Here’s my first post of my first day at AAAS, with lots of interesting topics topics that probably deserve their own posts I’m going to have to split this up. A reminder if you want more current info about what I’m up to I’m updating our twitter account.
As tends to be the case here the first day started with a tough decision about which panels to try to attend. I started out with Learning Research and Educational Practice: How Can We Make Better Connections. This dealt with the basic research on learning and cognition feeds into the people who actually use that info for teaching, and how it feeds the other way back into basic research. This was the big theme that was the importance of the feedback loop illustrated by every talk using a cycle image at some point. Initially I came in expecting to see how educators are using basic science, but saw that basic science has as much to gain from learning how educators are using research.
The first speaker was Ken Koedinger from Science of Learning Center in Pittsburgh. He focused on describing Algebra Cognitive Tutor, a program that is based on cognitive models to help teach algebra to high school students. It also allows the collection of data about how the students are using the program and what they need help with. One interesting result he presented was from how students differed in success with different formats for a math question. One idea was difficulties resulted in understanding the question, so a hint for comprehension was added, but had little effect on the success rate. However providing the equation had a large improvement. Similar success was also possible simply by splitting the question to explicitly ask what the equation was, followed by asking for the final answer rather than implying the two stages.
The second speaker was Philip Bell, from the Learning in Informal and Formal Environments Center here at UW. He described a few examples of the his work looking at how students can learn from the 79% of the time not spent in school, and how to take advantage of their interests. He also explained a bit about work modifying kits used in elementary schools to teach science. He also pointed out that this work could be harmed by the school district he is working with cutting the funds given to teachers for K-7 science lessons.
The last talk was by Javiar Movellan from UCSD , an engineer working on robotics and artificial intelligence and recently has been looking to see how the robots could be used for education. He noted how this experience has made him realize the importance of trying to bring the research to the field early on, rather than trying to develop basic research to a large extent than moving to the field. The first lesson was with their smile detection, which was known as being particular sophisticated in the lab, but completely failed once they took the robot into a classroom. Working closely with the students also led to a new discovery of how timing of motion is important for having the children treat the robot as a social entity. They also learned from the children how to have the robot respond to mistreatment by mimicking an emotional response like crying when hurt. This work has led to Dr. Movellan to question dichotomies between ideas like basic and applied research or what’s considered science vs. engineering.