The decision to check in multialternative choices and limited sensitivity to default

Abstract

Given the rapid proliferation of smartphone applications and data aggregation websites, in many situations people can use decision aids to guide their choices. For example, they may consider whether to use a navigation device to check the fastest route or whether to use a price comparison website to find the cheapest deal. In what circumstances will subjects use a costly comparison decision aid (which I refer to as "checking") to choose for them? In six studies, I investigate the impact of the number of available alternatives and checking's attractiveness on the decision to check. While at first increasing the attractiveness of checking led to higher checking rates, a further increase in the number of available alternatives (and thus checking's attractiveness) did not increase the checking rate. Surprisingly, even when checking had a 40% higher expected value compared with not checking, the observed checking rate was below 45%, contrary to risk and ambiguity aversion predictions. Furthermore, labeling the checking alternative as the default had no impact on its choice rate. I find large individual differences in decisions to check. Surprisingly, subjects' initial decisions had high predictive power over their subsequent checking rates, even after 100 trials with full feedback. I propose two simple learning models that capture well the aggregated results. K E Y W O R D S checking decisions, decisions from experience, inertia, information search, two-sample 1 | INTRODUCTION "Comparison is the death of joy." Mark Twain Imagine that you want to book accommodation in a foreign city. Even a small city may have scores of different hotels, not to mention hundreds of other alternatives, such as apartments for rent through Airbnb. These alternatives may differ in many dimensions-price, location, cleanliness, etc. Today's consumers constantly face choice decisions between a varying number of alternatives: which road to drive to work, where to book a room for their next vacation, or which breakfast cereal to choose. To help them resolve these decisions, consumers also have access to an exploding number of decision aids, such as price and quality comparison websites and apps. The present paper asks how consumers make decisions in this environment, and in particular, under what circumstances they use available decision aids. An abundant literature examines how consumers choose between products in multidimensional (multiattribute) settings, going back to the famous MouseLab studies (e.g., Payne, 1976; Payne, Bettman, & Johnson, 1988). The general findings of these studies show that people often make suboptimal decisions, whether because of confusion, required cognitive effort or task complexity (Stark & Choplin, 2009). More recent studies propose additional factors behind consumers' suboptimal decisions, such as time constraints (Dhar & Nowlis, 1999),

Attention Checks in Decisions From Experience: The Case of Checking Experiments

“…The initial checking rate of the attentive participants was 0.49 (SD = 0.51), and the mean checking rate was 0.25 (SD = 0.29). The mean checking rate was almost identical to 24% (SD = 0.35), the checking rate observed in Roth (2020). Appendix 2 shows that the results replicated on other measures as well including inertia rates and the first choice as a predictor of sequential choices.…”

“…3 Participants then proceeded to the main task. This task was a variant of the clicking paradigm (Barron & Erev, 2003;Erev & Haruvy, 2016); namely, the checking paradigm employed in Roth (2020) which was programmed using the oTree platform (Chen, Schonger, & Wickens, 2016). On each of the 100 trials in this task, the participants had to choose one out of five or 25 alternatives (depending on the experiment) or the checking alternative (see Figure 1).…”

“…The data from the laboratory experiments are taken from Roth (2020). Thus, the experiments serve as a non-trivial replication since they were not only administered in different media, but also tapped a completely different population composed of students in Israel as compared to an older and probably less educated USA MTurk population (e.g., Bolton, Dimant, & Schmidt, 2020;Huff & Tingley, 2015) with different psychological traits (Ophir, Sisso, Asterhan, Tikochinski, & Reichart, 2019;Stone, Walentynowicz, Schneider, Junghaenel, & Wen, 2019).…”

“…To help shed light on this issue, this study examined whether subjects who read the instructions carefully, as assessed on a simple attention check, differ from those who do not read the instructions on a series of five decisions from experience checking experiments (Roth, 2020;Roth, Wänke, & Erev, 2016). Checking experiments are particularly well suited for this comparison.…”

“…Data are available from similar, recent experiments run in the lab, thus enabling a clear comparison between the two settings. Decisions from experience in general, and checking experiments in particular, allow participants to acquire abundant experience during the experiment from their own behavior (see recent reviews in Erev &Haruvy, 2016 andRoth &Yakobi, 2022). Thus, the decision not to read the experimental instructions carefully and instead learn through the experiment might be perfectly reasonable based on rational ignorance (Downs, 1957;Jerath & Ren, 2021) or certain beliefs (Erev, 2020;Gill & Rosokha, 2020); e.g., that the instructions are uninformative, useless, or even deceptive, similar to the decision not to read the instructions for a new appliance but rather learn how to use it by trial and error.…”