ballard and colleagues predicted that prioritizing the more difficult goal, at a certain point, would increase the likelihood that both goals would be achieved, but it also runs the risk that neither goal would be achieved. This would happen if a participant made too many decisions trying to gain points toward the more difficult goal in the trial instead of securing enough points toward the easier goal first. The participant would run the risk of allocating too many decisions toward the difficult goal with not enough decisions left in the trial to satisfy the easier goal. In contrast, prioritizing the easier goal would increase the likelihood that one goal would be achieved, but decrease the likelihood that both goals would be achieved.
In their study, participants were asked to make multiple decisions in an effort to earn (or maintain) a certain amount of blue points (goal 1) and green points (goal 2). Importantly, one goal was always more difficult to achieve than the other. An example of the options was “80% chance to gain 1 blue point and a 20% chance to gain 1 green point versus the opposite”.
The experimental design used three pertinent manipulations: goal frame, incentive structure, and start position. The figure below shows a flow diagram of the task with all manipulations.
The goal frame had two levels: approach and avoidance. In the approach condition, participants started the task with an insufficient amount of points and they needed to perform an action (i.e. select an option) that would gain points toward each of their goals. Suppose for instance a participant selected the option “80% chance to gain 1 green point and 20% chance to gain 1 blue point” (see the top-left panel in the preceding flow diagram). There are three possible outcomes for selecting this option: they could gain 1 green point and 1 blue point, gain only a green point (since a 20% chance for gaining a blue point is relatively low), or they could gain no points for either goal (since both goals are based on chance and not 100% guaranteed).
In the avoidance condition, participants started with a sufficient number of points and needed to avoid losing points for either goal. In this case, they could lose points toward both goals, lose points towards only one goal, or not lose points toward either goal.
The incentive structure varied across participants. People were told that they would be compensated based on their performance on a randomly selected trial. If both goals were achieved on that trial, they would receive $10, and if only one goal (regardless of which one) was achieved, they would receive one of the following amounts: $0, $2.50, $5, $7.50, or $10. The incentive amounts varied between participants, but not between trials. If the incentive for achieving only one goal was $0, then participants could not earn any money unless they achieved both goals and if the incentive was $10, participants could earn the full amount by achieving the first goal, and no additional money was allocated for achieving both goals.
The last manipulation was the starting point, where the experimenters varied the number of blue and green points participants received at the beginning of each trial series.
There were several interesting and noteworthy results. First, people were found to focus more on achieving only one goal as the incentive to achieve only one goal increased. Similarly, people prioritized the second goal less frequently as the incentive for achieving the first goal increased. Also, people prioritized the more difficult goal more often in the avoidance condition than in the approach condition. In the avoidance condition, people still prioritized the more difficult goal even when the full reward could be earned from achieving a single goal.
To provide a more rigorous interpretation of the results, Ballard and colleagues developed a model for assessing participants’ subjective values for achieving various goal outcomes and compared the results of the subjective model to a reward maximizing model. As expected, they found that the subjective value of achieving goals increased as the monetary incentive increased.
Unexpectedly, compared to the maximizing reward model, participants placed less subjective value on achieving only one goal (and thus prioritizing both goals), even in conditions where the incentive for achieving only one goal was higher. Although participants prioritized achieving only one goal as the incentive increased, they did not place a higher subjective value on achieving only one goal, relative to achieving both goals.
For example, in the $10 incentive condition, participants received the full amount of reward for achieving only one goal, but behaved as if they valued achieving the first goal as 60% (in the approach condition) as valuable as achieving both goals. This suggests that people downgrade the value of achieving the first goal, and are even willing to forego a monetary reward in order to achieve both goals.
So the moral to the story is that, yes sometimes people will only put forth as much effort as required to achieve the maximum amount of reward, but when they have different goals in mind, say for example, winning by as many points as possible (as opposed to just winning), people may deviate from this way of thinking due to the psychological value placed on achieving goals.
Psychonomics article featured in this post:
Ballard, T., Farrell, S. & Neal, A. (2017). Quantifying the psychological value of goal achievement. Psychonomic Bulletin & Review, DOI: 10.3758/s13423-017-1329-1.