Embedded Files
Since September 2024, I have been a post-doctoral researcher at GATE Lyon Saint-Étienne (GATE Lab team) and at emlyon business school (BRIO research group). I work with Astrid Hopfensitz and Fabio Galeotti on projects examining beliefs and decision-making within couples (ANR-23-FRAL-0013). In parallel, I am conducting research on gender biases in managerial decision-making with Gabriel Bayle and Astrid Hopfensitz, on overconfidence and deterrence in contests with Alice Soldà, and on methodological questions examining how the quality of user interfaces (UI) in economic experiments affects the reliability of behavioral data.
Does Discussion with Affected Others Influence Competitiveness? with Fabio Galeotti and Astrid Hopfensitz
Data collection completed → Data analysis underway
ABSTRACT
A large experimental literature documents the determinants of individuals’, and in particular women’s, competitiveness. Yet most studies examine competitiveness as an individual decision made in isolation, whereas real-life career decisions, such as applying for a promotion or changing jobs, are often discussed with others (e.g. romantic partners, friends or colleagues). These discussions may matter because others, although often less informed than the decision-maker, may be affected by the decision, may hold different preferences, and may provide perspectives that shape how risks and competitive environments are evaluated. Despite this relevance, the role of discussion with affected but imperfectly informed others remains largely understudied.
To fill this gap, we study how discussion with a male partner and responsibility for this partner’s earnings affect women’s competitiveness using a 2×2 experimental design. In a laboratory experiment, female decision-makers first perform a math task and choose how to allocate their compensation between a piece-rate and a tournament scheme (INDIVIDUAL condition). They are then randomly paired with a male participant and make a second competitiveness decision with earnings accruing also to the male partner (RESPONSIBILITY condition). Crucially, between these two decisions, pairs engage in a live audio-video discussion. In the RELEVANT condition, participants are aware of the upcoming competitiveness decision and can discuss it; in a control IRRELEVANT condition, participants also discuss but are unaware of the subsequent decision and therefore cannot discuss it. We additionally elicit confidence, risk preferences and beliefs about the partner’s preferences, allowing us to explore potential mechanisms through which discussion may affect decisions.
Although, on average, neither responsibility nor discussion affects women’s competitiveness, relevant discussion significantly increases the extent to which women adjust their competitiveness when they are responsible for their male partner’s earnings, compared to when they are not. Importantly, this adjustment is not driven by decision noise, but by a stronger influence of the male partner’s believed risk preferences, facilitated by more accurate inference of those preferences following decision-relevant discussion. As a result, being matched with a more risk-tolerant man increases women’s competitiveness in this condition.
Overconfidence and Deterrence in Contests with Si Chen and Alice Soldà
Design validated → Data collection imminent
Who Takes the Risk? Gender Bias and Preference Formation in Organizations with Gabriel Bayle and Astrid Hopfensitz
Design validated → Data collection imminent
ABSTRACT
Persistent gender inequalities in the workplace are shaped not only by differences in pay or promotion, but also by the assignment of workers to different professional roles. In particular, managers often assign workers to tasks that vary in the level of risk they are expected to take. A key concern is that these assignments, based on imperfect information about workers, may reflect stereotypical gendered beliefs rather than actual differences in preferences or abilities. Drawing on existing literature, we hypothesize that managers may perceive women as systematically less willing to take risks than men, even when little or no objective differences exist between genders. Such beliefs matter because they can limit women’s access to high-reward opportunities, which are often associated with riskier tasks and roles. They may also generate self-fulfilling dynamics if workers adjust their behavior to the roles they are assigned.
Our research addresses two central questions. First, from the manager’s perspective, we examine whether managers exhibit gender bias by systematically assigning men to riskier positions, and whether this bias is driven by exaggerated beliefs about gender differences in risk attitudes. Second, from the worker’s perspective, we test whether being assigned to a specific role shapes subsequent risk preferences, and whether this effect is amplified when workers are explicitly told that the assignment was made by a manager who reviewed their CV—potentially creating a self-fulfilling prophecy in which assigned roles reinforce the stereotypes that led to the assignment.
To test these hypotheses, we conduct a three-session experiment. First, we profile workers to construct CVs. Managers then assign these workers to either a position where risk-taking is expected or one where it is not, based on the presented CVs, allowing us to measure gender bias in assignments. Finally, workers perform their assigned roles, and we elicit their risk preferences. Crucially, we manipulate whether workers are informed that their role was assigned by a human manager, allowing us to identify the causal effect of role assignment on preference formation.
This study aims to provide causal evidence on whether managerial role assignment decisions are gender-biased and to explore the mechanisms through which biased beliefs may perpetuate gender disparities in the workplace by shaping individual preferences.
Do Partners Know Best? Willigness to Compete within Couples with Fabio Galeotti and Astrid Hopfensitz
Design validated → Recruiting couples
ABSTRACT
Women of similar ability are less likely than men to choose competitive compensation schemes, a gap with important implications for career advancement (Niederle & Vesterlund, 2007). This project investigates how interpersonal discussions influence individuals’ decisions in this domain. Many real-life career decisions—such as negotiating a raise or applying for a new job—are made after discussing with a romantic partner, yet the influence of such discussions remains largely understudied.
We design an experiment in which participants are assigned to male–female dyads, with one partner (the Decision Maker) choosing between a piece-rate and a tournament compensation scheme for a task, first individually, then after a conversation with the other partner. By comparing real-life couples to stranger pairs, we assess whether intimate partners are better positioned to improve decision quality and reduce the gender gap in willingness to compete.
Our design includes rich measurements of individual preferences and beliefs about the partner’s preferences to uncover potential mechanisms. We explore two main channels: (1) a debiasing effect, where the partner helps shift the Decision Maker’s confidence (e.g., women gain confidence and become more willing to compete); and (2) preference transmission or integration, where individuals align with their partner’s preferences (e.g., risk-averse women agree to take more risk and enter the tournament). These mechanisms may increase WTC for women, decrease it for men, and ultimately reduce the gender gap.
How Couples Share and Process Information: Home Price Perceptions and Expectations with Astrid Hopfensitz
Design validated → Recruiting couples
ABSTRACT
Couples frequently face major household decisions—such as buying a home, planning a move, or making a career change—that depend on interpreting complex, ambiguous, and often fragmented information. In these situations, each partner may have access to different, potentially relevant information, but this information is rarely common knowledge. For instance, one partner may have better insight into housing market trends through peers, while the other tracks household finances. While communication is a natural solution, recent research suggests that information exchange within couples is often imperfect—even when incentives are fully aligned.
Recent evidence highlights important gender asymmetries in how information flows within households. Fehr et al. (2024) show that men are less likely to integrate information provided by their wives, despite mutual incentives. Similarly, Conlon et al. (2021) find that women incorporate their partner’s signals when updating beliefs, whereas men tend to underweight their partner’s input. However, these findings have not been consistently replicated (e.g., Mustafi, 2024), and the underlying mechanisms—whether men discount input or women under-communicate—remain unclear.
This project contributes to this literature by examining how couples communicate and process information about home price dynamics, a domain that is both high-stakes and widely relevant. In a controlled experiment, participants are paired in gender-mixed dyads (either real-life couples or strangers) and receive different pieces of information about past housing price inflation. Drawing on prior evidence that perceived past inflation shapes beliefs about future prices (Armona et al., 2019), we elicit participants’ beliefs about past and future home price changes before and after receiving the information, and again after a recorded discussion between partners. Final individual beliefs allow us to identify how discussion affects belief updating and whether gender patterns emerge in responsiveness to a partner’s input.
With this design, we aim to disentangle gender differences in information pooling—do men disregard input, or do women under-communicate? Does this vary by partner type (spouse vs. stranger)? More broadly, we shed light on how households handle real-world ambiguity and whether communication effectively helps combine dispersed, noisy information.
Poor Design, Poor Data? Examining the Behavioral Impact of User Interface Quality in Economic Experiments
Design in development
ABSTRACT
This project investigates whether the quality of user interfaces (UI) in economic experiments affects the reliability of behavioral data. While experimental economists devote significant attention to designing robust incentives and clear instructions, comparatively little scrutiny has been given to the interfaces through which participants interact with experimental tasks. Drawing on insights from web design and human-computer interaction, we test whether UI flaws—ranging from poor visual design to inconsistent input methods—can degrade data quality by increasing noise, reducing engagement, or inducing bias.
In our experiment, participants complete three widely used experimental tasks—a Bayesian updating task, a real-effort task, and a risk preference elicitation task—under either a well-designed or a poorly designed interface condition. The two versions differ along several dimensions: text presentation (e.g., inconsistent fonts and typos), visual layout (e.g., misaligned elements and clashing colors), input methods (e.g., inconsistent use of keyboard and mouse), and responsiveness (e.g., unnecessary confirmation steps). In addition, this design allows us to examine both involuntary effects (e.g., distraction or cognitive load) and voluntary reactions (e.g., reduced effort or strategic disengagement).
The project aims to inform best practices for experimental design in economics and behavioral sciences, particularly in online settings where UI issues are both more likely and more consequential. By identifying the behavioral impact of seemingly superficial design flaws, we seek to promote greater attention to interface quality as a key determinant of data validity.
Page updated
Google Sites
Report abuse