Daniel J. Acland

I use data from a short, repetitive online game to explore the role of Flow Theory in motivation and game play. For each player, the Flow-Theory channel in which they are most likely to continue playing the game is identified, and players are categorized into types accordingly. Boredom and Relaxation types are most common. Flow types are among the least common, making up 11% of players. Flow types have the lowest skill level, but challenge themselves the most, and are most likely to make use of self-control devices available within the game. Control types play most frequently and over a longer period of weeks. Apathy types are high skill but seek out low challenges and are least likely to make use of self-control devices. Flow and control types are more likely to play during the workday. Relaxation, boredom and apathy types are more likely to play during workday evenings.

Benefit-cost analysis (BCA) is typically defined as an implementation of the potential Pareto criterion, which requires inclusion of any impact for which individuals have willingness to pay (WTP). But this definition is incompatible with the widely agreed upon exclusion of certain impacts such as rights and distributional concerns, because, as I demonstrate, individuals do have WTP for these impacts. I propose a new definition: BCA should include only those impacts for which we believe consumer sovereignty should govern. This definition is rooted in the observation that WTP, as a measure of value, implicitly preserves consumer sovereignty, and is thus only useful and appropriate for impacts for which that is what we desire. I propose a practical rule for implementing my definition in real-world BCA, based primarily on the contention that any rule should be biased against inclusion of impacts that should not be included, so-called type II error. I test my rule against some common impacts that, by consensus, are either included or not included, and include that in practical application it performs well.

We conduct an experiment on an online game, exploring the effect on
gameplay behavior of voluntary commitment devices that allow players to limit their gameplay. Approximately 25% of players use the devices. Median and 75th percentile device users use devices approximately 60 and 100% of the time, respectively. Players who chose to use the device were those who had previously played longer and more frequently than those who chose not to use the device. Offering the commitment devices decreased session length and session frequency by 2.8 and 6.1%, respectively, while increasing weeks of play by 5.5%. Our results are consistent with some players having self-identified self-control problems, leading to longer and more frequent play than they would prefer, and to demand for commitment, and also with commitment devices creating a more rewarding experience, leading to longerlasting involvement with the game. Our results suggest incentivizing or requiring commitment devices in computer games.

Party polarization is a central feature of American political life, and a robust literature has shown that citizens engage in partisan-confirmation bias when processing political information. At the same time, however, recent events have highlighted a rising tide of anti-government populism that manifests on both sides of the aisle. In fact, data show that large proportions of both Democrats and Republicans hold negative views of government. Using an original set of survey experiments, we examine the psychology of public-sector evaluation. We find that citizens engage in a process of confirmation bias when they encounter new information, which is driven not by party and ideology but by beliefs about the quality and efficiency of government. Taken together, our findings suggest important limitations to citizens’ capacity to learn about public administration, and expand our understanding of what drives confirmation bias with respect to public and private service provision.

Le Grand and New, in their recent book, “Government Paternalism: Nanny State or Helpful Friend,” present a novel definition of paternalism and a framework for thinking about whether any given paternalistic policy can be considered justifiable. I show that their framework is flawed in that it restricts justifiable paternalism to that which is intended to alter individuals’ judgment about the means they use to pursue their self-determined ends. I show that the principles they use to justify certain kinds of means paternalism also justify certain kinds of ends paternalism. In particular, when there is a body of rigorous social-science evidence that individuals select ends that they themselves, if they had adequate information or experience would prefer not to pursue, and when other conditions are met, ends paternalism may be considered to improve the wellbeing of the individual as determined by the individual themselves. I present examples of policies that could be justified under this framework, and offer cautionary notes.

We implement a gym-attendance incentive intervention and elicit subjects' predictions of their postintervention attendance. We find that subjects greatly overpredict future attendance, which we interpret as evidence of partial naiveté with respect to present bias. We find a significant postintervention attendance increase, which we interpret as habit formation, and which subjects appear not to predict ex ante. These results are consistent with a model of projection bias with respect to habit formation. Neither the intervention incentives, nor the small posttreatment incentives involved in our elicitation mechanism, appear to crowd out existing intrinsic motivation. The combination of naiveté and projection bias in gym attendance can help to explain limited take-up of commitment devices by dynamically inconsistent agents, and points to new forms of contracts. Alternative explanations of our results are discussed.