Thibaud Latour

COPSE: Rapidly Instantiating Problem Solving Activities based on Tangible Tabletop Interfaces

Collaborative problem solving is a skill that has become very important in our everyday lives and is constantly gaining attention in educational settings. In this paper, we present COPSE: a novel and unique software framework for instantiating Microworlds as collaborative problem solving activities on tangible tabletop interfaces. The framework provides three types of building blocks: widgets (provide input and localized feedback), equations (define the model), and scenes (visualize feedback), which can be specified in the form of structured text. Aim of COPSE is to simplify processes of creating, adjusting, and reusing custom Microworlds scenarios. We describe the structure of the framework, provide an example of a scenario, and report on a case study where we have used COPSE together with 33 teachers to build new scenarios on the fly.

The Genetics Lab: An Innovative Tool for Assessment of Intelligence by Mean of Complex Problem Solving

The purpose of this paper is to present the important characteristics of the intelligence measurement tool called “The Genetics Lab”. This tool, based on web technologies, allows one to assess general intelligence using complex simulations. It has been developed and evaluated to measure cognitive skills of students in Luxembourg. Behind the tool itself, we propose a generic and clear architecture that can be used as groundwork for other evaluation solutions. Previous tools have been affected by various technical weaknesses. The Genetics Lab brings a major contribution by providing a clear architecture and an efficient implementation to address these common issues. To achieve this, we will explore in depth the main fields of e-assessment such as traces, instructions, scoring, localized and multilingual content. The proposed tool is reusable and highly adaptive. It also makes data collection resulting from the user’s test a flawless and one-step process.

Semi-automatic Generation of Competency Self-assessments for Performance Appraisal

Competency self-assessment for Performance Appraisal is receiving increasing attention from both researchers and practitioners. Nevertheless, the accuracy and supposed legitimacy of this type of assessment is still an issue. In the context of an industrial use case, we aim to develop and validate a computer-based competency self-assessment technology able to import any type of competency document (for performance appraisal, training need identification, career guidance) and generate semi-automatically self-assessment items. Following the model of Appraisal Effectiveness designed by Levy and Williams, our goal was to build an effective tool meaning that several perspectives must be taken into account: psychometric, cognitive, psychological, political and the reaction’s perspective. In this paper, we will only focus on one specific psychometric property (interrater reliability between an employee and its supervisor). According to a specific rating process and format, our Cross Skill™ technology showed promising results related to interrater reliability in a use case with bank officers and their supervisor.

Designing a Collaborative Problem Solving Task in the Context of Urban Planning

The construct to assess collaborative complex problem solving has two dimensions: the collaboration and the complex problem solving construct. Both have been defined in the past in the literature, but unfortunately no common model exists. In addition, current assessments lack of authentic tasks which enforce both face-two-face collaboration while solving a complex task. The paper presents a scenario where a task was designed to offer best conditions for assessing collaborative complex problem solving. The principal idea is that the actors play a certain role with specific objectives and different constraints. In addition, different implementations of feedback cues are provided.