Research Areas

Mechanisms of Statistical Learning

My dissertation research is focused on the mechanisms of pattern learning. I use statistical learning paradigms, in which subjects are briefly exposed to information that contains some unknown structure and then are tested on how well they learned the structure. I am investigating how basic cognitive processes like memory, attention, and perception constrain statsitical learning, and how statistical learning itself alters these constraints.

  • de Leeuw, J. R., & Goldstone, R. L. Memory constraints affect statistical learning; Statistical learning affects memory constraints. To be presented at the 2015 Cognitive Science Society Meeting.

  • de Leeuw, J. R., & Goldstone, R. L. (2014). Predictable sequences promote the learning of visual statistical regularities. Presented at the 4th Annual Midwest Cognitive Science Conference, Dayton, OH.

Tools for Running Psychology Experiments Online

The increasing sophistication of web browsers and the emergence of online labor marketplaces like Mechanical Turk allow for running complex behavioral experiments online. I have developed an open-source JavaScript library called jsPsych that helps researchers build experiments that run in a web browser. More information is available at the jsPsych website - http://www.jspsych.org.

  • de Leeuw, J. R. (2015). jsPsych: A JavaScript library for creating behavioral experiments in a Web browser. Behavior Research Methods, 47(1), 1-12. doi:10.3758/s13428-014-0458-y

  • de Leeuw, J. R., Coenen, A., Markant, D., Martin, J. B., McDonnell, J., Rich, A., & Gureckis, T. (2014). Online experiments using jsPsych, psiTurk, and Amazon Mechanical Turk. In P. Bello, M. Guarini, M. McShane, & B. Scassellati (Eds.), Proceedings of the 36th Annual Conference of the Cognitive Science Society (pp. 42-43). Austin, TX: Cognitive Science Society.

Accuracy of Response Time Measurements in Online Experiments

Online experiments are typically built with software that is different from the widely-adopted laboratory standards. One of the most popular options is programming in JavaScript. Ben Motz and I conducted an experiment in which we measured response times during a visual search task with both JavaScript and Psychophysics Toolbox for MATLAB (PTB; a common platform in laboratory-based studies). We set up the experiment so that participants could simultaneously do the visual search task in both software environments, providing a robust control for non-software effects. Using a hierarchical Bayesian model of the results, we showed that there are small delays in RT measurements collected with JavaScript relative to PTB, but there is no difference in the variability of measurements or estimated search function parameters.

  • de Leeuw, J. R., & Motz, B. A. (2015). Psychophysics in a web browser? Comparing response times collected with JavaScript and Psychophysics Toolbox in a visual search task. Behavior Research Methods. In press.

Large-scale Classroom Experiments to Improve Teaching

Laboratory studies of learning often lack ecological validity, making it difficult to form prescriptive recommendations for teaching practices. I am working with Paulo Carvalho, David Braithwaite, Ben Motz, and Rob Goldstone on methodological innovations that bring carefully controlled studies of learning into the classroom. In one such experiment, we created an online homework app that students used to learn about concepts of central tendency. We were able to manipulate various aspects of the homework and then check for subsequent effects on students' in class exams. Our ongoing work is focused on getting tools in the hands of teachers to allow them to create their own experiments so that they can scientifically explore how best to teach their own classes.

  • Braithwaite, D. W., Carvalho, P. F., de Leeuw, J. R., & Goldstone, R. L. (2014). Effectiveness of learner-regulated study sequence. Poster presented at the 55th Annual Meeting of the Psychonomic Society, Long Beach, CA.

  • Carvalho, P. F., Braithwaite, D. W., de Leeuw, J. R., Motz, B. A., & Goldstone, R. L. (2014). Sequencing examples during concept learning. Poster presented at the 2014 CIRCLE Conference - Integrating cognitive science with innovative teaching in STEM disciplines, St. Louis, MO.

Methodological Issues in Categorical Perception Experiments

Learned categorical perception (CP) effects occur when learning to categorize objects in a new way changes how people perceive and/or make judgments about those objects. Jan Andrews, Ken Livingston, and I are investigating how often ignored or downplayed aspects of methodological design change the kinds of learned CP effects that are observed. We are finding that different tasks for assessing CP find different effects, and that the effects found are often contingent upon the kinds of stimuli that are used. We are systematically investigating these methodological design decisions to better understand the mechanisms that produce CP.

  • de Leeuw, J. R., & Andrews, J. Using a task-filled delay during discrimination trials to examine different components of learned visual categorical perception. To be presented at the 2015 Cognitive Science Society Meeting

  • de Leeuw, J. R., Andrews, J., & Livingston, K. (2014). Learned visual categorical perception effects depend on method of assessment and stimulus discriminabilitiy. In P. Bello, M. Guarini, M. McShane, & B. Scassellati (Eds.), Proceedings of the 36th Annual Conference of the Cognitive Science Society (pp. 391-396). Austin, TX: Cognitive Science Society.

  • de Leeuw, J. R., Andrews, J., & Livingston, K. (2013). Variables influencing the nature of learned categorical perception effects. Poster presented at the 35th Annual Meeting of the Cognitive Science Society, Berlin, Germany.

How does Hunger Shape our Perceptual Processing of Food-like Stimuli?

It's a common television and movie trope: One character is near starvation, and suddenly objects and other characters start to look like food. While such an extreme effect is clearly hyperbole, a more subtle version in which food-like aspects of a stimulus are emphasized with increasing hunger seems plausible. Peter Todd and I are conducting an ongoing research project in which we ask subjects to decide whether a blended image of a food object and a non-food object is more like the food object or more like the non-food object. Our results so far indicate a small but reliable effect of hunger on these judgments. Hungry subjects tend to pick the food item as being more similar more often than satiated subjects.

  • de Leeuw, J. R., & Todd, P. M. (2013). The influence of hunger on categorical perception of food and non-food items. Presented at the 3rd Annual Midwest Cognitive Science Conference, Columbus, OH.

EEG Correlates of Learning with Naturalistic Educational Stimuli

Studies of learning in neuroscience are usually unrealistically narrow: learners are exposed to isolated learning opportunities with materials that are designed first and foremost for the purpose of a well controlled experiment. The learning that takes place in the classroom is very different. Rather than being isolated, concepts are embedded in a rich context. Ben Motz, Rob Goldstone, Adam Maltese, Joshua Danish, Jennifer Eastwood, and I are investigating the feasibility of identifying EEG correlates of successful learning with naturalistic educationally-relevant materials. We show subjects pre-recorded lectures on a computer monitor while recording EEG from a mobile EEG system. At key points during the lecture, concepts are introduced that the subjects are later tested on. We then compare aspects of the EEG signal during the points in the lecture that students succesfully learned and the points in the lecture that they failed to learn.

  • de Leeuw, J. R., Motz, B. A., Eastwood, J. L., Maltese, A. V., Goldstone, R. L., Danish, J. A. (2015) Needle in the neural haystack: Electroencephalograph signatures of concept learning while viewing naturalistic educational materials. Proceedings of the 2015 American Educational Research Association Annual Meeting.

Emergent Group Behavior in Aquatic Robots

Schooling behavior is often modeled as an emergent phenomenon that results from many agent-to-agent interactions. These models usually make the assumption that agents in the group can detect and react to other agents in the group. This project, a collaboration with John Long, Marianne Porter, and Ken Livingston, tested the hypothesis that the ability to sense other group members is not necessary to produce coordinated behavior when the members of the group share a common goal. We built a set of aquatic robots that can perform simple behaviors such as tracking a light source. We measured the coordination of the group by tracking the location of each individual robot over time and computing how much the shape of the group changed. When the group members all shared the goal of swimming towards a stationary light source, the group's behavior became coordinated as a result of the physical bumping and jostling of the robots as they tried to swim towards the same target.

  • de Leeuw, J. R., Livingston, K. R., Porter, M. E., & Long, J. H. Jr. (2013). When swarm intelligence isn't: Common goals alone explain emergence of group coordination in asocial embodied robots. Presented at the Society for Integrative and Comparative Biology, San Francisco, CA.

Publications & Presentations

Journals

de Leeuw, J. R., & Motz, B. A. (2015). Psychophysics in a web browser? Comparing response times collected with JavaScript and Psychophysics Toolbox in a visual search task. Behavior Research Methods. In press.

Goldstone, R. L., de Leeuw, J. R., & Landy, D. H. (2015). Fitting perception in and to cognition. Cognition, 135, 24-29. doi:10.1016/j.cognition.2014.11.027

de Leeuw, J. R. (2015). jsPsych: A JavaScript library for creating behavioral experiments in a Web browser. Behavior Research Methods, 47(1), 1-12. doi:10.3758/s13428-014-0458-y

Long, J.H. Jr., Krenitsky, N. M., Roberts, S. F., Hirokawa, J., de Leeuw, J. R., and Porter, M. E. (2011). Testing biomimetic structures in bioinspired robots: how vertebrae control the stiffness of the body and the behavior of fish-like swimmers. Integrative and Comparative Biology, 51(1), 158-175.

Proceedings

de Leeuw, J. R., & Goldstone, R. L. Memory constraints affect statistical learning; Statistical learning affects memory constraints. To be presented at the 2015 Cognitive Science Society Meeting.

de Leeuw, J. R., & Andrews, J. Using a task-filled delay during discrimination trials to examine different components of learned visual categorical perception. To be presented at the 2015 Cognitive Science Society Meeting.

de Leeuw, J. R., Motz, B. A., Eastwood, J. L., Maltese, A. V., Goldstone, R. L., Danish, J. A. (2015) Needle in the neural haystack: Electroencephalograph signatures of concept learning while viewing naturalistic educational materials. Proceedings of the 2015 American Educational Research Association Annual Meeting.

de Leeuw, J. R., Andrews, J., & Livingston, K. (2014). Learned visual categorical perception effects depend on method of assessment and stimulus discriminabilitiy. In P. Bello, M. Guarini, M. McShane, & B. Scassellati (Eds.), Proceedings of the 36th Annual Conference of the Cognitive Science Society (pp. 391-396). Austin, TX: Cognitive Science Society.

Krishnamurthy, P., Khorrami, F., de Leeuw, J. R., Porter, M. E., Livingston, K., & Long, J. H. (2010). An electric ray inspired biomimetic autonomous underwater vehicle. In Proceedings of the American Control Conference pages 5224-5229.

de Leeuw, J. R., & Livingston, K. (2009). A self-organizing autonomous prediction system for controlling mobile robots. In Chen, K., Moustafa, K. A. F., and Karras, D. A., editors, Proceedings of the International Conference on Automation, Robotics and Control Systems (ARCS-09), pages 123-129. ISRST.

Krishnamurthy, P., Khorrami, F., de Leeuw, J. R., Porter, M. E., Livingston, K., & Long, J. H. (2009). A multi-body approach for 6dof modeling of biomimetic autonomous underwater vehicles with simulation and experimental results. In Control Applications, (CCA) & Intelligent Control, (ISIC), 2009 IEEE, pages 1282-1287.

de Leeuw, J. R., & Livingston, K. (2007). When less is more: Sensor resolution and learning. In Berthouze, L., Prince, C. G., Littman, M., Kozima, H., and Balkenius, C., editors, Proceedings of the Seventh International Conference on Epigenetic Robotics: Modeling Cognitive Development in Robotic Systems.

Workshops & Tutorials

de Leeuw, J. R. Programming online experiments with jsPsych. Tutorial to be presented at the 2015 Cognitive Science Society Meeting.

de Leeuw, J. R., Coenen, A., Markant, D., Martin, J. B., McDonnell, J., Rich, A., & Gureckis, T. (2014). Online experiments using jsPsych, psiTurk, and Amazon Mechanical Turk. In P. Bello, M. Guarini, M. McShane, & B. Scassellati (Eds.), Proceedings of the 36th Annual Conference of the Cognitive Science Society (pp. 42-43). Austin, TX: Cognitive Science Society.

Presentations & Published Abstracts

Braithwaite, D. W., Carvalho, P. F., de Leeuw, J. R., & Goldstone, R. L. (2014). Effectiveness of learner-regulated study sequence. Poster presented at the 55th Annual Meeting of the Psychonomic Society, Long Beach, CA.

Carvalho, P. F., Braithwaite, D. W., de Leeuw, J. R., Motz, B. A., & Goldstone, R. L. (2014). Sequencing examples during concept learning. Poster presented at the 2014 CIRCLE Conference - Integrating cognitive science with innovative teaching in STEM disciplines, St. Louis, MO.

de Leeuw, J. R., & Goldstone, R. L. (2014). Predictable sequences promote the learning of visual statistical regularities. Presented at the 4th Annual Midwest Cognitive Science Conference, Dayton, OH.

de Leeuw, J. R., Andrews, J., & Livingston, K. (2013). Variables influencing the nature of learned categorical perception effects. Poster presented at the 35th Annual Meeting of the Cognitive Science Society, Berlin, Germany.

de Leeuw, J. R., & Todd, P. M. (2013). The influence of hunger on categorical perception of food and non-food items. Presented at the 3rd Annual Midwest Cognitive Science Conference, Columbus, OH.

de Leeuw, J. R. (2013). Common goals coordinate groups of asocial embodied robots. Video and poster presented as part of the 2013 IGERT National Poster & Video Competition, online at http://posterhall.org/igert2013/

de Leeuw, J. R., Livingston, K. R., Porter, M. E., & Long, J. H. Jr. (2013). When swarm intelligence isn't: Common goals alone explain emergence of group coordination in asocial embodied robots. Presented at the Society for Integrative and Comparative Biology, San Francisco, CA.

de Leeuw, J. R., Porter, M., Livingston, K., & Long, J.H. Jr. (2010). Evolving intelligence in autonomous, fish-like biorobots: does competition for resources matter? Poster presented at the Society for Integrative and Comparative Biology, Seattle, WA.

Hirokawa, J., Roberts, S., Frias, C., Krenitsky, N., de Leeuw, J. R., Long, J.H., Jr., & M.E. Porter. (2010). A self-propelled robotic swimmer as a biomechanical testbed: how swimming performance is modulated by the axial length of the intervertebral joints in a biomimetic vertebral column. Poster presented at the Society for Integrative and Comparative Biology, Seattle, WA.

Non-refereed

de Leeuw, J. R., & Goldstone, R. L. (2014). Context effects in visual statistical learning. Poster presented at the Annual IGERT in Brain-body-environment Systems Symposium, Bloomington, IN.

de Leeuw, J. R., & Todd, P. M. (2013). Meat-O-Vision: Testing a literary trope in the lab. Poster presented at the Annual IGERT in Brain-body-environment Systems Symposium, Bloomington, IN.

de Leeuw, J. R., Livingston, K., Porter, M. E., & Long, J.H. Jr. (2012). Common goals coordinate groups of asocial embodied robots. Poster presented at the Annual IGERT in Brain-body-environment Systems Symposium, Bloomington, IN.

de Leeuw, J. R. (2011). Testing selection pressures for small-world neural networks in virtual robots. Poster presented at the Annual IGERT in Brain-body-environment Systems Symposium, Bloomington, IN.

de Leeuw, J. R. (2010). Lessons from Creating Minds. The Vassar Alumni Club of Sarasota Springs. Invited presentation.

de Leeuw, J. R., & Livingston, K. (2007). Autonomous Learning in Mobile Robots. The 2007 Undergraduate Research Summer Institute Symposium. Oral Presentation.

Livingston, K., & de Leeuw, J. R.(2007). Virtual bottom-up robotics lab (with physical robot component). Workshop on Interactive Computer-Base Activities for Undergraduate Cog Sci Instruction at the 29th Annual Meeting of the Cognitive Science Society. Invited presentation.