The SESAME Colloquium

The SESAME Colloquium offers talks on a variety of subjects related to the learning sciences.

Unless otherwise noted, in-person talks are given on Thursdays from 12:00 to 1:30 pm in room 4101, Berkeley Way West (2121 Berkeley Way, Berkeley, CA 94704). Here is the Zoom link for joining talks that are held remotely or in hybrid format.

We are now starting most of these colloquium sessions close to 12:00 (not "Berkeley Time") with a brief puzzle, icebreaker, or other social activity.

Contact sesame.colloquium@berkeley.edu to be added to our mailing list, if you are interested in speaking in this series, or if you'd like to suggest a speaker.

Videos of many past talks may be found on our Colloquium YouTube channel.

Here is information for speakers.

August 27: TBA

Title TBA

Speaker TBA

Abstract

tba

About the speaker

tba

September 3: Jonathan Pedroza

Examining Student Internal Motivators and Ways For Improving Competency in an Accelerated Statistics Course

Jonathan Pedroza (JP)

Berkeley School of Education, University of California, Berkeley

Abstract

With interest in data science growing, there is an influx of students pursuing data science programs from diverse academic backgrounds. This diversity attracts students with varying levels of mathematics and statistics competencies, which requires instructors to adapt courses to be equitable for all students. While external factors, like teacher and school resources, directly impact student mathematics achievement, my focus will be on what instructors can do once students arrive in their classrooms. This focus includes investigating changes in students’ internal motivators and discussing potential changes that instructors can implement for courses to reduce the mathematics and statistics competency gap.

In this talk, I will talk about findings from surveys shared with students on self-efficacy and motivations at three points in an accelerated summer semester, design changes to make statistics courses more equitable, findings on some of the equitable components of the course, and a tool for instructors interested in implementing some of the components covered in the course. Lastly, I’ll discuss some of the lessons learned about incorporating surveys into the classroom and implementing a hybrid grading approach.

About the speaker

Jonathan A. Pedroza (JP) is a postdoctoral scholar in data science education in the School of Education at UC Berkeley. JP’s research interests include developing methods for early identification of struggling students in college courses, and utilizing alternative course design and grading practices.

September 10: Devin T. White

BRACEing for Liberation: The Black Resistance and Care in Education Model

Devin T. White

College of Education and Human Development, University of Maine

Abstract

In this talk, I will discuss the development of the Black Resistance and Care in Education (BRACE) model, a recursive theoretical framework that theorizes care labor as a strategy of resistance. The BRACE model is explicitly adapted from Margaret Beale Spencer’s Phenomenological Variant of Ecological Systems Theory (PVEST), which centers culturally grounded, identity-driven responses to risk and resilience among Black youth and families (Spencer, 1995, 2006). Drawing on this lineage, BRACE reorients analysis toward the recursive and co-constitutive nature of care and resistance in the lives of Black individuals in education.

Built from the findings of three qualitative studies, BRACE reveals that care and resistance function not as opposites but as interwoven strategies in response to institutionalized anti-Blackness. The BRACE model thus serves as both a reflection of participants lived realities and a theoretical offering for guiding future research, institutional transformation, and culturally responsive praxis.

About the speaker

Devin T. White was born in Oakland, California, and raised on the same block as the Melrose Branch of the Oakland Public Library. Currently, Dr. White is a Postdoctoral Scholar at the Center for Empowered Learning and Development with Technology (CELDTECH) in the Rossier School of Education at the University of Southern California.

At CELDTECH, Dr. White supports the research of Dr. Brendesha Tynes in the development and piloting of CRITmetic, a digital media literacy and mental health platform designed with and for adolescents to cope with online racial discrimination and misinformation. Ultimately, Dr. White is committed to amplifying the voices of Black STEMmers K-20.

Septmber 17: Shaun Liverpool

Technology and Youth Mental Health: Bringing the Evidence Together

Shaun Liverpool

Health Research Institute, Edge Hill University, UK

Abstract

Digital technologies are increasingly positioned as part of the solution to the growing mental health needs of children and young people. From apps and online therapy to virtual reality, games, and emerging artificial intelligence tools, the possibilities are expanding rapidly. But what does the evidence actually tell us about whether these technologies improve young people’s mental health, who benefits from them, and under what circumstances?

In this talk, I will bring together findings from research on digital mental health interventions for children and young people, drawing particularly on evidence syntheses alongside examples of different technologies and approaches. I will explore what we currently know about their effectiveness, acceptability, engagement, and potential to increase access to mental health support. I will also consider where the evidence is less certain, including questions around longer-term outcomes, safety, cost-effectiveness, implementation, and whether digital approaches are reaching young people who experience the greatest barriers to traditional mental health services.

The talk will argue that the key question is no longer simply whether technology can support youth mental health, but how we ensure that it does so in ways that are effective, safe, equitable, and responsive to young people’s lives. I will conclude by highlighting priorities for the next generation of research, including greater involvement of young people in designing technologies, more attention to marginalised and underserved populations, and a shift from evaluating individual digital tools towards understanding how technology can be meaningfully integrated within wider systems of mental health support.

About the speaker

Dr Liverpool was born in Trinidad and Tobago. He started his Psychology studies at the University of the West Indies in Barbados before moving to the UK for postgraduate studies. Currently, Dr Liverpool is an Associate Professor at the Health Research Institute at Edge Hill University and an Atlantic Fellow for Health Equity at George Washington University. Dr Liverpool holds leadership roles within the Research Centre for Arts and Wellbeing and teaches on the Child and Adolescent Mental Health programmes.

September 24: Fall Welcome Event

Fall Welcome Event

October 1: STEM Ed Panel #1

A's For All + Computer-Based Testing = Mastery Learning

STEM Ed Panel 1 (Dan Garcia and Armando Fox)

Electrical Engineering & Computer Sciences, University of California, Berkeley

Abstract

Research on Mastery Learning (dating to Bloom's work in 1968) shows that by giving students multiple opportunities to practice and demonstrate mastery, including “second-chance” exams and flexible assessment deadlines, we can narrow equity gaps and raise the performance of all students. In principle, this means anyone willing to work for an A should be able to achieve it (“A’s For All”), without lowering standards. But actually implementing this at scale requires the judicious application of technology and automation to manage flexible deadlines and multiple-chance assessments at scale, and to evaluate which policies actually achieve the goals of mastery learning.

We’ll discuss key milestones and lessons learned from putting Mastery Learning into practice at UC Berkeley, including:

Bootstrapping a Computer-Based Testing Facility from one course with 230 students in 2023 to over 16 courses serving over 5,500 students in 2026, both to deliver exams economically at scale (allowing second-chance retakes) and to improve the academic integrity of assessments in the AI era [1]
Numerous policy experiments, both successful and less so, around eliminating curved grading, flexible deadlines, clobber exams, and second-chance exams in large computer science courses [2]

References

Armando Fox. 2026. Bootstrapping a Computer-Based Testing Facility Using Shared Lab Space. In Proceedings of the 2nd ACM Virtual Global Computing Education Conference V.1 (SIGCSE Virtual 2026), November 12–15, 2026.
Dan Garcia, Armando Fox, Solomon Russell, Edwin Ambrosio, Neal Terrell, Mariana Silva, Matthew West, Craig Zilles, and Fuzail Shakir. 2023. A's for All (As Time and Interest Allow). In Proceedings of the 54th ACM Technical Symposium on Computer Science Education V. 1 (SIGCSE 2023). Association for Computing Machinery, New York, NY, USA, 1042–1048. https://doi-org.libproxy.berkeley.edu/10.1145/3545945.3569847

About the speakers

tba

October 8: Brendan Henrique

From Ordinality to Linearity: Rethinking How We Estimate Numbers

Clarissa Thompson

University of Notre Dame, Institute for Educational Initiatives (IEI) and Center for Educational Research and Action (Era)

Abstract

How do people learn to estimate numerical magnitudes, and what can their performance reveal about the cognitive processes underlying numerical estimation? Long-standing debates about the number-line estimation task have focused on whether children’s estimates reflect a shift from logarithmic to linear numerical representations or instead arise from biased magnitude representations and response strategies.

To move beyond this debate, we formed an adversarial collaboration and developed the nlFit model (Thompson & Cohen, 2025; American Psychologist), a computational cognitive model that does not assume an underlying numerical representation but instead captures a progression from knowledge of ordinality to knowledge of linearity. The model has successfully captured children’s and adults’ performance across ages, tasks, and experimental manipulations, and new data provide an opportunity to ask what the model’s parameters reveal about individual differences in numerical estimation.

In this talk, I will present findings from new datasets (Cohen et al., 2026, invited revision; Journal of Educational Psychology) examining whether nlFit parameters vary with gender identity and self-reported ratings of math anxiety across bounded and universal number-line tasks. I will discuss how this preliminary research might inform future approaches to improving students’ estimation skills and decreasing their self-reported math anxiety.

About the speaker

Dr. Clarissa A. Thompson completed her Ph.D. in Developmental and Cognitive Psychology in 2008 at The Ohio State University. After spending two years as a postdoctoral research associate at Carnegie Mellon University, she became an Assistant Professor at the University of Oklahoma in 2010. While at OU, Dr. Thompson coordinated the Elements of Psychology Program and taught graduate students how to teach for the first time. She joined the Kent State Department of Psychological Sciences in 2014, where she served as Undergraduate Curriculum Coordinator, Chair of the KSU IRB, and Co-Director of the Science of Learning and Education Center until 2026. Dr. Thompson is now a Research Professor in the Institute for Educational Initiatives and Director of Research Programs in the Center for Educational Research and Action (Era).

Her research focuses on how students understand number magnitudes and best practices at improving math competencies. She has been recognized by the American Psychological Association for Distinguished Contributions of Applications of psychology to education and training, and has earned several grants from the National Science Foundation and the U.S. Department of Education.

If you have questions about the Colloquium series, please contact sesame.colloquium@berkeley.edu.