Developing Emotional Expression in Painting Through Guided Pedagogies

Abstract

This study demonstrates a method of integrating machine learning (ML) with pedagogies to enhance emotion expression in painting. Much traditional art education focuses on skill and technique, leaving less emphasis on emotional expression, where most beginner artists are interested. This research uses an analysis of emotional content in painting with a CNN model so that participants receive personalized feedback in real-time, potentially filling the gap between emotion and artistic output. In total, two groups were observed during the experiment. These are the traditional art educational input group and the experimental machine learning-guided input group. The result from this was that the improvement in emotional expression was very clear; hence, emotional scores from the experimental group changed from 0.47 to 0.75 against 0.45-0.54 of the control group. Additionally, 85% of participants in the experimental group felt more confident about expressing their emotions through their paintings, and their emotional expression was still stronger without further guidance. This study implies that incorporating machine learning into art education may help improve emotional depth and artistic confidence, providing a tool for artists to express themselves better in visual form. This approach offers a novel framework for rethinking art pedagogy, with technology creatively merged to foster deeper emotional connections within artistic practices.