Posted on Feb 9, 2015

Summary

We are looking for outstanding students interested in a Research Experience for Undergraduates (REU) at the CatLab at Vanderbilt University this summer 2015. Our REU is part of an NSF-funded project entitled Perceptual Categorization in Real-World Expertise. This project uses online behavioral experiments to understand the temporal dynamics of perceptual expertise, measuring and manipulating the dynamics of object recognition and categorization at different levels of abstraction and assessing how those dynamics vary over measured levels of expertise, using computational models to test hypotheses about expertise mechanisms. Students have opportunities to work on projects ranging from the development of online experiments, development of analysis routines, and development and testing of computational models. This REU is especially appropriate for students interested in applying to graduate programs in psychology, vision science, cognitive science, or neuroscience. The REU provides a $5000 summer stipend, $500 per week for ten weeks; an additional $150 per week helps offsets the cost of housing and meals; a $250 travel allowance is also provided. REUs are restricted to undergraduate students currently enrolled in a degree program and must be U.S. citizens, U.S. nationals, or permanent residents of the United States. Undergraduates who are students at Vanderbilt may have the opportunity to continue the REU experience into the 2015-16 academic year.

Project Description

People with perceptual expertise are skilled at making rapid identifications of specialized objects at a glance, often in poor light and camouflage. Forensic experts can accurately match exemplars to latent fingerprints that may be small, distorted, or smudged. Expert radiologists can quickly categorize medical images as normal or cancerous. Bird experts can identify species at long distances and in poor light. This project examines perception, categorization, and identification along the continuum from novice to expert performance in two real-world perceptual domains: analysis of latent fingerprints and zoological identification of birds. Forensic expertise was chosen because of its real-world importance in criminal and civil investigations and homeland security. Bird expertise was chosen not because it is important to understand bird identification per se, but because it is an excellent domain for studying a broad continuum of real-world expertise with a large and willing subject population. The overall aim is to understand how fundamental perceptual and cognitive mechanisms are tuned and modified by experience and expertise. The models arising from this project will enable us to understand the development of real-world perceptual expertise and to validate theoretically-grounded measures of expert performance.

Why study perceptual expertise? Just as gifted athletes push the limits of their bodies, or prize-winning mathematicians push the limits of their minds, perceptual experts push the limits of their perceptual systems. Perhaps with better markers of perceptual expertise and a better understanding of how people become perceptual experts, we could identify potential perceptual experts more effectively, train new perceptual experts more efficiently, and evaluate existing perceptual experts more thoroughly. Studying perceptual expertise can also help inform our understanding of the kinds of everyday expertise that we all have, such as recognizing faces or reading words. This can yield new insights into education and workforce training along with new insights into how the ravages of brain damage or disease might lead to perceptual and learning deficits and potentially inform future breakthroughs in evaluation, intervention, or treatment.

REU Experience

There are several components of the ongoing research that an REU could be involved in: The research in my lab requires developing computerized experiments for online testing using Adobe Flash, Javascript, or the Amazon Mechanical Turk API. It requires analyzing expertise performance data using psychometric tools such as IRT. It requires analyzing behavioral data using customized programs in Matlab or R. It requires characterizing behavioral data by fitting computational models, such as the diffusion model, which in turn requires adapting and writing simulation and model fitting code in Matlab or R. It requires developing and testing new models that explain how behavior changes over the course of expertise. Developing computer programs for experimental design, data analysis, and computational modeling; every graduate student and postdoctoral fellow in an active lab spends considerable time writing computer code in Matlab, R, Flash, or other platforms in order to conduct our research. Some programming experience is expected; while experience with some of these particular programming platforms is desirable, it is not necessary.

Following standard practice in my lab for the past two decades, undergraduates will be paired with a senior graduate student or postdoctoral fellow and work with them on a specific concrete project. They will read relevant research pertaining to the project underway and they will attend lab meetings. They will meet regularly with the graduate student or postdoctoral fellow as well as myself to discuss goals, achievements, and challenges. Undergraduates in my lab typically begin by working on an ongoing project but as their skills develop and their interests blossom, they often end up working on new projects more independently. Lab meetings often offer opportunities to discuss broader issues related to topics like responsible conduct of research and professional development.

Application Procedure

Please send the following to Professor Thomas Palmeri at thomas.j.palmeri@vanderbilt.edu; we will begin reviewing applications March 1, 2015.

  • 1-2 page cover letter describing your educational experience and research background, your interest in the research in the CatLab, and your future goals; please also describe your computer programming experience.
  • Resume or vita.
  • Transcripts.
  • Name of 2 individuals who could comment on your educational background, experience, and potential for research.

Eligibility

Undergraduate student participants supported with NSF funds in REU programs must be U.S. citizens, U.S. nationals, or permanent residents of the United States. An undergraduate student is a student who is enrolled in a degree program (part-time or full-time) leading to a baccalaureate or associate degree. Students who are transferring from one college or university to another and are enrolled at neither institution during the intervening summer may participate. High school graduates who have been accepted at an undergraduate institution but who have not yet started their undergraduate study are also eligible to participate. Students who have received their bachelor’s degrees and are no longer enrolled as undergraduates are generally not eligible to participate.