This episode's title refers to a psychological phenomenon where people overestimate how widespread a behavior is because the most visible members of their social circle are doing it. Liz surveys 76 students to examine whether teens are overestimating how common professional signaling behaviors like having a job, dressing for work, and building a LinkedIn profile actually are. The most memorable moment comes when we turn the lens on the room she's sitting in and we ask whether AP Research itself might be subject to the same forces she spent a year studying.

Scrub to 1:10 to skip the introduction.

This episode's title refers to the themes Geffen uncovered in over 3,000 complaints filed with the Consumer Financial Protection Bureau. Using a machine learning method called Latent Dirichlet Allocation, he analyzes what consumers are actually saying about Buy Now, Pay Later companies and finds that the complaints go beyond hidden fees and confusing payment structures to something newer: the ways technology itself may be used to obscure what consumers are really agreeing to. His closing advice is worth the listen alone.

Scrub to 1:10 to skip the introduction

This episode's title refers to the idea that word choice in news reporting shapes how audiences understand a conflict without ever stating a position. Joe analyzes five news outlets from the five countries contributing the most funding to the Russo-Ukrainian war and finds that each one has its own distinct form of bias, not just in what it says, but in what it chooses to focus on at all.

Scrub to 1:10 skip the introduction.

This episode's title refers to the funding and treatment gap between two chronic diseases that share surprising biological similarities. Chloe uses a thematic analysis of posts from disease-specific Reddit communities to compare how patients with sickle cell disease and cystic fibrosis describe their experiences with healthcare providers. What she finds is not just a difference in outcomes, but a difference in the nature of the barriers each group faces, one rooted in logistics, the other in something far more personal.

Scrub to 1:10 to skip the introduction

This episode's title refers to Disney's ongoing project of remaking its animated classics into live action films, and the question of whether those remakes actually change how gender is portrayed. Sophia has been studying this since ninth grade, and this year she applies a quantitative coding framework to Snow White and Beauty and the Beast, comparing their animated and live action versions side by side. Her most surprising finding is that the biggest shift becomes visible only when you stop looking at the princess alone and start looking at how gender is represented across both the prince and the princess together.

Scrub to 1:10 to skip the introduction

This episode's title refers to the distinction Malia draws between parasocial relationships, which occur between humans at a distance, and pseudosocial attachment, which forms between a person and a non-human entity like an AI. Malia conducts a content analysis of ChatGPT's responses to escalating prompts modeled on teen mental health struggles, examining how well the platform's guardrails actually hold up.

Scrub to 1:10 to skip the introduction.

This episode's title borrows from Kevin's current undergraduate research paper, which recontextualizes the 2008 candlelight protests in South Korea through the lens of neoliberalism and globalization. Kevin was the first student ever interviewed for this podcast, and he returns to talk about where his research has taken him since graduating, what college-level independent research actually looks like, and answers questions from high school students interested in pursuing college-level work. The episode closes with a discussion on AI and how it is currently impacting both sides of the classroom and lecture hall.

This episode features two student projects examining discourse on X. Andrew analyzes how American fans perceive cultural appropriation in K-pop, while Jeffrey looks at how English-speaking audiences respond to Arabian Peninsula countries owning major European football clubs. Together, their findings raise a question neither X nor the research fully answers: once a problem is identified, what comes next?

Scrub to 1:10 to skip the introduction!

This episode's title refers to the way oversexualization shows up across very different cultural products, often carrying the same message to the same audience. Katie analyzes objectification in popular contemporary R&B/hip hop lyrics across gender lines, while Kailey compares the sexualization of women's and girls' Halloween costumes on Spirit Halloween's website. Together, their findings point to how these messages get normalized long before adolescents have the tools to recognize them.

Scrub to 1:10 to skip the introduction.

This episode's title refers to the recurring use of depression as a narrative device in modern horror filmmaking. Rebecca conducts a content analysis of three horror films, The Babadook, The Night House, and Lights Out, and examines how each one portrays depression through emotional, behavioral, and cinematic cues. She finds that horror's tendency to externalize mental illness as something dangerous and uncontrollable may be shaping how audiences understand depression in real life.

Scrub to 1:10 to skip the introduction.