Hitoshi Terai

An utterance attitude model in human-agent communication: from good turn-taking to better human-agent understanding

In this study, we discuss a novel expression and comprehension model of the utterance attitude of speaking/hearing during conversations. Humans who participate in conversation display these implicit and explicit attitudes, and use them to understand the other participants in advance of turn-taking. We design abstract animated agents that mimic human turntaking in conversations to confirm the validity of our model. The subjective evaluation tests show that the expressions of the agents are understandable. The model may facilitate turn-taking in human-agent interaction.

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.

サーチエンジン検索結果ページにおける視線情報の分析( 第17回(2009年度)年次大会(研究報告会&総会))

*1 物質・材料研究機構科学情報室 National Institute for Materials Science E-mail: TAKAKU.Masao@nims.go.jp 2 国立教育政策研究所教育研究情報センター National Institute for Educational Policy Research E-mail: yuka@nier.go.jp 3 東京電機大学情報環境学部情報環境学科 Tokyo Denki University E-mail: terai@sie.dendai.ac.jp 4 愛知教育大学教育学部 Aichi University of Education E-mail: hsaito@auecc.aichi-edu.ac.jp 5 放送大学 ICT活用・遠隔教育センター The Open University of Japan E-mail: miwamaki@code.u-air.ac.jp 6 国立情報学研究所情報社会相関研究系 /総合研究大学院大学 National Institute of Informatics / Graduate University for Advanced Studies E-mail: kando@nii.ac.jp

Analyzing driver gaze behavior and consistency of decision making during automated driving

We investigate a possible method for detecting a drivers negative adaptation to an automated driving system by analyzing consistency of driver decision making and driver gaze behavior during automated driving. We focus on an automated driving system equivalent to Level 2 automation per the NHTSAs definition. At this level of automation, drivers must be ready to take control of the vehicle in critical situations by monitoring the driving environment and vehicle behavior. Since drivers are not required to operate the pedals or steering wheel during automated driving, a drivers negative adaptation to an automated system needs to be detected from behavior other than vehicle operation. In this study, we focus on driver gaze behavior. We conduct a simulator study to compare the gaze behavior of fifteen drivers during conventional and automated driving. We also analyze the consistency of driver decision making when changing lanes during conventional and automated driving. Experimental results show that drivers who pay less attention to the road ahead during automated driving tend to be less sensitive to risk factors in the surrounding environment and also tend to make inconsistent lane change decisions during automated driving.

Connecting Qualitative and Quantitative Analysis of Web Search Process: Analysis Using Search Units

Our final goal is to understand exploratory searches as four levels of search processes: search task, intent unit, search unit, and link unit. To complete these objectives, we used qualitative data to categorize participants’ information needs for search units and quantitatively analyzed whether differences in the information needs of search units influence users’ search processes and how task types and groups affect search units. In the experiment, eleven undergraduates and five graduates conducted information gathering task for writing a report and trip planning. We recorded their verbal protocols during the tasks and post interviews, browser logs, screen captured video, and eye-tracking data. We divided the process of exploratory searches into search units. Then search units were classified into the two types of information needs, navigational and informational, based on qualitative data. We conducted a quantitative analysis to compare between tasks and groups and types of search units. The results showed that there were many differences between the information and navigation search units.

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.

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.

A Chance Favors a Prepared Mind: Chance Discovery from Cognitive Psychology

A chance has two contrary aspects: suddenness as an accidental finding and gradualness as the result of a prepared mind. Such duality of chance discovery resembles the insight process treated by problem solving researches. In this paper, we focus on the insight process in human problem solving, present a broad overview of its suddenness and the gradualness, and introduce our experimental results from the viewpoint of the duality of insight. We believe that our research findings will contribute to studies of chance discovery.

An experimental study on collaborative scientific activities with an actual/imaginary partner

In this study, we experimentally investigate collaborative scientific activities that are undertaken through a virtual space such as the Internet. In such cases, a partner has two aspects: an imaginary partner with whom the problem solver seems to work together, and an actual partner with whom he/she actually works. We design an experimental environment in which we can control the two factors independently. The experimental result shows: (1) a bias appearing in human behavior, such as the positive test bias in hypothesis testing, was not influenced by the change of an actual partner, however (2) the degree of using information given by a partner, such as reference to a partners hypothesis, varied considerably with the change of an actual partner. Neither phenomenon above depended on the type of imaginary partner.