Kehon hyödyntämisen mahdollisuudet luonnontieteiden oppimisessa

Abstract

In recent years, Finnish pupils’ interest in natural sciences education has been declining. For this study, new bodily methods were developed in which learning is based on the bodily learning experience and its reflection. In Finnish science education, the effect of learning methods that utilize bodily learning experiences has not been investigated. The purpose of this study is to explore the possibilities of utilizing the body in multidisciplinary science education and to find out how students experience bodily learning in natural sciences. In addition, the aim is to find out what new information machine learning–based artificial intelligence solutions can process based on the collected data. The thesis a multi-method study based on four research articles in which students’ (n = 611) experiences of utilizing the body in learning were investigated with questionnaires. The responses were analyzed using traditional statistical methods and unsupervised machine learning–based cluster analysis. The main result of the research is that new methods that utilize the body create a meaningful and memorable learning experience that helps concretize the abstract phenomenon. Bodily learning can take place in various learning environments, and variability as well as increased physical activity can increase students’ alertness and interest in instruction. Overall, more than 80% of students found the new methods more interesting compared to the traditional classroom work, which can increase students’ motivation and interest in science education. Research shows that the bodily methods could motivate students to learn, especially boys and weaker students, for whom learning abstract science concepts using traditional methods is challenging. Most students felt that learning and remembering phenomena are more effective when the body is utilized in the learning process. Examining the ability of machine learning–based artificial intelligence applications to process new information from survey data, it was found that unsupervised machine learning is a useful tool for identifying students with different learning profiles. Based on the theoretical framework of the work and the students’ experiences, a learning model that utilizes the body was created. This model can be utilized in the design of meaningful science education.