Learning Sciences and Technologies

Faculty

Learning Sciences & Technologies Core Faculty

N. T. Heffernan, The William Smith Dean's Professor of Computer Science and Director of Learning Sciences and Technologies; Ph.D., Carnegie Mellon University; Intelligent tutoring agents, artificial intelligence, cognitive modeling, machine learning
J. E. Beck, Associate Professor; Ph.D., University of Massachusetts, Amherst; educational data mining, student modeling, Bayesian Networks, student individual differences
L. Harrison, Associate Professor; Ph.D., University of North Carolina at Charlotte; Information visualization, visual analytics, human-computer interaction.
E. Ottmar, Assistant Professor; Ph.D., University of Virginia; mathematics teaching and learning; mathematics development and cognition; interventions in schools; instructional quality; social and emotional learning; motivation and engagement; perceptual learning; teacher/child interactions; observational measurement development
A. C. Sales, Assistant Professor; Ph.D., University of Michigan, Ann Arbor; analysis of educational data, log data from intelligent tutors, causal inference, causal mechanisms, randomized experiments, observational studies.
S. T. Shaw, Assistant Professor; Ph.D., University of California, Los Angeles: teaching and learning in mathematics, creativity, statistics education, math anxiety.
J. R. Whitehill, Assistant Professor; Ph.D., University of California, San Diego; Machine learning, crowdsourcing, automated teaching, human behavior recognition.

Learning Sciences & Technologies Associated Faculty

D. C. Brown, Professor; Ph.D., Ohio State University; Knowledge-based design systems, artificial intelligence
J. K. Doyle, Associate Professor; Ph.D., University of Colorado/Boulder; judgement and decision making, mental models of dynamic systems, evaluation of interventions
S. Djamasbi, Professor; Ph.D., University of Hawaii at Manoa; Usability, decision science.
G. T. Heineman, Associate Professor; Ph.D., Columbia University; Component-based software engineering, formal approaches to compositional design
A. C. Heinricher, Professor; Ph.D., Carnegie Mellon University; applied probability, stochastic processes and optimal control theory
C. Ruiz, Professor; Ph.D., University of Maryland; Data mining, knowledge discovery in databases, machine learning
J. L. Skorinko, Professor; Ph.D., University of Virginia; social environmental cues, stigmas and stereotyping, perceptions of others
G. Smith, Associate Professor and Director of Interactive Media and Game Development; Ph.D., UC Santa Cruz, 2012. Computational creativity, games and social justice, tangible computing, computer science education, computational craft, procedural generation.
G. B. Somasse, Associate Teaching Professor; Ph.D., Clark University; Development economics, applied econometrics, policy evaluation, public policy.
S. Stanlick, Assistant Professor and Director of the Great Problems Seminar; Ph.D., Lehigh University; learning sciences and technology, public interest technology, global citizenship, digital sociology, ethics, transformative learning
J. Zou, Associate Professor; Ph.D., University of Connecticut; Financial time series (especially high frequency financial data), spatial statistics, biosurveillance, high dimensional statistical inference, Bayesian statistics.

Program of Study

The Learning Sciences and Technologies (LS&T) program offers graduate studies toward the M.S. and Ph.D. degrees. Our state-of-the-art facilities, faculty and strong relationships with K-12 schools provide students with the resources to perform innovative scientific research at the highest level. The diverse learning environment that characterizes our program promotes easy exchange of ideas, access to all the necessary resources, and encourages creative solutions to pressing scientific questions. The LS&T program is based on three affiliated areas – Computer Science, Cognitive and Educational Psychology, and Statistics – and provides opportunities for advanced course work and research for highly qualified students.

Admissions Requirements

Applicants must apply directly to the LS&T program. In order to be capable of performing graduate level work, applicants should have background in at least one of the core disciplines of LS&T, namely, Cognitive/Educational Psychology, Computer Science, or Statistics. We will also consider applicants whose academic background is in Science or Math.

A student may apply to the Ph.D. program in LS&T after completing a bachelor’s degree (in which case a master’s degree must first be completed) or a master’s degree in one of the affiliated areas (Computer ­Science, Cognitive or Educational Psychology or Statistics) or a closely related area. Applicants with other degrees are welcome to apply if they can demonstrate their readiness through other means, such as GRE Subject exams in an affiliated area, or through academic or professional experience. GRE scores are strongly recommended, but not required, for all applicants. Inquiries about the GRE should be made to Dr. Neil Heffernan, the Program Director.

Research Labs/Research Groups

Causal Modeling Research Group

Prof. Sales
We build, analyze, and evaluate statistical causal models, primarily for large, complex, or messy datasets such as log data from EdTech or state administrative data. Our research includes developing novel principal stratification models for implementation and/or computer log data from randomized trials; methods for incorporating auxiliary data and machine learning into classical analyses of A/B tests, RCTs, or observational matching designs; and regression discontinuity analysis with flexible covariance models.


Creativity, Education, Affect, and Reasoning (CEDAR) Lab

Prof. Shaw
The CEDAR Lab conducts research on creative and flexible thinking in mathematics, reasoning of complex concepts, and how student experiences shape thinking and learning in STEM education. Our lab uses experimental methods, observational data, learning analytics, and qualitative methodologies in an effort to better understand teaching and learning in STEM subjects. See more at cedarlab.org

Educational Data Mining Research Group

Profs. Beck, Heffernan, & Whitehill 
Large datasets of students’ fine-grained interactions (e.g., student S answers math problem X with answer Y at time T) with intelligent tutoring systems, educational interventions, and massive open online courses (MOOCs) enable the exploration and optimization of how learners learn and how teachers teach. By harnessing methods from machine learning -- such as probabilistic graphical models, Markov chains, and deep neural networks -- we can develop more accurate predictors of which and when students will succeed, fail, persist, need help, etc. These predictors can, in turn, serve as the basis for both human-assisted and automated interventions to improve students’ learning outcomes and the personalization of learning.

Embodied Cognition In Mathematics Research Group

Prof. Ottmar
This research group carries out research about new ways of learning, using motor actions as well as cognitive thought. We investigate how children may better learn mathematics while exploring the physical space, getting a different understanding of math learning by gesturing, and using technology to guide them through 3D spaces.

Machine Perception of Human Learning Group

Profs. Whitehill, Heffernan & Beck
This group uses machine learning and computer vision to study how learners learn and how they emote while they learn. Particular interests include the training of deep neural networks to recognize students’ facial expressions during learning tasks, and the development of real-time cyberlearning systems that respond instantaneously to learners’ current cognitive, affective, and linguistic needs.

Math, Abstraction, Play, Learning, And Embodiment (MAPLE) Lab

Prof. Ottmar
Teaching and learning mathematics is a highly complex social, exploratory, and creative process. We design novel dynamic technologies that make “math come alive” (Graspable Math, From Here to There!) and use eye tracking, mouse gestures, and log files to explore the coordination of attention, cognition, gestures, and strategies when solving mathematical equations. We also use a variety of applied multilevel quantitative methods, observational measures, and assessments to examine the efficacy of instructional, social, and emotional classroom Interventions that can improve K-12 math teaching, learning, and engagement. https://sites.google.com/site/erinottmar/

Quantitative Methods in the Learning Sciences

Profs. Sales, Ottmar, Heffernan, Somasse & Zou
This research group is focus on rigorous quantitative methods such as hierarchical linear models (which is a typical method to use when students are nested inside teachers and teachers are nested inside schools). Other topics include issues that are used a lot in Learning Science like structural question modeling, longitudinal data analysis, propensity score matching, regression discontinuity designs, quasi-experimental designs and advanced topics like principal stratification. The faculty in this group like to apply (and adapt) statistical methodologies to solves the problems they are working on.

Running Classroom Experiments on the Web

Profs. Heffernan, Beck, Ottmar, Shaw & Sales
We use a number of web-based platforms and technologies (i.e. ASSISTments, MathSpring, GraspableMath) to conduct randomized-controlled trials in K-12 mathematics classrooms. These studies help us understand “what works” with regards to different pedagogical techniques, content, feedback, and tasks, and helps us develop a better understanding of the mechanisms guiding learning. Together the group has over 100 randomized controlled trials running each year. There are a set of methodologic issues that their research group tackle related to student-level randomized controlled assignment.
https://www.etrialstestbed.org/