Events Old

Register here for the Spring 2022 Behavioral Science Authors Virtual Event Series

All BCFG Virtual events are recorded. Videos will be posted below within 7 days of the event. If you have questions, or need additional information, please email bcfg-comms@wharton.upenn.edu.

Spring 2022 BCFG Virtual Events

Date

Speaker

Topic

*video links below

February 4, 2022

John List, Kenneth C. Griffin Distinguished Service Professor of Economics, University of Chicago

Interviewed by Dana Suskind, Professor of Surgery, University of Chicago

The Voltage Effect

January 21, 2022

Ayelet Fishbach, Jeffrey Breakenridge Keller Professor of Behavioral Science and Marketing, University of Chicago

Get it Done

Fall 2021 BCFG Virtual Events

Date

Speaker

Topic

*video links below

September 24, 2021

Jay Van Bavel, Associate Professor of Psychology & Neural Science, New York University

The Power of Us

Listen to the interview on Knowledge@Wharton

October 15, 2021

Eric Johnson, Norman Eig Professor of Business, Columbia University

Interviewed by Colin Camerer, Robert Kirby Professor of Behavioral Finance and Economics, Caltech

The Elements of Choice

November 5, 2021

Alison Buttenheim, Associate Professor of Family and Community Health and Health Policy, Scientific Director of CHIBE, School of Nursing & Perelman School of Medicine, University of Pennsylvania

Angela Duckworth, Rosa Lee and Egbert Chang Professor of Operations, Information, and Decisions & Psychology, Co-Director of Behavior Change for Good Initiative, The Wharton School & The School of Arts and Sciences, University of Pennsylvania

Linnea Gandhi, PhD Candidate in Operations, Information, and Decisions, The Wharton School, University of Pennsylvania

Katherine Milkman, James G. Dinan Endowed Professor of Operations, Information, and Decisions, Co-Director of Behavior Change for Good Initiative, The Wharton School, University of Pennsylvania.

Harsha Thirumurthy, Associate Professor of Medical Ethics and Health Policy, Associate Director of CHIBE, Perelman School of Medicine, University of Pennsylvania

Kevin Volpp, Mark V. Pauly Presidential Distinguished Professor of Medicine, Medical Ethics and Health Policy, & Health Care Management, Director of CHIBE, Perelman School of Medicine & The Wharton School, University of Pennsylvania

BCFG x CHIBE Virtual Briefing on Tactics Tested for Encouraging COVID-19 Vaccination

Agenda:

Welcome (Professors Angela Duckwork and Alison Buttenheim)
Can Text Message Nudges Increase Vaccination Rates? Results from Two Megastudies (Professor Katherine Milkman)
Does Increasing the Odds of Winning a Vaccine Lottery Increase Vaccine Uptake? (Linnea Gandhi)
Are Statewide Incentive Programs for COVID-19 Vaccinations Effective (Professor Harsha Thirumurthy)
Audience Q+A (led by Professor Alison Buttenheim)
Key Takeaways and Closing Remarks (Professor Kevin Volpp)

December 3, 2021

Dolly Chugh, Associate Professor of Management & Organizations, New York University

Robert Cialdini, Regents’ Professor Emeritus of Psychology & Marketing, Arizona State University

Wendy Wood, Provost Professor of Psychology & Business, University of Southern California

Panel Discussion featuring authors of: The Person You Mean to Be, Influence, and Good Habits, Bad Habits

Spring 2021 Virtual Seminar Series

Date

Speaker

Topic

*video links below

January 25, 2021

Judd Kessler, Associate Professor of Business Economics and Public Policy

University of Pennsylvania

The Gender Gap in Self-Promotion

Abstract: In applications, interviews, performance reviews, and many other environments, individuals are explicitly asked or implicitly invited to evaluate their own performance and ability. In a series of experiments, involving over 4,000 participants, we find that women evaluate their performance less favorably than equally performing men. This gender gap in self-evaluations is notably persistent.

February 1, 2021

Ethan Kross, Professor of Psychology and Management/Organizations

University of Michigan

Self-Talk: How You Do It Matters

Abstract: We all have an inner monologue that we engage in from time to time. Yet, people often refer to themselves in strikingly different ways when they engage in this introspective process. Whereas people typically use 1st person singular pronouns to refer to themselves (e.g., Why am I feeling this way?), they at times use their own name and other non-1st-person pronouns as well (e.g., Why is Ethan feeling this way or Why are you feeling this way?). Far from representing a simple quirk of speech or epiphenomenon, I will suggest that these linguistic shifts serve a basic self-control function, enhancing people’s ability to reason wisely and control their thoughts, feelings and behaviors under stress.

February 8, 2021

Gretchen Chapman, Professor of Psychology, Social and Decision Sciences

Carnegie Mellon University

Numerical Cognition and Federal Budgetary Expenditures

Abstract: Understanding federal budgetary expenditures is an inherent part of good citizenship, and yet research on numerical cognition points to limitations in how people process very large numbers. Three experiments compared judgments about US federal budgetary expenditures that were presented in total or per capita terms. Per capita presentation improved accuracy of numeric processing and also altered support for the funded government programs.

February 22, 2021

Betsy Levy Paluck, Professor of Psychology and Public Affairs

Princeton University

Prejudice reduction: Progress and challenges

Abstract: The past decade has seen rapid growth in research that evaluates methods for reducing prejudice. We review 418 experiments reported in 309 manuscripts from 2007 to 2019 to assess which approaches work best and why. Our quantitative assessment uses meta-analysis to estimate average effects. Our qualitative assessment calls attention to landmark studies that are noteworthy for sustained interventions, imaginative measurement, and transparency. However, 76% of all studies evaluate light touch interventions, the long-term impact of which remains unclear. The modal intervention uses mentalizing as a salve for prejudice. Although these studies report optimistic conclusions, we identify troubling indications of publication bias that may exaggerate effects. Furthermore, landmark studies often find limited effects, which suggests the need for further theoretical innovation or synergies with other kinds of psychological or structural interventions. We conclude that much research effort is theoretically and empirically ill-suited to provide actionable, evidence-based recommendations for reducing prejudice.

March 1, 2021

Francesca Gino, Tandon Family Professor of Business Administration

Unit Head, Negotiation, Organizations & Markets

Harvard University

Piqued Curiosity

Abstract: Integrating the differentiated knowledge, skills, and experiences of its members is critical to the effective functioning of an organization. For this reason, a principle focus of organizational scholarship is the interactions spanning people, groups, role sets, and organizational units. Decades of research have substantiated not only the range of important outcomes deriving from, but also the formidable barriers to initiating and maintaining, the interactions that foster integration across formal organizational, informal social, and demographic boundaries. And while our knowledge of organizational (e.g., propinquity) and relational (e.g. homophily) obstacles is extensive, our understanding of psychological barriers faced by individuals when contemplating the prospect of engaging in boundary-spanning networking is less advanced. We argue that a chief impediment to boundary-spanning networking in organizations is the lack of motivation on the part of individuals to cultivate such interactions. We further theorize that curiosity heightens individuals’ motivation to network across boundaries in organizations, due to the drive to learn and the desire to discover novel ideas, solutions, and opportunities. Moreover, we propose that the state of curiosity is piqued by cues in the organizational environment. To test our claims, we designed and conducted a randomized-control-trial field experiment involving over 2,200 middle-managers in a North American financial services organization. To further substantiate the operative causal mechanisms, we designed and conducted two additional field experiments involving over 600 working professionals. The results from the three experiments provide strong support for our claims and illuminate the psychological foundations to boundary-spanning networking in organizations.

March 8, 2021

Susan Athey, The Economics of Technology Professor

Senior Fellow, Stanford Institute for Economic Policy Research

Stanford University

Designing and Analyzing Behavioral Experiments with Machine Learning

Abstract: When studying the impact of behavioral interventions, it is important to recognize that different treatments can work differently for different types of people. Understanding treatment effect heterogeneity can be useful for several reasons. First, understanding for whom a treatment works can provide insight about why it works, as well as lay the foundation for improving future treatment designs. Second, it can help assess the validity of a treatment effect estimate when applied to different populations. Third, treatment effect heterogeneity can be used as an input to an optimal treatment assignment policy. For each of these exercises, it is possible to explore the questions using pre-specified observable characteristics of individuals, but that approach leaves open the question of whether the targeting is as good as possible. Using pre-specified covariates, it is not possible to answer the question, “Does there exist a targeted assignment policy that improves over giving everyone the same treatment?” In the last few years, many new methods based on machine learning have emerged that, unlike off-the-shelf prediction methods, are tailored to the context of an analyst who wishes to discover treatment effect heterogeneity, estimate optimal treatment assignment policies, and test hypotheses. This research provides theoretical guarantees as well as practical methods. In my work I have developed software packages to implement methods, including generalized random forests (GRF) and policyTree, available in R on GitHub. To use machine learning for heterogeneous treatment effects, we need sufficient observations as well as observable characteristics of individuals, and enough of each that we can split datasets (in order to assess how well calibrated the machine learning model is, as well as to enable hypothesis testing; and so that there is something for the machine learning to discover with respect to characteristics.) In applying these tools, one potential avenue of exploration is to revisit previously designed large-scale experiments, where there are individual characteristics observable. However, many published studies were designed for interventions that work well for everyone, and were sized to have just-sufficient power for average effects, and thus it may be rare to find a public data set that is well powered and appropriately designed to find heterogeneity. When there is low signal, machine learning can give spurious results, overstating heterogeneity. In several studies, we demonstrate how to analyze heterogeneous treatment effects, assess calibration and magnitudes, and estimate the benefits of targeted policies. In our analysis, we do not assume that the model is well-calibrated, and we provide interpretations reflecting that covariates may be correlated with one another, so that many different models may lead to similar targeting. The BCFG video seminar focuses on the application to a fintech company seeking to increase donations. In a practitioner’s guide, we include applications to behavioral nudges for financial aid. The practitioner’s guide also incorporates an introduction to adaptive experiments and an application to pilot studies.

March 15, 2021

Colin Camerer, Robert Kirby Professor of Behavioral Finance and Economics

MacArthur Genius Award Winner

California Institute of Technology

New Perspectives on Habit Formation from Machine Learning and Neuroscience

Abstract: We introduce a machine learning approach to characterizing individual trajectories of habit formation in the wild: Predictable context-sensitivity (PCS) identifies a person-specific set of variables that maximizes prediction of whether or not that person will perform a behavior on subsequent occasions. We apply PCS to two large, long-term panel data sets tracking (1) hospital caregivers’ hand-sanitizing and (2) gym members’ gym attendance. We find that while past behavior is nearly-universally predictive of future behavior, different subsets of context variables (e.g., day of the week, peer behavior) are predictive for different people. We also find that the time it takes to form a habit is not universal: we estimate that it takes 3 to 7 months to develop a gym habit, but only 4 to 9 weeks to develop a handwashing habit. We conclude with a brief discussion of how a neural autopilot theory, based on reliability of reward as the causal driver of habit, asks new questions.

March 22, 2021

Dean Karlan, Professor of Economics and Finance

Founder of Innovations for Poverty Action

Northwestern University

Nudge Today, Almost Gone Tomorrow: Evidence from Savings Messaging Experiments with 1.2 Million People in 6 Countries

March 29, 2021

Marissa Sharif, Assistant Professor of Marketing

University of Pennsylvania

Leveraging Flexibility to Increase Goal Persistence

Abstract: Throughout long term goal pursuit, it is inevitable that people will experience small, short-term goal failures along the way. Some days it might be impossible to make it to the gym and some days people might have to splurge on that dessert. Unfortunately, these small short-term failures can derail people in reaching their long-term goals. Missing a few days of going to the gym can lead to months without returning. In this talk, I will discuss two interventions that leverage flexibility in order to help people persist in the face of these small failures. First, I will reveal how framing goals with emergency reserves (e.g., reach your step goal 7 days of the week with 2 emergency skip days) lead people to persist more than goals framed without them, including objectively equivalent goals (e.g., reach your step goal 5 days of the week) and more difficult goals (e.g., reach your step goal 7 days of the week). Second, I will discuss how encouraging people to “make up” for their past goal failures can also lead to greater persistence. I will demonstrate the implications of these strategies for achieving fitness and language learning goals.

April 5, 2021

Muriel Niederle, Professor of Economics

Stanford University

The Role of Competitiveness in Education and Labor Market Outcomes (based on joint work with Thomas Buser and Hessel Oosterbeek)

Abstract: We assess the predictive power of two measures of competitiveness for education and labor market outcomes using a large, representative survey panel. The first is incentivized and is an online adaptation of the laboratory-based Niederle-Vesterlund measure. The second is an unincentivized survey question eliciting general competitiveness on an 11-point scale. Both measures are strong and consistent predictors of income, occupation, completed level of education and field of study. The predictive power of the new unincentivized measure for these outcomes is robust to controlling for other traits, including risk attitudes, confidence and the Big Five personality traits. For most outcomes, the predictive power of competitiveness exceeds that of the other traits. Gender differences in competitiveness can explain 5-10 percent of the observed gender differences in education and labor market outcomes.

April 12, 2021

Duncan Watts, Stevens University Professor

Principal Researcher at Microsoft Research

University of Pennsylvania

The Effects of Task Complexity on Group Synergy

Abstract: Complexity—defined in terms of the number of components to a problem and the nature the inter-dependencies between them—is clearly a relevant feature of all tasks performed by groups. Yet the role that task complexity plays in determining group performance remains poorly understood, in part because no clear language exists to express it in a way that allows for straightforward comparisons across tasks. Here we avoid this analytical difficulty by identifying a class of tasks for which complexity can be varied systematically while keeping all other elements of the task unchanged. We then test the effects of task complexity in a preregistered, two-phase experiment in which 1,200 individuals were evaluated on a series of tasks of varying complexity (phase 1) and then randomly assigned to solve similar tasks either in interacting groups or as independent individuals (phase 2). We find that groups are faster and more efficient than similar sized collections of independent problem solvers for complex tasks but not for simpler ones. Leveraging our highly granular digital data, we define and precisely measure group process losses and synergistic gains, and show that the balance between the two switches signs for intermediate values of task complexity. Finally, we find that groups generate more solutions more rapidly, and explore the solution space more broadly, than independent problem solvers, finding higher quality solutions than all but the best individuals.

April 19, 2021

John Beshears, Terrie F. and Bradley M. Bloom Associate Professor of Business Administration

Harvard University

Consumption Responses to Mortgage Payments: Evidence and Implications

Abstract: We empirically examine how spending on nondurable consumption items responds to the timing of mortgage payments. Contrary to standard economic models and some leading alternative models from the behavioral science literature, consumption declines immediately after a mortgage payment relative to immediately before. The effect is primarily driven by spending on debit cards (instead of credit cards) and is concentrated among households who pay interest on credit card debt. We discuss the implications of these findings for interventions that may help households better plan their spending and ultimately reach better financial outcomes.

April 26, 2021

Modupe Akinola, Sanford C. Bernstein & Co. Associate Professor of Leadership and Ethics

Columbia University

Thriving Under Pressure: The Effects of Stress-Related Wise Interventions on Affect, Sleep, and Exam Performance for College Students from Disadvantaged Backgrounds

Abstract: Nearly all students—and especially students from socially disadvantaged backgrounds—experience stress as they pursue important academic goals. But evidence suggests that adopting a stress-is-enhancing mindset can help improve responses to stress. We find that a stress-mindset intervention delivered to freshmen students from disadvantaged backgrounds many months before finals increased exam-week positive affect and decoupled the negative association between exam-week stress and exam performance, relative to a control group. An alternative stress management intervention also mitigated the negative association between stress and performance for students from disadvantaged backgrounds relative to controls, but had a non-significant effect on positive affect relative to controls and was associated with directionally less sleep than controls and significantly less sleep than stress mindset. This study suggests the potential value of targeting mindsets about stress using a “wise intervention” approach to try to improve well-being and sleep, and to prevent stress-related performance declines for students from disadvantaged backgrounds during stressful times.

May 3, 2021

Alison Wood Brooks, Associate Professor of Negotiation, Organizations, and Markets

Hellman Faculty Fellow

Harvard University

How to Talk Gooder: The Science (and Serendipity) of Conversation

Abstract: Behavioral science tells us much about the decisions we make in our lives, but less about the conversations that surround the decisions we make—who we talk to, what we say, and how we say it (especially based on conversations between real people). But a data-driven revolution on the psychology of conversation is afoot! Why? Conversation is an uncertain, unfolding coordination environment that demands a relentless stream of micro-decisions—rich with possibilities for behavioral research. In this talk, I’ll zoom in on a fundamental micro-decision we make at every turn of every conversation (often without noticing): Should we stay on this topic or switch to another one? Humans instinctively use topics as a “chunking heuristic” to index (organize and remember) conversational content after the fact (“We talked about the weather, his mom, chili recipes, and fantasy football”). I’ll share data that suggests that topics provide a useful heuristic during conversations, too, because the topic level of analysis is more fine-grained than thinking about whole conversations (e.g. as “difficult”), and less cognitively overwhelming than thinking about specific words or phrases to use on the fly (though some words and phrases can be helpful, too). First, I’ll present one barrier that stands in the way of making good topic choices: we struggle to detect what others want to talk about. Second, I’ll offer two strategies that can help us navigate topics more effectively: brainstorming possible topics before conversations begin, and asking follow-up questions once they’re underway. Together, these strategies can help us both prepare and improvise (and reduce anxiety, disfluencies, and egocentrism along the way).