Integrated Development Environments (IDE) generate multiple graphical and textual representations of programs. Co-ordination of these representations during program comprehension and debugging can be a complex task. In order to better understand the role and effectiveness of multiple representations, we conducted an empirical study of Java program debugging with a professional, multi-representation IDE. We found that program code and dynamic representations (dynamic viewer, variable watch and output) attracted the most attention of programmers. Static representations like Unified Modeling Language (UML) Diagrams and Control Structure Diagrams (CSD) saw significantly lesser usage. We analyzed gaze patterns by segmenting the debugging sessions into three, five and fifteen minute intervals, and classifying gazes into short and long gazes. Novel data mining techniques were used to detect high frequency patterns from eye tracking data. Visual pattern differences were found among participants based on their programming experience, familiarity with the IDE and debugging performance.

This BOF is related to the Special Session Transforming the CS Classroom with Studio-Based Learning (SBL). SBL promotes learning in a collaborative context by having students construct, present, review and refine their work with the guidance of peers and teachers. A team of CS educators and education experts have been implementing and evaluating SBL in CS courses over the past five years. The BOF will introduce SBL to the SIGCSE audience, and engage them in a discussion of the potential of, evidence for, and practical advice regarding SBL as an instructional approach that can motivate as well as teach students. Discussions will include "war stories" from teachers who have adopted the approach in their courses and hands-on activities to help participants apply SBL to their courses.

The studio-based learning (SBL) model aims to promote learning in a social and collaborative context by having learners construct, iteratively refine, and critically review design artifacts under the guidance of instructors and disciplinary experts. Recognizing the potential of SBL to foster the development of not only technical design skills, but also increasingly-coveted communication and collaboration skills, we have been adapting the approach for computing education, implementing it in various courses, and evaluating its impact over the past 5 years. This effort now involves 26 computing courses at 15 institutions in seven states. To our knowledge, this represents the most systematic implementation and evaluation of a pedagogy for computing education to date. This special session will introduce SBL to the SIGCSE community and facilitate a discussion and exchange of ideas. In addition to oral and poster presentations of the SBL model and its evaluation results, the session will feature "war stories" from teachers who have applied the model to their courses and hands-on activities to help attendees apply SBL to their courses.

This paper presents an experience in designing, implementing, and evaluating a studio-based learning model for CS2. Adapted from architecture and art education, as well as from collaborative problem-solving pedagogies, studio-based learning has shown great promise for computing education. Key elements of studio-based learning include exploring multiple solutions to a problem, justifying the choice of one solution, and being subject to, as well as providing, peer reviews. We describe the design, implementation, and revision of a studio model for CS2, and then present the results of an evaluation of the model when compared to traditional instruction in CS2.

While the demand for college graduates with computing skills continues to rise, such skills no longer equate to mere programming skills. Modern day computing jobs demand design, communication, and collaborative work skills as well. Since traditional instructional methods in computing education tend to focus on programming skills, we believe that a fundamental rethinking of computing education is in order. We are exploring a new "studio-based" pedagogy that actively engages undergraduate students in collaborative, design-oriented learning. Adapted from architectural education, the studio-based instructional model emphasizes learning activities in which students (a) construct personalized solutions to assigned computing problems, and (b) present solutions to their instructors and peers for feedback and discussion within the context of "design crits." We describe and motivate the studio-based approach, review previous efforts to apply it to computer science education, and propose an agenda for multi-institutional research into the design and impact of studio-based instructional models. We invite educators to participate in a community of research and practice to advance studio-based learning in computing education.

Cognitive models and empirical studies of problem solving in visuo-spatial and causal domains suggest that problem solving tasks in such domains invoke cognitive processes involving mental animation and imagery. If these internal processes are externally manifested in the form of eye movements, such tasks present situations in which the trajectory of a user's visual attention can provide clues regarding his or her information needs to an Attentive User Interface [Vertegaal 2002]. In this paper, we briefly review research related to problem solving that involves mental imagery, and describe an experiment that looked for evidence and effects of an imagery strategy in problem solving. We eye-tracked 90 subjects solving two causal reasoning problems, one in which a diagram of the problem appeared on the stimulus display, and a second related problem that was posed on a blank display. Results indicated that 42% of the subjects employed mental imagery and visually scanned the display in a correspondingly systematic fashion. This suggests that information displays that respond to a user's visual attention trajectory, a kind of Attentive User Interface, are more likely to benefit this class of users.

This research began by investigating the literature on student learning from algorithm animations and conducting experimental studies of an algorithm visualization system. The results led us to develop CAROUSEL (Collaborative Algorithm Representations Of Undergraduates for Self-Enhanced Learning), using which students created expository representations of algorithms, shared their representations with others, evaluated each other's representations and discussed them. The system and the activities of representation creation, sharing, evaluation and discussion that it supports were then studied in three experiments, which are summarized. They show a significant positive relationship between these constructive and collaborative activities and algorithm learning, which suggests that this is a beneficial pedagogical approach for introductory courses on algorithms.

Algorithm animation systems have not met the initial promise they seemed to hold for teaching algorithms to undergraduate computer science students. A qualitative study of algorithm learning strategies of students and a usability study of a hypermedia algorithm animation system led to the conclusion that current algorithm animations may actually do more to hinder than help undergraduates develop a solid understanding of the algorithm, because of their learning strategies. Students relied on a single representation of an algorithm, misinterpreted the limitations and specifics of that representation, avoided using outside media sources and worked primarily within groups to understand algorithms. Thus, algorithm animation may have limited usefulness for this population. Therefore, to discourage students from relying on a single representation and to encourage them to be more self-directed and engaged learners, we have students create, share and collectively interpret their own algorithm representations.

With the development of novel interfaces controlled through multiple modalities, new approaches are needed to analyze the process of interaction with such interfaces and evaluate them at a fine grain of detail. In order to evaluate the usability and usefulness of such interfaces, one needs tools to collect and analyze richly detailed data pertaining to both the process and outcomes of user interaction. Eye tracking is a technology that can provide detailed data on the allocation and shifts of users' visual attention across interface entities. Eye movement data, when combined with data from other input modalities (such as spoken commands, haptic actions with the keyboard and the mouse, etc.), results in just such a rich data on set. However, integrating, analyzing and visualizing multimodal data on user interactions remains a difficult task. In this paper we report on a first step toward developing a suite of tools to facilitate this task. We designed and implemented an Eye Tracking Analysis System that generates combined gaze and action visualizations from eye movement data and interaction logs. This new visualization allows an experimenter to see the visual attention shifts of users interleaved with their actions on each screen of a multi-screen interface. A pilot experiment on comparing two interfaces — a traditional interface and a speech-controlled one — to an educational multimedia application was carried out to test the utility of our tool.

Models that are constructed within the bounds of a single paradigm are not sufficient for modeling all aspects of complex systems. Therefore, even though reasoning and simulation systems that utilize a single modeling paradigm are the current norm, we explore a multimodel approach in this paper. A multimodel approach is defined as one in which more than one model-each derived from a different perspective, and utilizing correspondingly distinct reasoning and simulation strategies-are employed. By describing four models which illustrate the use of different modeling techniques, we show how a multimodel approach can enrich the modeling environment and make it correspond better with real world information. Our models come from many sources-Systems and Simulation literature for the modeling of natural phenomena and artificial devices, and Artificial Intelligence and Cognitive Science for the modeling of human intuition and expertise in reasoning. Generalizing from these four models, we suggest that modeling complex systems may best be approached from an integrated architectural viewpoint which combines multiple modeling paradigms.>

The problem of modeling knowledge about the fault behavior of a system and utilizing this model for reasoning about and diagnosing failures is addressed. A solution that merges graph and fault-tree-based failure analysis with rule-oriented reasoning is presented. Failure analysis is divided into two phases, a failure source location phase and a failure cause identification phase. Each phase consists of a failure model and a process that operates on it. The failure models for the first and second phases are based on lesel-structured fault propagation digraphs and augmented fault trees, respectively. The augmented fault tree (AFT) is a conceptual structure that encodes probabilistic, temporal, and heuristic information in addition to the causal aspects of failures modeled by conventional fault trees. The two models are combined to form a novel hierarchical failure knowledge representation scheme. Upper levels of this hierarchy are made up of the fault propagation digraphs. Each level represents a view of the system under a particular granularity, and the granularity increases with levels. This feature permits control over the resolution of fault diagnosis. The lowest level consists of a set of cause-consequence knowledge bases containing production rules. These production rules are derived from augmented fault trees and represent the cause-effect relations among failure events that lead to the corresponding subsystem's failure. A knowledge acquisition procedure to generate these failure models and failure analysis processes that operate on them are described. The methodology proposed is inherently parallel as the processes may operate on different levels independently.