Kazuhisa Miwa

A cognitive study of information seeking processes in the WWW: the effects of searcher's knowledge and experience

We investigated, through a cognitive psychological experiment and its protocol analysis, human cognitive processes of seeking information on the WWW and the effects of a subjects knowledge and experience on the information seeking processes and performance. In our experiment, the subjects were divided into two groups: one comprising expert subjects and the other novice subjects. All subjects were given a general search task and a specific search task. In the experimental results, except for one exceptional subject, we could confirm significant differences between the two groups in the solution time, the number of pages searched, and the kind of pages accessed. We also propose a behavioral schema for tracing a subjects searching processes. The behavioral schema consists of four behavior levels on the WWW: search, results-of-search, page-following results, and page-following pages. Each subjects behavior was described as a transition of nodes, each representing the subjects behavioral state, and six kinds of operator connecting two nodes: search, link, return, jump, browse, and next among the four behavior levels. The results of an analysis using the schema showed distinctive subject behavior such as a breadth-first search or depth-first search.

Experimental study of learning support through examples in mathematical problem posing

When using mathematics to solve problems in everyday life, problem solvers must recognize and formulate problems by themselves because structured problems are not provided. Therefore, in general education, fostering learner problem posing is an important task. Because novice learners have difficulty in composing mathematical structures (solutions) in problem posing, learning support to improve the composition of solutions is required. Although learning by solving examples is adopted in general education, it may not be sufficiently effective in fostering learner problem posing because cognitive skills differ between problem solving and problem posing. This study discusses and experimentally investigates the effects of learning from examples on composing solutions when problem posing. We studied three learning activities: learning by solving an example, learning by reproducing an example, and learning by evaluating an example. In our experiment, undergraduates were asked to pose their own new, unique problems from a base problem initially presented after the students learned an example by solving, reproducing, or evaluating it. The example allowed the undergraduates to gain ideas for composing a novel solution. The results indicated that learning by reproducing the example was the most effective in fostering the composition of solutions.

Learning through intermediate problems in creating cognitive models

Cognitive modelling is one of the representative research methods in cognitive science. It is believed that creating cognitive models promotes learners’ meta-cognitive activities such as self-monitoring and reflecting on their own cognitive processing. Preceding studies have confirmed that such meta-cognitive activities actually promote learning effects. However, there are some difficulties in bringing about learning by creating cognitive models in an educational context. To overcome the difficulties, we propose an innovative learning design, ‘learning through intermediate problems’ and also developed a web-based production system called DoCoPro that can be used anywhere and anytime in an environment connected to the Internet. We performed three introductory cognitive science classes in which the participants learned cognitive modelling and constructed running computer models using our system. In the first and second classes, the participants were required to construct production system models that solve pulley problems. They also posed their original pulley problems that their own models were subsequently able to solve. These generated problems were distributed to the other members. The participants were able to find incompleteness in their cognitive models, revise them to remove the incompleteness, and improve their models while solving the given problems. The participants, by successfully creating sophisticated models, acquired a deeper knowledge of the learning domain. The class practices confirmed the utility of ‘learning through intermediate problems’ when constructing an educational environment for learning creating cognitive models. In the third class, the participants constructed cognitive models solving addition and subtraction problems using DoCoPro. The cognitive processing underlying such problem solving is automated, therefore it may be difficult to verbalize and externalize such cognitive processes. The post-questionnaire showed evidence that the participants actually performed meta-cognitive activities while monitoring their own internal information processing.

Evaluation of a system that generates word problems through interactions with a user

In mathematical learning, it is important to give learners a number of problems that have various features in both surface problem situations and deep mathematical solution structures. In this study, we implement a system that generates various word problems by using episodes, which are knowledge regarded as cases of problem generation. Our system interacts with a teacher as a user to acquire the common knowledge needed to generate word problems. We performed experimental evaluations to verify problem generation by our system, with the results indicating that our system can successfully expand the variety of problems from the initial ones stored in the system. We also found that our system needs interactions with a knowledgeable user because novice users cannot necessarily provide the system with effective knowledge.

Development of a Usability Questionnaire for Automation Systems

In this study, we positioned automation systems as the third-generation artifacts and developed a generalized usability questionnaire with 18 questions for automation systems as daily used artifacts. This questionnaire could be used to evaluate various types of automation systems and is useful for the development and improvement of automation systems as artifacts used in our everyday life.

Learning Mental Models of Human Cognitive Processing by Creating Cognitive Models

We investigated how creating cognitive models enhances learners’ construction of mental models on human cognitive information processing. Two class practices for undergraduates and graduates were performed, in which participants were required to construct a computational running model of solving subtraction problems and then develop a bug model that simulated students’ arithmetic errors. Analyses showed that by creating cognitive models, participants learned to identify buggy procedures that produce systematic errors and predict expected erroneous answers by mentally simulating the mental model. The limitation is that this benefit of creating cognitive models was observed only in participants who successfully programmed a computational model.

Use of a Cognitive Simulator to Enhance Students' Mental Simulation Activities

We developed a cognitive simulator of the dual storage model of the human memory system that simulates the serial position effect of a traditional memory recall experiment. In a cognitive science class, participants learned cognitive information processing while observing the memory processes visualized by the simulator. Through the practice, we confirmed that participants learned to predict experimental results in assumed situations implying that participants successfully constructed a mental model and performed mental simulations while running the mental model in various settings. We discuss the possibility that a cognitive model can be used as a learning tool and, more specifically, as a mediator tool connecting theory and empirical data.

Supporting Mathematical Problem Posing with a System for Learning Generation Processes through Examples

Problem posing, by which learners create new problems by themselves, is an important activity in mathematics education. However, novice learners have difficulty in posing problems, particularly when formulating appropriate solution structures of problems. Although they are provided with example problems that can serve as hints for composing novel problems, they do not necessarily understand the key ideas used to generate the examples. To improve problem posing for novices, this study discusses an approach that supports learning from examples as a production task. We propose a method of learning from examples through imitation, where a learner reproduces problems identical to given examples. We implement a system that presents examples of problem posing and supports learners in understanding the examples by having the learners reproduce them. We conducted an experimental evaluation in which learners learned from an example that embeds useful ideas to alter solution structures in the system. The results demonstrated that the learners successfully adapted the example when posing their own problems if they learned the example by the reproduction method. Thus, learning from examples through reproduction appears to be effective in the domain of problem posing as a production task.

Experimental investigation of misuse and disuse in using automation system

In this study, we experimentally investigated human use of automation systems and the selection strategies of such usage. We used two different types of tracking tasks. As a result, we found that the participants neither tended to misuse nor disuse the automation system. Also, we confirmed that they tended to select to use the automation system depending on their manual performance rather than the system performance. Moreover, we found that there is a relationship between the tendency to use the automation system and the selection strategy.

Source orientation in communication with a conversational agent

Do people directly interact with a computer as an independent social actor during simple conversations, or do they orient to the unseen programmer behind the program or an imagined person operating from another room? In the former case the CAS (Computers As Source) model would be valid, while in the latter it would be the CAM (Computers As Media) model that describes the interaction [1]. This study specifically attempts to address the following research questions: (a) Whether the CAM model adequately describes bi-directional communication, and (b) Whether the CAS model adequately describes one-directional communication when the representation of the agent is unfamiliar to the user. Two factors were thus investigated: (1) directions of communication (bi-directional vs. one-directional), and (2) familiarity of the agent’s representation to the user (familiar vs. unfamiliar).