In previous studies, we developed various learning support systems for engineering students. These systems include those for learning university lecture content, such as Intellectual Property (IP) law, statistics, and collaborative chemistry experiments. Additionally, we created systems for enhancing the learning environment, such as a Virtual Reality (VR) content creation system for lectures and a system that allows a single teacher to monitor and instruct a large number of students on their programming status online. Previous research has shown that the effectiveness of these systems varies depending on student’s learning style. Our earlier studies identified that the "Visual-Verbal" and "Active-Reflective" dimensions of Felder's learning style are closely related to the use of IP law learning support systems. Furthermore, another study suggested that the "Brain-Physical (which activity is preferred as the primary one)" and "PC-Smartphone (which is easier to input)" dimensions, could be optimize various learning support systems.

Building on these findings, this study analyzed which learning support systems students prefer based on their classification within the four dimensions of the learning style model. Each dimension was divided into + and - cells, resulting in 16 combinations (2^4). For convenience, in the Active-Reflective dimension, Active is denoted as -, and Reflective as +; similarly, in the Physical-Brain dimension, Physical is +, and Brain is -.A learning style survey was administered to 121 students to determine their classification within the four-dimensional model. Subsequently, we analyzed the relationship between the survey results on students' interests, usage of learning support systems, and preferred input devices, with their classification in the four-dimensional model.

The results indicated, for example, that students in the Active-Visual-Brain-PC cell are faster at reading horizontal text, more proficient with PCs, better at typing with a mouse, and appreciative of visual content. These traits helped reveal the characteristics of each cell. Based on these findings, we can now provide more effective learning support systems tailored to the four-dimensional learning style model. Comparing data from the past four years, we found an increasing number of students proficient in smartphone input, suggesting that learning support systems are less effective if they are not smartphone-compatible.. Forced online education during the COVID-19, beginning in 2020, seems to have increased students' dependence on smartphones. Students likely use smartphones more frequently than PCs because their accessibility and easy of communication. Therefore, it is recommended that future learning support systems be designed to be smartphone compatible.

Keywords: Learning Support system, Learning Style Model, System Development.