Frank E. Ritter
January 1999
I have had a wide range of teaching experience in the last six years, from experimental psychology to user interface design, essentially the range of cognitive science. It includes group teaching, module development, use of the web and computers as delivery media, external teaching, and external evaluation of teaching. This teaching was done in a slightly different system, so some explanation and equivalencies will be included, for example, the use of 'modules' in the UK to mean 'classes' in the US. These difference have helped me understand the many types of instruction and their relative advantages.
My early teaching experiences were to provide of subsections of larger modules, and to run tutorial groups (similar to US discussion sections). While I enjoy teaching in teams of two, some modules had up to five instructors, which led to a fragmented presentation. I developed two second year modules, one on visual cognition and one a 6-week practical (lab methods) class on Card, Moran and Newell's (1983) keystroke level model of computer users. In the practical, students ran an experiment, and analyzed the results with respect to keystroke level model. Both modules were dropped when the second year was reorganized, but after teaching the keystroke-level model several times I was able to create a teaching aid/design tool (Nichols & Ritter, 1995) .
I continue to teach and develop modules on cognitive modeling as AI programming. Two of these modules continue to be used, one for our final year undergraduates and a paired module for graduate students. These modules include the Psychological Soar Tutorial that I developed with Richard Young. The tutorial is designed to introduce the Soar AI and cognitive architecture. It includes a web-based set of notes and exercises to be performed while running Soar, or with the static traces that are included. I know that it is used in at least three other universities. With Fernand Gobet, I have developed a set of additional exercises based on other Soar models. These models include a model of categorization (Miller & Laird, 1996), a model of expert to novice transitions, and a model of subtraction (Ritter, Jones, & Baxter, in press) . We have disseminated this tutorial and its lessons widely (Ritter, 1997; Ritter, Jones, & Young, 1997) . Students in this module are assessed by coursework (which is essential when teaching a procedural skill like this), a project, and a project presentation. I have been told that the experience my students gain by presenting their work orally has gained employment for more than one of them. It is used by the university as a case study in combinging teaching and research in the use of technology for new lecturers.
The module I created with the largest impact on the department is Introduction to British Psychology. This is a first year module to help incoming international students. It provides additional support and feedback that is important to help them adjust to what is perhaps a new system in a second language. The feedback includes additional feedback on written work and on an oral presentation. This in particular is important because in UK universities students receive only their marks as feedback -- no comments, model answers, or their exam. This type of support helps a slightly wider range of students adjust to university, and has allowed me to increase our international admissions by a factor of ten. This has had a significant impact on the department's finances, now providing £240,000 ($360,000) per year.
I convened (organized) several modules outside the department, and help liase with other departments about teaching. The modules I taught in CS were good for Psychology, in that we got credit for the teaching, and good for CS who needed them taught. These modules have been mostly taken by students doing a masters in information technology or other engineering course. These modules help teach psychology to designers who will not get this information in their other modules. I was successful enough that computer science hired someone from our department to teach two courses in this area. I picked up some of their teaching in the health psychology masters program (i.e. Issues in Ergonomics). There is a draft of a textbook as web pages that David Gilmore, Elizabeth Churchill and I are writing. Teaching an HCI module would encourage progress on the book.
The module People and Computers in the Workplace allows me to think about the larger scale issues that computers and their use raise, such as workload, CSCW, and management of error. This module is taught in a graduate style -- after an introductory lecture in the first hour of each week, students from 4-10 people groups present a paper or concept to complement, contradict, or supplement the lecture. Students find this nerve-racking at first, but their speaking skills and presentations improve over the semester. Their evaluation forms that the end of the module indicate that they value it.
I have supervised about ten final year projects. These come out at a level between the US final year and US (one-year) master projects. In one case the work was good enough that we are attempting to publish it. I have also supervised five students on intercalated year (co-op) programs. They get a lot out of it, and I have too. One student project we eventually worked up into a journal article (Kuk, Arnold, & Ritter, in press) , another we have submitted to Communications of the ACM.
I also oversee the cognitive science degree program at the U. of Hertfordshire as one of two external examiners. External examiners, usually at the full professor level, are the way that the UK system attempts to keep parity between university degrees, including content and level. This basically means that I proofread half their exams, confirm that their marks are consistent within their program and consistent with UK standards, and advise on disciplinary issues. This has allowed me to see a wide range of modules in another university.
To summarize, I have taught in the areas of experimental psychology, cognitive psychology, HCI, programming, AI, and cognitive modeling. I am happy to teach in these areas, they all support my research in understanding learning, modeling learning, and applying models of learning. In most modules since 1996 I have had web pages at least as a home for the module but often to provide additional material. I have tried including student presentations. In most modules they appear to work well. I would be interested in providing similar modules to undergraduates and graduates. There are other areas of CS that I could teach as well.
I have taught on several master modules. The UK has only started to introduce a taught component to the PhD programs, so I have only taught PhD students, per se, at special workshops organized by the Society for AI and Simulation of Behaviour.
I have been working on a book (O'Shea, Lehtinen, Ritter, & Langley, in preparation) to motivate the importance of sequence effects in learning for graduate students in education, psychology, and AI. This project arose out of a European Science Foundation program on learning in humans and machines. Our task force was charged to look at how the order that material, for both facts and skills, is presented or explored by a learner can strongly influence what is learned, how fast performance increases, and sometimes, even that the material is learned at all. The book designed to look like Mindstorms when it is completed, in that it includes numerous questions that should lead to further work by PhD students and provide food for thought for professionals working in these areas such as educational researchers.
I have supervised four PhD students, one who has submitted, and one who should submit in early 1999. I have enjoyed this, and in one case the student is has already been and will be very productive. I have also sponsored five visiting graduate students. Four students came from the Technical University of Berlin, as their degree required them to work in an outside lab as a type of co-op experience. One came from Georgia Tech to gain cognitive modeling skills. I have supervised several HCI and AI related master degree projects from the Information Technology Institute in the CS department. These students are more mixed in their abilities and are more challenging to supervise. They all did well in the end, and one was able to publish his results, and one has a thesis that I am able to cite.
I enjoy teaching. It is important, and I have found ways to tie it to my research so that both benefit.
Card, S., Moran, T., & Newell, A. (1983). The psychology of human-computer interaction. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc.
Kuk, G., Arnold, M., & Ritter, F. E. (in press). Using event history analysis to model the impact of workload on an air traffic tactical controller's operations. Ergonomics.
Miller, C. S., & Laird, J. E. (1996). Accounting for graded performance within a discrete search framework. Cognitive Science, 20, 499-537.
Nichols, S., & Ritter, F. E. (1995). A theoretically motivated tool for automatically generating command aliases. In CHI '95, Human Factors in Computer Systems. 393-400. New York, NY: ACM.
O'Shea, T., Lehtinen, E., Ritter, F., & Langley, P. (Eds.). (In preparation). In order to learn: How ordering effects in machine learning illuminates human learning and vice versa.
Ritter, F. (1997). WWW presentation of overheads & exercises. CTI Psychology Software News, 7(2), 46.
Ritter, F. E., Jones, G., & Young, R. M. (1997). Report on Tutorial 1: Introduction to the Soar cognitive architecture. AISB Quarterly, 98, 50.
Ritter, F. E., Jones, R. M., & Baxter, G. D. (in press). Reusable models and graphical interfaces: Realising the potential of a unified theory of cognition. In U. Schmid, J. Krems, & F. Wysotzki (Eds.), Mind modeling - A cognitive science approach to reasoning, learning and discovery. Lengerich: Pabst Scientific Publishing.