Mindpainter: a system for enhancing children's emotional expression and creativity through brush generation

How can technology enhance children’s emotional expression and creativity through drawing? We proposes MindPainter, a machine learning-driven approach that generates children's paintbrushes based on emotional signals. In Study 1, we examine the relationship between six basic emotions and thirteen brush shapes, using emotional stimuli, galvanic skin response (GSR) data, and self-annotations from 51 children aged 9–11. Results show connections between brush attributes and four emotions (happiness, surprise, fear, and anger). Building on these findings, Study 2 explores emotion brushes using GSR data, employing a machine learning model trained on the YAAD dataset and real-time brush generation with p5.js. User feedback confirms the efficient differentiation of emotional parameters and satisfaction, which drives continued system usage for artistic creation and emotional expression and communication. This research introduces an interactive method of generating paintbrushes based on children's emotions, fostering emotional expression, creativity, and enhancing both emotional health and art education.⁣