EEG signals respond differently to idea generation, idea evolution, and evaluation in a loosely controlled creativity experiment
Abstract
We investigated whether EEG signals are significantly different during three modes of thinking: idea generation, idea evolution, and evaluation in a loosely controlled creativity experiment. In the experiment, participants accomplished a modified figural Torrance Test of Creative Thinking (TTCT-F) while their EEG signals were recorded. During the idea generation, a participant was instructed to complete a sketch that immediately triggered by a sketch stimulus at first sight. During the idea evolution, a participant was instructed to complete a sketch that is radically distinctive from what was the immediately triggered by the sketch stimulus. During the evaluation, a participant was instructed to evaluate difficulties of thinking and drawing during the idea generation and evolution. It is expected that participants would use their past experience to intuitively complete a sketch during the idea generation while they could use more divergent and imaginative thinking to complete a possible creative sketch during the idea evolution. Since creativity is not a repeatable, reproducible, and controllable act leading to a possible creative sketch during either idea generation or idea evolution, the duration of each task was extended to three minutes in that participants could have enough time to freely explore and generate possible creative ideas. Such an experimental design is named as a loosely controlled creativity experiment, which offers an approach to studying creativity in an ecologically valid
manner.
The validity of the loosely controlled creativity experiment could be verified through comparing its findings on phenomena that have been effectively studied by validated experimental research. It was found from our experiment that alpha power decreased significantly from the rest to three modes of thinking. These findings are consistent with that from visual creativity research based on event-related (de)synchronization (ERD/ERS) and task-related power changes (TRP). Specifically, in the lower alpha band (8-10 Hz), the decreases of alpha power were significantly lower almost over the entire scalp during the idea evolution compared to the other modes of thinking. The finding indicated that the idea evolution requires less generation attention demands than the other two modes of thinking, since lower alpha ERD has been reported as being more likely to reflect general task demands such as attentional processes. In the upper alpha band (10-12 Hz), the decreases of alpha power were significantly higher over frontal-central, frontal-temporal, central, and central-parietal cortical sites during the evaluation compared to the other modes of thinking. The finding indicated that evaluation involves more task-specific demands, because upper alpha ERD has been found as being more likely to reflect task-specific demands such as memory and intelligence as was defined in the literature. In addition, new findings were obtained since the loosely controlled creativity experiment could activate multiple brain networks to accomplish the tasks involving the three modes of thinking. EEG microstate analysis was used to structure the unstructured EEG data to detect the activation of multiple brain networks. Combined EEG-fMRI and EEG source localization studies have indicated that EEG microstate classes are closely associated with resting-state network as
identified using fMRI. It was found that the default mode network occurred frequently during the idea evolution compared to the other two modes of thinking while the cognitive control network occurred frequently during the evaluation compared to the other two modes of thinking. The finding indicated that idea evolution might be more associated with unconscious and internal directed attention process. Taken together, the loosely controlled creativity experiment with the support of EEG microstate analysis appears to offer an effective approach to investigating the real-world complex creative activity.
Repository structure
Main directory
The directory of raw_data includes all of raw EEG data from 29 subjects, while the directory of clean_data consists of clean EEG data after removing noise and artifacts.
The directory of alpha_power_changes includes the results of alpha power changes, while the directory of microstate_maps consists of individual EEG microstate classes, condition-wise EEG microstate classes, and global EEG microstate classes.
The directory of results includes of behavioral results and EEG microstate analysis results.
The directory of experiment_config includes the makers of all of the subjects and experimental names.
The directory of code includes python codes, which are responsible for data preprocessing, alpha power changes, and EEG microstates.
Support
For questions on the dataset or the code, contact zeng@ciise.concordia.ca
Licensing
This work is licensed under a Creative Commons Attribution 4.0 International License.
