Abstract

Design is a recursive process to create original and useful artifacts to meet human needs. Creativity is a critical ability for designers to create original and useful artifacts. Design creativity focuses on creativity mechanisms during the design process. Even though many studies endeavor to understand cognition in design creativity by protocol analysis, neurocognitive underlying design creativity remains unknown. Such scarcity can be explained by the intrinsic characteristics of the design process and the data analysis approaches. Loosely controlled experiments are capable of simulating the intrinsic characteristics of the design process, whereas they would increase difficulties of data analysis due to unstructured data and hidden causal relationships between a stimulus and its response. Therefore, this research aims to: (i) test the effectiveness of loosely controlled experiments through comparing its findings on phenomena that have been effectively studied by validated experimental research; (ii) structure and segment unstructured electroencephalography (EEG) signals through EEG microstate analysis; (iii) identify EEG-defined large scale brain networks and uncover their temporal dynamics in design creativity; (iv) capture the temporal dynamics of EEG defined large-scale brain networks to classify different cognitive activities in design creativity through RNN techniques under the autoencoder framework. The loosely controlled experiments supported by EEG microstate analysis appear to offer an effective approach to facilitating an ecologically valid neurocognitive study.