Schedule for Fall 2011

September

September 9, 2011 
Opening Session, State of the Institute
--Marie Banich, Ph.D.
University of Colorado, Professor and Director, Institute of Cognitive Science 
--Tammy Sumner, Ph.D.
University of Colorado, Associate Professor, Cognitive and Computer Science 
--Mike Mozer, Ph.D.
University of Colorado, Professor, Computer Science

September 23, 201
--Sean Kang, Ph.D.
University of California, San Diego, Post-doctoral Research Scholar, Department of Psychology 
Title: "The Benefits of Retrieval Practice and Spacing for Learning"

Abstract: 
A wealth of evidence from basic memory research has revealed two robust phenomena that have direct implications for education: (1) Testing is not a neutral event in which one's knowledge is merely assessed; the act of retrieving information from memory enhances the later retention of that information, and (2) practice that is distributed or spaced out in time (relative to practice that is crammed/massed) leads to more durable learning. However, the potential for retrieval practice and spacing to be potent learning tools has not been fully realized in practice, perhaps in part because laboratory research has often not focused enough on the sorts of concrete procedural choices that arise in real-world learning contexts. I will present new research demonstrating the utility of retrieval practice and/or spacing for learning across diverse domains (i.e., language acquisition, inductive/category learning). Importantly, the current studies were designed to better reflect realistic learning situations, and I will emphasize the potential application of these findings for improving educational practice.

September 30, 201
--Paul Cohen, Ph.D.
University of Arizona, Professor and Director, School of Information: Science, Technology and Arts  
Title: "Verb Meanings for Robots"

Abstract:
By now we know that robots can associate sensory patterns with words, thereby "grounding" word meanings. However, these patterns might not function as words in language (e.g., patterns associated with verbs might not have anything like a case structure); and they do not necessarily function as word meanings, either, in that they might be "semantically impotent," doing nothing for the robot.   I will talk about what we want word meanings to do for robots (i.e., words help robots imagine scenes) and sketch some of our recent work on learning word meanings.  The last part of my talk will be about nonliteral language, such as metaphor, and escaping from the "grounding" we tried so hard to achieve.  

Bio:
Paul Cohen is Professor of Computer Science and Director of the School of Information: Science, Technology and Arts. His research is in AI and Cognitive Science, and focuses on two broad questions:  How can robots learn natural language and communicate with us in language, and how can information science and technology help to fix the problems of public education?

October

October 7, 2011 
--Carol Seger, Ph.D.
Colorado State University, Professor, Cognitive Psychology 
Title: "Decision Making and Learning in the Corticostriatal System" 

Abstract:
The corticostriatal system is a recurrent network connecting basal ganglia with cortex through which the basal ganglia exert a plastic modulatory influence on cortical representations.  Corticostriatal networks participate in a broad variety of cognitive processes, including executive functions, decision-making, response selection, and sequence and syntax processing.  I will discuss several recent studies from my laboratory exploring the roles played by corticostriatal systems during categorization and other decision-making tasks.  I will address how dopamine mediated plasticity within the corticostriatal system allows for continuous learning, which I will illustrate with studies using reinforcement learning approaches to model brain activity during category learning.  If time permits, I will conclude with a discussion of how the corticostriatal system interacts with the medial temporal lobe memory system during learning and memory tasks.

October 28, 2011 
--Ken Koedinger, Ph.D.
Carnegie Mellon University, Professor, Human Computer Interaction Institute 
Title: "The Knowledge-Learning-Instruction (KLI) Framework: Bridging the Science-Practice Chasm to Enhance Robust Student Learning"

Abstract:
Despite the accumulation of substantial cognitive science research relevant to education, there remains confusion and controversy in the application of research to educational practice. In support of a more systematic approach, my colleagues and I in the Pittsburgh Science of Learning Center (see learnlab.org) have developed the Knowledge-Learning-Instruction (KLI) framework.  KLI promotes the emergence of instructional principles of high potential for generality, while explicitly identifying constraints of and opportunities for detailed analysis of the knowledge students may acquire in courses. Drawing on research across domains of science, math, and language learning, we illustrate the analyses of knowledge, learning, and instructional events that the KLI framework affords. We present a set of three coordinated taxonomies of knowledge, learning, and instruction. For example, we identify three broad classes of learning events: a) memory and fluency processes, b) induction and refinement processes, c) understanding and sense-making processes, and we show how these can lead to different knowledge changes and constraints on optimal instructional choices. 

November

November 4, 2011 
--Tor Wager, Ph.D.
Institute of Cognitive Science, Psychology & Neuroscience, University of Colorado
Title: "Shared and Divergent Representations of Physical and Emotional Pain in the Central Nervous System"

Abstract:
Pain is a subjective experience created at the intersection of somatosensation and meaning.  Because of its complex origin in the central nervous system, pain has defied objective measurement, making it difficult to study its genesis and develop effective treatments.  Here, I present new efforts to use machine learningtechniques to develop objective, interpretable biomarkers for pain in the human brain. I demonstrate that physical pain can be differentiated from socially induced emotional distress by fine-grained patterns of activity within both somatosensory and meaning-generation systems.  The resulting biomarkers for physical and social "pain" could be used as targets for future studies of how both psychological and pharmacological treatments influence each type of distress.   

November 11, 2011 
--John Lynch, Ph.D.
University of Colorado, Leeds School of Business, Professor and Director of Center for Research on Consumer Financial Decision Making

November 18, 2011
--Tamara Sumner, Donna Caccamise, Lee Becker, Les Sikos, and Michael Mozer
Institute of Cognitive Science
Title: "Findings and Recommendations from the ICS Self Study"

Abstract:
Self Study Committee: Tamara Sumner, Donna Caccamise, Lee Becker, Les Sikos, Michael Mozer

In this colloquium, we will present and discuss the results and recommendations emerging from the ICS Self Study process, taking place this semester as part of the CU Academic Review Process. The recommendations include expansions to our research programs, changes to our degree and certificate programs, and changes to our communication processes. We will also review and discuss the resource requests the Institute is asking of the University, including additional faculty lines, graduate student support, administrative support, and research facilities. This process will inform the Institute's Strategic Plan, so please be sure to come and contribute your energy and ideas about how to improve the Institute.

November 21-25, 2011 
--Fall Break/Thanksgiving Break

December

December 2, 2011 
--William Penuel, Ph.D.
University of Colorado, Professor, Educational Psychology and Learning Sciences, School of Education

December 9, 2011 
--Last Day of Classes