Our Research

We conduct research on core cognitive processes involved in learning, memory, and decision making, and explore their role in real-world problems related to education, health, and human-computer interaction. How do people adapt to their environment to be more successful? How do we represent and reason about uncertainty? How do our existing beliefs and attitudes shape the way we explore and interact with others? We examine these fundamental questions about human cognition from several angles as seen in the project descriptions below. Ultimately, our research has the common goal of enhancing human learning and wellbeing through a better understanding of how people interact with and acquire new knowledge from the world around them.

In support of the core scientific questions above, the lab engages in activities related to:

• Behavioral experiments in the lab and online
• Design and programming of experimental tasks and games
• Computational modeling of cognition and behavior
• Open and reproducible science
• Data science-y skills involved in working with behavioral data, including statistical programming
• Pedagogical research on research methods and statistical thinking

The lab typically includes students from graduate programs in Cognitive Science, Psychology, Health Psychology, and Computer Science. Undergraduate research assistants are welcome from any area as long as they have an interest in cognitive science or a closely related field and are motivated to learn about the work we’re doing in the lab. If you are a prospective student, read more about how to get involved.

See below for summaries of some current research directions, or the list of publications.

Constructing abstract conceptual knowledge through exploration

A hallmark of human intelligence is the ability to construct abstract knowledge out of related experiences. For instance, by simply exploring a physical space, the mind naturally forms a “cognitive map” that serves as a model of the environment. This internal model enables flexible reasoning (e.g., route planning) and the discovery of new relationships that haven’t been directly experienced (e.g., unexplored shortcuts).

Research in cognitive psychology has shown that people also create internal “maps” to organize other kinds of experiences. For instance, when starting at a new school or company, we acquire new knowledge about the social and cultural environment by observing how people interact: Which individuals seem more popular or powerful? What sorts of behaviors are rewarded vs. punished? And perhaps most salient: Where do I fit in the social hierarchy? The beliefs that we form about the social environment exert a powerful influence on how we behave and interact with others.

An ongoing focus of the lab is to understand how people acquire this kind of abstract relational knowledge through experience. Work thus far has focused on basic cognitive mechanisms involved in combining memories of related experiences into an integrated “map” of novel social hierarchies, with our work showing that people adaptively explore novel relationships for the purpose of learning an underlying social hierarchy. An ongoing focus of the lab is to understand how these processes allow people to rapidly adapt to novel social environments.

Related works:

• Markant, D. (2022). Modeling the effect of chained study in transitive inference. Proceedings of the 44th Annual Conference of the Cognitive Science Society. Austin, TX: Cognitive Science Society.
• Markant, D. (2021). Chained study and the discovery of relational structure. Memory and Cognition. doi: 10.3758/s13421-021-01201-1 [data repository]
• Markant, D. (2020). Active transitive inference: When learner control facilitates integrative encoding. Cognition 200. doi: 10.1016/j.cognition.2020.104188

Learning about the self: Cognitive and affective processes in self-verification

Research in psychology has shown that people often exhibit a range of confirmation biases, such that they seek out information or interpret evidence in a way that reinforces their existing beliefs. One manifestation of this is known as self-verification, in which people (knowingly or not) favor interactions or feedback that that are expected to confirm existing self-views. For instance, when given the opportunity to hear positive or negative evaluations, people with low self-esteem have been shown to prefer negative feedback, even though it reinforces existing, negative beliefs about themselves. Self-verification can create a self-fulfilling feedback loop in which people with negative self-views pay more attention (or put more trust in) negative feedback, which strengthens those views further.

Why do people seek out self-verifying feedback, especially when it means hearing something negative about oneself? In ongoing projects, we are examining the role of emotion regulation processes in decisions about whether to seek out positive or negative evaluations from others, including artificial agents like AI coaches. The eventual goal of this line of research is to identify strategies for providing feedback that build trust and help people to be more receptive to alternative perspectives about themselves.

Related works:

• Glass, S., Levens, S., and Markant, D. (February, 2023). “The relationship between emotion regulation and self-verification.” Poster presented at the Annual Meeting of the Society for Affective Science: Positive Emotion Pre-conference Workshop. Long Beach, CA, USA.

Persuasive data visualization

Data visualizations are increasingly central to public discourse about issues of societal concern, including public health, social inequity, climate change, and political polarization. An effective data visualization can seem to “speak for itself”: To show evidence so compelling that it seems any viewer should take away the same message. Yet, just like many other kinds of evidence, data visualizations are interpreted through the lens of a viewer’s existing attitudes and beliefs. Data that seems self-evident to one viewer might have little impact on someone with different experiences or worldview.

In a collaboration with colleagues in Computer Science, this line of research examines the cognitive mechanisms of persuasive data visualization. How do people evaluate visual representations of statistical evidence, and do they change their minds when that evidence conflicts with their existing beliefs? In a recent study, we found that people with anti-vaccination attitudes made smaller adjustments to their beliefs about the effectiveness of vaccines when presented data in scatterplots with uncertainty representations. Ongoing work in this area aims to use theories of belief and attitude change to understand how data visualizations can be used to persuade among audiences of diverse viewpoints and motivations.

Related works:

• Markant, D., Rogha, M., Karduni, A., Wesslen, R., and Dou, W. (2023). When do data visualizations persuade? The impact of prior attitudes on learning from visualizations of correlations. Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (CHI 2023).
• Karduni, A., Markant, D., Wesslen, R., & Dou, W. (2021). A Bayesian cognition approach for belief updating of correlation judgment through uncertainty visualizations. IEEE Transactions on Visualization and Computer Graphics (TVCG).