Representing counter-stereotypical scientists in undergraduate biology curricular materials is a powerful way to support student persistence along STEM pathways. The Costello lab uses experimental approaches to identify the specific features of curricular materials showcasing scientists that are responsible for creating change in student attitudes towards science and scientists. 

In collaboration with the DataVersify team, Costello found that including information about scientists' life experiences improves student engagement with quantitative biology activities and that this humanizing information was especially impactful for students who shared excluded identities with the featured scientists. Currently, the DataVersify team is extending this work to better understand how scientist role model stories impacts student perceptions of the attainability of a career in science. This work was funded by NSF's Improving Undergraduate STEM Education program.

The lab is currently engaged in several other collaborative projects to understand the impacts of representing scientists. In collaboration with Christine Lattin at Louisiana State University, the lab is exploring the impact of profiling scientists' hobbies and interests on burnout in graduate students. And in collaboration with Lora Park in UB's Department of Psychology, the lab is interested in understanding what information STEM instructors disclose to students, student perceptions of instructor disclosures, and impacts of those disclosures on student attitudes towards STEM.

Read more about the lab's work on the representation of scientists in Proceedings of the Royal Society – B and CBE – Life Sciences Education. 

Including photos of scientists in course materials is not sufficient to increase student engagement with quantitative biology activities. Instead, by providing information about scientists' life experiences, students find the activities more interesting, more relevant to their future careers, and put more effort into the activities. These results were strongest for students who shared excluded identities with the featured scientists. From: Costello et al. (2025). More than a token photo: Humanizing scientists enhances student engagement. Proceedings of the Royal Society: B.

Demographic identifiers influence the likelihood that a student persists in a STEM major and graduates with a STEM degree. Identifying explanatory mechanisms underlying this pattern is rarely done and is vital for understanding the institutional and classroom interventions most likely to support students along STEM pathways. 

The lab employs structural equation modeling to identify underlying factors responsible for demographic disparities in STEM graduation rates.  Analyzing institutional data from Auburn University in collaboration with Eric Burkholder and Cissy Ballen, Costello found that gaps in STEM graduation rates were mediated by factors that differ among marginalized groups. Women were less likely to receive undergraduate STEM degrees because they were less likely to pursue undergraduate STEM degrees. However, persons excluded because of their ethnicity or race (PEERs) and first-generation college students were just as likely to pursue STEM majors but did not attain STEM degrees because of the performance challenges of early undergraduate courses.

Currently, the lab is extending this work of identifying mechanisms that underlie demographic disparities in STEM graduate rates to UB. Specifically, the lab is exploring how required courses for degree programs impact persistence in STEM fields. Students interested in STEM  are required to take many STEM courses across STEM fields, often simultaneously and in the first year at university. Our work investigating the impacts of required courses on persistence in STEM will help inform the ways that departments structure their degree programs.

Read more about the lab's work on persistence in STEM fields in International Journal of STEM Education.

Alluvial plots indicating students' initial majors, and their pathways into and out of STEM fields over 6 years of college. From: Costello et al. (2023). Pathways of opportunity in STEM: Comparative investigation of degree attainment across different demographic groups at a large research institution. International Journal of STEM Education.

Pursuing a career in STEM is difficult, and students must navigate and overcome these challenges to persist in STEM. The lab is currently launching several projects aimed at characterizing the obstacles that scientists face. Specifically, we want to understand both student perceptions of scientific challenges and how scientists navigate through the challenges they face. Doing so will point towards classroom interventions that can support students along STEM career trajectories.

One classroom intervention the lab plans to develop and evaluate will feature scientist stories of unexpected results. Scientists often obtain results that are not consistent with their predictions (i.e., results that are unexpected). Resilience to unexpected outcomes is an important determinant of success in science, yet students rarely see adaptive responses to unexpected results modeled by instructors or featured scientists. Our work will investigate the impact of featuring unexpected results on student mindsets towards research challenges and their success in STEM. This work will be led by postdoctoral researcher Carolyn Graham.

Most students think they will face obstacles in science and cite obstacles due to the challenging scientific process, their own personal traits, and societal discrimination. From: Manuscript in preparation led by Prathibha Chandran.

Traditional curricula present biology as an objective field and disregard the impact of human biases on the creation, application, and representation of our biological knowledge base. The Costello lab explores how explicitly acknowledging these biases, stereotypes, and assumptions can increase student awareness of the impact science has on social issues. 

Currently, the lab is engaged in multiple collaborative projects aimed at supporting biology curricula that extend beyond traditional content. In collaboration with the Resources for Inclusive Evolution Education Group (RIE2), the lab is evaluating evolution education resources aimed at addressing how exclusionary systems influenced foundational work in the field of evolution. And, in collaboration with the Theobald lab at the University of Washington, the lab is exploring how and whether assessments test students on the societal implications of scientific knowledge. 

Read more about the lab's work on how human biases shape scientific knowledge in Research in Science Education.

Example activities that explicitly acknowledge how biases influence science. From: Costello et al. (2024). Re-envisioning biology curricula to include ideological awareness. Research in Science Education.

Multilevel selection theory is a foundational evolutionary framework that expands the concept of natural selection beyond the traditional focus on individual organisms to also include groups of organisms. Although social behaviors inherently involve interactions among individuals, surprisingly little research investigates both individual- and group-level (multilevel) selection acting on social behaviors. Costello's doctoral research explored the evolution of social networks (a comprehensive description of social behaviors) in experimental populations of forked fungus beetles (Bolitotherus cornutus) at Mountain Lake Biological Station. She found that patterns of multilevel selection on social networks vary across habitats and between sexes.

Although multilevel selection now has strong empirical support, it has a controversial history that has impeded its wide adoption in evolutionary biology. Currently, the lab, in collaboration with ProSocial World's Multilevel Selection Initiative and members of the Brodie lab, is exploring how multilevel selection is presented in undergraduate evolutionary biology textbooks. Given multilevel selection's value for understanding evolutionary change, portraying multilevel selection with modern evidence in evolutionary biology courses is critical, and we are working to develop recommendations on how to teach multilevel selection to undergraduate biology students.

Read more about Costello's work on multilevel selection and forked fungus beetles in Evolution, Journal of Animal Ecology, Journal of Heredity, Biology Letters, The American Naturalist, and Philosophical Transactions B.

When resources are patchily distributed, socially interactive male beetles mate more, but female beetles in more interactive groups lay fewer eggs. From: Costello et al. (2023). Multilevel selection on social network traits differs between sexes in experimental populations of forked fungus beetles. Evolution.