Case Studies

The Environment

Research Question: We know that our past actions can influence our future choices. Do past self-interested behaviors contribute to difficulty in getting people to act in the interest of society as a whole? In particular, how do past actions hinder the likelihood of someone acting in line with their reported concern for the environment?

Hypothesis: Past actions often serve as a strong signal of who we are and what we care about. Because of this, we predicted that out of a desire to maintain a (temporarily) consistent sense of self, people would be less willing to engage in pro-environmental behavior immediately after taking a self-interested (and anti-environmental action).

Experiment: Subjects were recruited from Amazon Mechanical Turk and assigned to one of three conditions: no hazard note, hazard note, or hazard note with prior spraying. All subjects were given a description of a new pesticide spray, and those in the hazard note and hazard note with prior spraying conditions read that the bug spray was toxic to pets and plants and could harm the environment. Those in the prior spraying condition were then told to imagine they had previously used the spray to kill a moth that morning and were asked to click a red dot (to simulate that action). At this point all subjects rated how likely they were to use the pesticide if they found a spider in their home on a 7-point scale (from very unlikely to very likely).

Results: We found that those in the prior spraying condition were significantly more likely to use the pesticide again than those in the hazard note condition. Specifically, those in the hazard note condition had an average likelihood of 3.17 out of 7 while those in the hazard note with spraying condition had an average likelihood of 7.23 out of 7. We also replicated earlier experiments showing that the tendency to use pesticides is significantly reduced with a hazard note indicating the risks to the environment (when prior spraying is not involved).


Application: As was replicated in our study, highlighting information about risks to the environment can increase pro-environmental actions. But our study also demonstrates that those effects can be cancelled out when people have recently taken anti-environmental actions. This suggests that attempts to change societal norms regarding the environment by providing information on harmful actions may not be as effective as we might have otherwise thought. Because many people will have already taken self-interested actions that harm the environment, the beneficial effect of information may be reduced or eliminated. If we wish to increase pro-environmental behavior, we may need to look to other nudges.