Tracking Real-Time Micro-Aggressions in STEM Learning Environments

This project aims to understand microaggressions in social media and lab spaces against women and/or students of color in STEM fields. It will integrate knowledge from social, behavioral, and computational sciences for acquiring new insights into the way microaggressions are expressed in real-life academic STEM environments and the way parties involved react to such episodes in-the-moment.

Minoritized students often experience negative exchanges in everyday social interactions that communicate denigration, discrimination, and exclusion [Keller et al., 2010], also referred to as microaggressions. There is strong evidence that microaggressions can create an atmosphere in which minoritized students feel they need to prove they belong [Sue et al., 2007], affecting their ability to concentrate, degrading academic motivation, and compromising success, with major psychological implications [Congleton, 2013; Ngalo-Morrison, 2017; Reynolds et al., 2010]. While existing studies on microaggressions focus on retrospectively documenting individuals’ experiences through self-assessment reports, interviews, and focus group discussions [Harwood et al., 2015; Rhoton, 2011], there has been no tracking of the naturalistic expression of microaggressions in real time, comprising a significant impediment for understanding and ultimately addressing this phenomenon.

Ambulatory assessment performed through wearable and sensing devices provides the unique opportunity to study students’ experiences of microaggressions in the academic environment and gather ecologically valid data related to the real-time expression and perception of such micro-processes by all parties involved. We will collect and analyze real-life longitudinal data from small-group student academic interactions, and develop novel natural language processing and machine learning for automatically detecting microaggressions, and quantitatively assessing their effect on individual and team outcomes. This pilot project will focus on racial/ethnic minorities, specifically African American and Hispanic/Latino(a) students, providing the foundation for future studies, which will examine women, LGBTQ+ students, and individuals with disabilities. Texas A&M University, a “historically White” military school, has observed a major demographic shift over the last decade (e.g., only 2% away from Hispanic Serving Institution designation [TAMU Accountability, 2020]), rendering this research strategically timely and placing our team in a unique position to study ecologically valid interactions from the academic body.

The project will collect a “first-of-its-kind” rich multimodal dataset of speech, physiological signals, and self-reports that will capture real-life experiences of microaggressions of African Americans and Hispanic/Latino(a) students in STEM via ambulatory assessment methods. Following that, it will design knowledge-driven linguistic and acoustic markers of microaggressions, that will allow researchers to automatically detect these micro-behaviors in longitudinal data. Using the collected data, we will study common themes of microaggressions and identify common victims, perpetrators, allies, and bystanders. We will investigate how involved parties react to microaggressions, and explore victims’ emotional reactions, quantified through physiological (e.g., heart rate) and conversational (e.g., speech intonation) measures. Finally, we will explore the effect of microaggressions on individual and team outcomes, such as mood, stress, and team function, as well as buffering factors that can protect against adverse effects.

The research team will pay utmost attention to ensure participants’ privacy and confidentiality via obtaining appropriate permissions from the institutional review board and continuously monitoring the data to identify any adverse events that could potentially emotionally harm participants.

Publication:
Paromita, P., Khader, A., Begerowski, S., Bell, S. T., & Chaspari, T. (2023). Linguistic and Vocal Markers of Microbehaviors Between Team Members During Analog Space Exploration Missions. IEEE Pervasive Computing.