Ohio State nav bar

Spring 2018

Winners of the 2018 DSC Research Forum Best Abstract

 

Daniel Zane, Robert Smith, and Rebecca Walker Reczek

The Meaning of Distraction: How Metacognitive Inferences from Distraction Affect Brand Attitudes
 
Consumers often encounter advertisements in the background while primarily focused on other tasks. In this research, we show that consumers draw inferences about how interested they are in a brand advertised in a background ad based on how distracted they are by the background ad. We demonstrate that when consumers find themselves more distracted than expected by a background advertisement, they often infer that they are interested in the advertised brand, based on a lay theory that distraction signals interest. However, we also explore various factors that determine whether consumers draw such inferences or whether they draw different inferences.

 

Kirby Nielsen

Preferences for the Resolution of Uncertainty and the Timing of Information 

We experimentally investigate preferences over the timing of non-instrumental uncertainty resolution. We examine whether individuals prefer uncertainty to resolve gradually or all at once, and whether they prefer early or late resolution. Additionally, we compare choices over two-stage lotteries to analogous choices over information structures. We find a persistent preference for early, informative signals, but a stronger preference for late resolving lotteries, with limited evidence of a preference for one-shot resolution in either domain. Subjects are willing to pay a substantial percentage of the lottery's expected value to obtain their preferred resolution, often paying to avoid information. 

 

DSC Research Forum Poster Abstracts

 

Lorraine Borghetti

Theory of Mind for Automated Agents:  Inferring the Intentions of an Imperfect Agent Teammate

Automated agents increasingly team with human operators in complex industries such as nuclear power plants, electrical grids, network security, etc. However, their partnership is challenged by the lack of transparency of and limited means to communicate with the agent. Using an air traffic control game for our experiment, we found that decreases in agent reliability (increases in mistakes) led human teammates to want to apply theory of mind to infer intentions of the agent and to communicate with the agent to a greater degree. We further found that intention inferences and communication needs increased as trust in the agent declined.

 

Gillian Davis, Ellen Peters, and Brittany Shoots-Reinhard 

Increasing Climate Change Belief Through Source Matching

Despite the scientific consensus that human-caused climate change is occurring, 21% of Americans disagree (Pew Research Center, 2016). Researchers have investigated ways to combat this disbelief, such as providing participants with this consensus information (e.g., van der Linden et al., 2015). However, this approach can backfire, especially for Republicans motivated to disagree with scientists (e.g., Bolsen & Druckman, 2016; Kahan, 2013). One possibility to reduce motivated cognition is to provide climate change messages from sources that Republicans trust (i.e., military vs. scientists). This study explores this method for increasing belief in climate change in those who are difficult to persuade.

 

Daniel Kent and Daniel Silverman

Can Entrenched Factual Misperceptions be Corrected? An Experiment on Public Fear of Terrorism
 
Can objectively false political beliefs be corrected? To address this question we conduct a survey experiment on 800 participants, investigating how responsive individuals are to factual information about the risks posed by terrorism, about which a particularly costly and sizeable misperception exists among the U.S. public. We find that misperceptions about the threat posed by terrorism are correctible, but corrections reinforced by a partisan cue are most effective and in some cases necessary. Moreover, despite minor attenuation, effects hold over two-weeks. These findings suggest the public is capable of accurately updating beliefs toward difficult issues when given the right information. 
 

 

Austin Knuppe 

Local Partners for Local Problems: When Does Security Assistance Undermine Public Support for Local Combatants? 
 
How does security force assistance affect civilian support for local combatants in fragile states? This projects investigates how foreign assistance affects civilian attitudes toward local combatants. Drawing on original experimental evidence from Baghdad, Iraq, I argue that public beliefs and attitudes toward foreign intervention are shaped by civilians’ exposure to and experience with local combatants. While most members of the public lack strong prior opinions about the legitimacy and efficacy of specific security policies, civilians’ relationship to local combatants provides a reference point for developing attitudes about which foreign actors to trust and under what specific conditions. I draw on a series of embedded survey experiments – including an endorsement experiment, conjoint analysis, and vignette experiment – as an empirical test of my claim that local combatants mediate civilian attitudes about foreign intervention. The pre-analysis plan and survey instrument were pre-registered with Experiments in Governance and Politics (EGAP) prior to the enumeration of the survey in November 2017. 
 
 
Cameron Luther and Ian Krajbich
 
Integrating Current and Prior Information in a Social Learning Game with the Drift Diffusion Model 
 
This research builds on a working paper by Krajbich and Frydman that investigates whether subjects learn from others’ response times in social learning situations. In this project we use the drift diffusion model to investigate how participants integrate current and prior information into their decisions. Specifically, we investigate whether prior information in favor of one decision would bias subjects’ choices or be integrated with the new information at the time of decision. We found evidence for a prior bias, but only when response times were available. 
 
 
Nicholas O’Dell and Duane Wegener
 
Decision-making Ability Beliefs
 
The Decision-making Ability Beliefs (DAB) scale assessed post-choice satisfaction in a two-alternative forced-choice paradigm. The speeded subscale (SDAB) was predictive of satisfaction (Speed X SDAB, F(1,298) = 8.34 p =.004.  Individuals who perceived themselves to possess a low self-efficacy in speeded contexts demonstrated significantly lower satisfaction, than their high-self efficacy counterparts, with the presence of time pressure.   The deliberative subscale (DDAB) was capable of predicting satisfaction in non-speeded contexts (Speed X DDAB, F(1,298) = 7.19 p = .008).  Satisfaction was reduced in speeded, or time constrained, contexts. The SDAB predicts satisfaction in speeded contexts above and beyond existing decision-making metrics.  
 

 

Danyi Qi

Messages to Diners about Food Waste Result in Smaller Orders, Cleaner Plates, but No Change in Total Food Intake
 
Consumer food waste reduction policies operate through one or both of the constituent behaviors of food selection and food intake, where changes in each behavior hold distinct implications for social welfare. To explore these policy tensions, we estimate the impacts of common consumer food waste interventions on each behavior in a dining experiment. We find out that informational interventions advance sustainability goals by reducing food waste, help foodservice managers reduce operation costs associated with food procurement, and help hunger relief agencies source more unserved food. Nutritionists may worry that consumers could develop a habit of cleaning their plates. 
 
 

Mary Kate Tompkins, Ellen Peters, Sarah Garver, Venson Banda, Sarah Huber, Kunuwo Fokong, Abigail Norris Turner, and Alison Norris

Numeric competencies and STI/HIV outcomes in rural Malawi

STIs are highly prevalent in Malawi; few studies have examined how decision-making abilities influence STI risk. This study examined the potential role of objective numeracy (ONS, objective mathematical skills) and subjective numeracy (SNS, perceived numeric ability) in health outcomes that relate to individuals’ utilization of strategies to avoid and treat STIs. In prior research, greater numeracy was associated with beneficial decision-making processes and a greater reported likelihood of condom use to protect against HIV. We predicted that greater ONS and/or SNS would be associated with fewer laboratory-confirmed STIs that result from sexual-health decisions. Data collection is in progress.