Trustworthy-looking faces are also perceived as more attractive, but are there other meaningful cues that contribute to perceived trustworthiness? Using data-driven models, we identify these cues after removing attractiveness cues. In Experiment 1, we show that both judgments of trustworthiness and attractiveness of faces manipulated by a model of perceived trustworthiness change in the same direction. To control for the effect of attractiveness, we build two new models of perceived trustworthiness: a subtraction model, which forces the perceived attractiveness and trustworthiness to be negatively correlated (Experiment 2), and an orthogonal model, which reduces their correlation (Experiment 3). In both experiments, faces manipulated to appear more trustworthy were indeed perceived to be more trustworthy, but not more attractive. Importantly, in both experiments, these faces were also perceived as more approachable and with more positive expressions, as indicated by both judgments and machine learning algorithms. The current studies show that the visual cues used for trustworthiness and attractiveness judgments can be separated, and that apparent approachability and facial emotion are driving trustworthiness judgments and possibly general valence evaluation.
3 November 2025 in Journal of Clinical and Translational Science
Introduction: Declining participant engagement threatens human subjects research. Participant feedback systems (PFS) may combat this decline by empowering participants to evaluate their research experiences and share that feedback with researchers to identify targets for improvement. PFS signal that participant experiences are prioritized, making the request for feedback itself an intervention. PFS design work remains largely confined to clinical research. This exploratory study investigates the design parameters of extending PFS to nonclinical research. We conducted focus groups with nonclinical stakeholders: Experienced research participants (ERP) and research team members (RTM). Methods: ERP focus groups were organized by affinity (LGBTQIA+, BIPOC, persons with disabilities, neurodivergent, and a general group). RTM focus groups were organized by unit within the University of Michigan. Transcripts were analyzed using inductive thematic analysis. Results: Ten focus groups (ERP: 5, n = 25; RTM: 5, n = 26) identified key PFS design considerations: (1) motivations for feedback, (2) feedback collection, and (3) feedback delivery. ERP and RTM collectively preferred anonymous web-based surveys with six potential topic areas: communication, respect, being valued, receiving value, burden, and safety. Feedback delivery faced two key design tensions: balancing institutional standardization with study-specific insights and aligning leadership’s preference for high-level summaries with frontline staff’s need for detailed, real-time feedback. Conclusion: Expanding PFS to nonclinical research requires balancing centralization and study-specific flexibility. While centralization enhances consistency, the diversity of nonclinical studies necessitates adaptable implementation. A hybrid model is proposed to optimize feasibility. Future research should refine and test this model.
1 January 2025 in Journal of Personality and Social Psychology
Face stereotypes are prevalent, consequential, yet oftentimes inaccurate. How do false first impressions arise and persist despite counter-evidence? Building on the overgeneralization hypothesis, we propose a domain-general cognitive mechanism: insufficient statistical learning, or Insta-learn. This mechanism posits that humans are quick statistical learners but insufficient samplers. Humans extract statistical regularities from very few exemplars in their immediate context and prematurely decide to stop sampling, creating and perpetuating locally accurate-but globally inaccurate-impressions. Six experiments (N = 1,565) tested this hypothesis using novel pairs of computer-generated faces and social behaviors by fixing the population-level statistics of face-behavior associations to zero (i.e., no relationship). The initial sample contained either 11, five, or three examples with either a positive, zero, or negative linear relationship between facial features and social behaviors. The sampling procedure contained a free-sampling condition in which participants were free to decide when to stop viewing more examples and a fixed-sampling condition in which participants were forced to view all stimuli before making decisions. Consistent with the Insta-learn mechanism, participants learned novel face stereotypes quickly, with as few as three examples, and did not sample enough when they were given the freedom to do so. This domain-general cognitive mechanism provides one plausible origin of false face stereotypes, demonstrating negative consequences when people learn too much from too little. (PsycInfo Database Record (c) 2025 APA, all rights reserved).
Trustworthy-looking faces are also perceived as more attractive, but are there other meaningful cues that contribute to perceived trustworthiness? Using data-driven models, we identify these cues after removing attractiveness cues. In Experiment 1, we show that both judgments of trustworthiness and attractiveness of faces manipulated by a model of perceived trustworthiness change in the same direction. To control for the effect of attractiveness, we build two new models of perceived trustworthiness: a subtraction model, which forces the perceived attractiveness and trustworthiness to be negatively correlated (Experiment 2), and an orthogonal model, which reduces their correlation (Experiment 3). In both experiments, faces manipulated to appear more trustworthy were indeed perceived to be more trustworthy, but not more attractive. Importantly, in both experiments, these faces were also perceived as more approachable and with more positive expressions, as indicated by both judgments and machine learning algorithms. The current studies show that the visual cues used for trustworthiness and attractiveness judgments can be separated, and that apparent approachability and facial emotion are driving trustworthiness judgments and possibly general valence evaluation.
Scientists have long used physiological recording to study bodily processes underlying emotions. However, sensors are often affected by hair or skin characteristics that impair the accuracy of their measurements in Black individuals. We draw attention to an underrecognized consequence of such recording biases: differential exclusion of Black participants during Quality Control (QC) stages of the data pipeline, what we call the QC exclusion gap. These routine procedures receive less attention than other stages of the research process; yet, disparate participant exclusions during QC can undermine the internal and external validity of research utilizing physiological recording. Using Electrodermal Activity (EDA) as an illustrative example, we walk through QC analyses of Black (N= 72) and White (N=68) participants’ Skin Conductance Levels (SCL) during a baseline passive listening task. Black participants exhibited significantly lower SCL (M = 4.13μS, SD = 2.49) than White participants (M = 7.95μS, SD = 3.19), p < .001. As a result, Black participants’ EDA data received lower quality grades during QC, leading to differential attrition during triage. We offer immediate recommendations to compensate for the QC exclusion gap, such as integrating quality monitoring feedback into recruitment procedures and reporting exclusion rates by demographic group, and identify broader directions for sensor development and analytical methods. Quality control is often considered a prelude to data analysis; by attending to who is excluded or included, however, we can design a more inclusive science.