An official website of the United States government
Mental health stigma prevents many individuals from receiving the appropriate care, and social psychology studies have shown that mental health tends to be overlooked in men. In this work, we investigate gendered mental health stigma in masked language models. In doing so, we operationalize mental health stigma by developing a framework grounded in psychology research: we use clinical psychology literature to curate prompts, then evaluate the models’ propensity to generate gendered words. We find that masked language models capture societal stigma about gender in mental health: models are consistently more likely to predict female subjects than male in sentences about having a mental health condition (32% vs. 19%), and this disparity is exacerbated for sentences that indicate treatment-seeking behavior. Furthermore, we find that different models capture dimensions of stigma differently for men and women, associating stereotypes like anger, blame, and pity more with women with mental health conditions than with men. In showing the complex nuances of models’ gendered mental health stigma, we demonstrate that context and overlapping dimensions of identity are important considerations when assessing computational models’ social biases.
more »
« less
Spatializing stigma-power: Mental health impacts of spatial stigma in a legally-excluded settlement in Mumbai, India
In disadvantaged neighborhoods such as informal settlements (or “slums” in the Indian context), infrastructural deficits and social conditions have been associated with residents’ poor mental health. Within social determinants of health framework, spatial stigma, or negative portrayal and stereotyping of particular neighborhoods, has been identified as a contributor to health deficits, but remains under-examined in public health research and may adversely impact the mental health of slum residents through pathways including disinvestment in infrastructure, internalization, weakened community relations, and discrimination. Based on analyses of individual interviews (n = 40) and focus groups (n = 6) in Kaula Bandar (KB), an informal settlement in Mumbai with a previously described high rate of probable common mental disorders (CMD), this study investigates the association between spatial stigma and mental health. The findings suggest that KB’s high rate of CMDs stems, in part, from residents’ internalization of spatial stigma, which negatively impacts their self-perceptions and community relations. Employing the concept of stigma-power, this study also reveals that spatial stigma in KB is produced through willful government neglect and disinvestment, including the denial of basic services (e.g., water and sanitation infrastructure, solid waste removal). These findings expand the scope of stigma-power from an individual-level to a community-level process by revealing its enactment through the actions (and inactions) of bureaucratic agencies. This study provides empirical evidence for the mental health impacts of spatial stigma and contributes to understanding a key symbolic pathway by which living in a disadvantaged neighborhood may adversely affect health.
more »
« less
- Award ID(s):
- 1918175
- PAR ID:
- 10494570
- Editor(s):
- Srinivas, Prashanth Nuggehalli
- Publisher / Repository:
- PLOS Global Public Health
- Date Published:
- Journal Name:
- PLOS Global Public Health
- Volume:
- 3
- Issue:
- 7
- ISSN:
- 2767-3375
- Page Range / eLocation ID:
- e0001026
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Mental health stigma prevents many individuals from receiving the appropriate care, and social psychology studies have shown that mental health tends to be overlooked in men. In this work, we investigate gendered mental health stigma in masked language models. In doing so, we operationalize mental health stigma by developing a framework grounded in psychology research: we use clinical psychology literature to curate prompts, then evaluate the models’ propensity to generate gendered words. We find that masked language models capture societal stigma about gender in mental health: models are consistently more likely to predict female subjects than male in sentences about having a mental health condition (32% vs. 19%), and this disparity is exacerbated for sentences that indicate treatment-seeking behavior. Furthermore, we find that different models capture dimensions of stigma differently for men and women, associating stereotypes like anger, blame, and pity more with women with mental health conditions than with men. In showing the complex nuances of models’ gendered mental health stigma, we demonstrate that context and overlapping dimensions of identity are important considerations when assessing computational models’ social biases.more » « less
-
Research has demonstrated that natural disasters, like flooding that are increasing with climate change, can have profound mental health effects. Moreover, these outcomes are not experienced evenly across the population with disadvantaged populations like racial/ethnic minorities and lower socio-economic status individuals being more likely to report psychological diagnoses and symptoms related to floods. However, the mechanisms that could account for the link between social vulnerability and worry about the threat of flooding remain poorly understood. In this analysis, we use a 2022 survey of Houston-area residents to examine how perceived flood risk and subjective flood preparedness relate to racial/ethnic differences in worry about the threat of flooding. We find that both individual-level and area-level race/ethnicity are significantly related to greater worry about the threat of flooding. Further, this is partially mediated by perceived flood risk, but not subjective flood preparedness. This suggests that policies and infrastructure priorities that reduce risk rather than prepare households for flooding would accomplish more in closing the gap in social disparities in mental health outcomes from flooding.more » « less
-
Mental health stigma manifests differently for different genders, often being more associated with women and overlooked with men. Prior work in NLP has shown that gendered mental health stigmas are captured in large language models (LLMs). However, in the last year, LLMs have changed drastically: newer, generative models not only require different methods for measuring bias, but they also have become widely popular in society, interacting with millions of users and increasing the stakes of perpetuating gendered mental health stereotypes. In this paper, we examine gendered mental health stigma in GPT3.5-Turbo, the model that powers OpenAI’s popular ChatGPT. Building off of prior work, we conduct both quantitative and qualitative analyses to measure GPT3.5-Turbo’s bias between binary genders, as well as to explore its behavior around non-binary genders, in conversations about mental health. We find that, though GPT3.5-Turbo refrains from explicitly assuming gender, it still contains implicit gender biases when asked to complete sentences about mental health, consistently preferring female names over male names. Additionally, though GPT3.5-Turbo shows awareness of the nuances of non-binary people’s experiences, it often over-fixates on non-binary gender identities in free-response prompts. Our preliminary results demonstrate that while modern generative LLMs contain safeguards against blatant gender biases and have progressed in their inclusiveness of non-binary identities, they still implicitly encode gendered mental health stigma, and thus risk perpetuating harmful stereotypes in mental health contexts.more » « less
-
Abstract This study explores the outcomes and impacts of sanitary sewer overflows (SSOs) and basement backups in underserved communities in Baltimore, Maryland. The larger effort is an environmental and community-driven mixed-methods project, however, the research in this manuscript focuses on the household survey portion with residents who have experienced SSOs or sewage backups. Based on the snowball sampling method applied, the resulting residents engaged are predominantly African-American individuals, females, homeowners, and residents between the ages of 50 and 69. Strikingly, 70% of respondents reported that their frequency of SSOs is between moderate to frequent. The findings reveal that SSOs are a pervasive issue affecting residents’ physical and mental health and overall quality of life. Despite residents’ perceptions that their household infrastructure is in good condition, the recurring nature of SSOs highlights systemic problems within the city’s aging sewer systems, urging a deeper understanding of the social and structural vulnerabilities involved. This research calls attention to the importance of comprehensive interventions, including effective risk communication strategies and substantial investment in infrastructure rehabilitation, to mitigate the risks posed by SSOs and promote long-term resilience in urban environments. Additionally, it emphasizes the importance of community-driven research in addressing engineering, urban planning, and public health challenges with particular support for the most affected populations.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1371/journal.pgph.0001026
