Learning social skills is an important part of the socialisation process of children, which should occur at school, at home and in any place where children live. There are very few studies on social interaction and collaboration roles with 3–4 year old. In this paper, we aim to understand collaboration in young children to help them develop their social skills and improve their overall development. To get this, we have designed an observational experiment to monitor and characterise group activity and roles, mediated by technology and using data mining techniques. First, we have designed a game as a free-play situation where the conditions require interplay of three children with toys and among interaction among peers. Children interacts with game through tangible toys. The environment collects accurate data on children’s actions automatically and non-intrusively. We also consider other data from direct observation by psychologists and educators. Then, we have organised a study for groups(triads) of 3 to 4-year-old children playing with this game. We analyse data from 81 children (51.9% boys and 48.1% girls) in groups of three randomly selected. The work proposes a set of actions in the game and from them a set of indicators, which are used as intermediate measures of observation to analyse the playing process. Social interaction is characterised in 5 levels: Coordination, Cooperation, Collaboration, Troubled and Unproductive; and five roles: Saboteur, Missing, Explorer, Actor, Collaborator and Director. We found that children interact socially, engage in play, help each other and mostly reach the level of collaboration. There are minority cases of non-cooperation (Troubled or Unproductive), with conflict situations or trial and error solving processes, which cause the task to last a long time before it is finally finished. We have also found that children can adopt different roles in the group. Occasionally, there are cases of children who act as conductors, organising the work while others follow. This work provides a contribution for teachers and educators in the preschool classroom, relating to social interaction. It tells us that young children can play in randomly organised groups, collaborate and take roles, help their peers and learn to do so in classroom games. It shows a learning situation that can serve as an example of how to conduct technology-mediated activities in the classroom to promote social interaction among children in an effective, engaging and motivating way.

Concept maps are commonly used as knowledge representation mechanisms to classify a given domain in the scope of a constructivist knowledge building process. In this paper we carry out social constructivist learning experiences using concept maps as a collaborative knowledge-sharing tool with applications in the coordination of virtual communities for learning. In our experiment, the community studies an environmental area-theoretically and in-situ, collects data from a variety of experiments and makes a representation of it. Then, they build a common concept map representing their understanding of this area using a set of tools developed ad-hoc and integrated by a metamodel wrapping approach. Finally, they complete it linking their node with the collected information as a way to organize and to navigate through the different pieces of knowledge that they have been accumulating during several fieldtrips.

The purpose of the paper is to describe TRAZO, a system aimed at the acquisition of handwriting skills using tablet-PC devices. This tool focuses on the pre-writing phase and is aimed at 3-year-old children, who practice different strokes following a sequence defined by the teacher. They usually draw straight lines first, then curves, and then move on to a combination of both. The system evaluates the exercises automatically, maintains a user model and includes a monitoring tool to show the learning processes graphically. The aim is to achieve hand coordination, visual perception; to learn to grip the pencil properly, and to practice the direction and pressure of the pencil (or any other object used to write).

The SIETTE collaborative environment allows students to take tests together in small groups. Individuals can communicate and hold discussions to solve test questions. This paper describes the summative evaluation of the SIETTE collaborative environment and the relationship between the collaboration and the actual improvement which occurs in the student knowledge level. We conclude that under certain conditions improvement occurs and can be measured.

In this paper we define some indicators that are useful to analyze the behavior of an individual within a learning community and hypothesize that there must exist some underlying factors relating to the user that explain the result obtained. We analyze the learner behaviors in the Comtella-D learning community and conclude that the actual data do not contradict our hypothesis.

This paper presents some experiences of learning with computers in a preschool setting. The MEDEA framework (a Spanish acronym of Methodology and Tools for the Development of Intelligent Environments of Teaching and Learning) is used to define the activities and construct the user models. The result is an environment that has been installed in the classroom as a new resource and which has been used in conjunction with traditional learning tasks. The system has been successfully evaluated with children from different countries and cultures.