Journal Publications
& Brief Summaries of Each

In 36 lab experiments and a field experiment with all US traffic on a major retailer's site, we compare two ways of structuring a purchase decision. People are much more likely to make a purchase when the whether to buy and how many to buy decisions are integrated into one simultaneous choice rather than being resolved sequentially. One reason for this is that "quantity-integrated" choices anchor people later in the buying decision process.

Conversations around improving mental health care typically focus on bolstering supply. We give a complementary perspective on how behavioral science principles and techniques can help manage demand and maximize returns from mental healthcare engagement.

We show that the "experientialness" and "materialness" of a good are better represented as separate features rather than opposing ends of the same construct. Through disentangling the two, and testing hundreds of different products, we show that purchases high on both factors tend to yield the highest happiness.

We show that the structure of a task can induce a feeling of fluency. When completing tasks that have superficially consistent structures (e.g., solving number sequences all with the same blank position), people experience "procedural fluency," which increases their confidence in their performance.

We show that social norms are subtly communicated through the structure of an incentive. Specifically, loss-framed incentives like surcharges convey injunctive and descriptive social norms about the incentivized behavior. As a result, having a surcharge (vs. discount) in an environment can have rippling effects on behavior elsewhere in the community.

We show that feelings of guilt come from two sources: “decision guilt” (tied to the initial choice) and “action guilt” (tied to the behavior itself). When time passes between when we decide to perform some guilt-inducing action and when it actually happens, decision guilt can decay in this interim period. As a result, experiences structured with "decision-enactment gaps" let people feel less guilty during their actions, leading them to indulge more and atone less.

We study the certainty effect: consumers' preference for certain over uncertain/risky rewards. We highlight how task structure features that lead to reliance on gist versus details (time pressure, intuition, reward values, etc.; design features that are used commonly in JDM research) accentuate—and hence may overestimate—the preference for certainty.

The mere presence of our smartphones in our vicinity, even if they are completely powered off, detracts from available cognitive resources. This happens because we automatically orient our attention toward monitoring our phones, just like parents automatically attend to the sound of a baby's cry. The detrimental effects of phone presence are more pronounced for people who have higher smartphone dependence.

Policy preferences can be asymmetric: a protest may feel ethical when a cause we support does it, but not when our opponent does. Will people openly endorse this asymmetry within-subjects? We find that within-subjects designs weaken, but do not eliminate, individuals' tendencies to "speak out of both sides of their mouths." Liberals (vs. conservatives) are more motivated to appear unprejudiced, but equally prone to this tendency. More broadly, we comment on the choice of between- vs. within-subjects research designs for psychological inference.