Quantitative and qualitative data from a functional magnetic resonance (fMRI) study of ideation in product design engineering




This study aimed to examine the brain regions activated during ideation in practising product design engineers, and to compare brain activation patterns for open-ended and constrained PDE ideation tasks. Brain activity was recorded using functional magnetic resonance imaging (fMRI).


A total of 32 designers participated in the study. Three participants were excluded from the analysis due to poor quality fMRI data. This dataset contains the data for the 29 participants included in the analysis.

Participants completed three types of task while undergoing fMRI scanning: creative (open-ended) ideation; innovative (constrained) ideation; and manipulate. In the ideation tasks, participants were asked to generate ideas for products to solve open-ended and constrained problems (up to three ideas per task, in a maximum of 85 seconds). In the manipulate tasks, they were asked to recall a known product from a given category, visualise it, and mentally rotate or resize a selected feature (up to three products per task, in a maximum of 30 seconds). Each participant completed 10 of each type of task (i.e. 30 tasks in total). In each task, participants indicated when they had generated an idea or formed an image by pushing a hand-held response button. Participants also completed twenty baseline tasks, where they responded each time a fixation cross presented on a black background changed from white to purple. The cross was presented for 30 seconds in total, and changed colour for 200 milliseconds at least three times. Colour changes were separated by intervals of 1 – 10 seconds.

The open-source Matlab toolbox Cogent (http://www.vislab.ucl.ac.uk/cogent.php) was used to present task descriptions to participants in the scanner, and to collect data on their response times. A Siemens 3T MRI scanner with a standard head coil was used to record T1-weighted anatomical and echoplanar T2*-weighted image volumes with BOLD contrast from each participant while they completed the above tasks.

After each ideation task, the participants were given 25 seconds to verbally summarise their ideas. After exiting the MRI scanner, they listened back to these recordings and produced a rough sketch of each idea they could recall. Each sketch was coded to identify what interpretation of the design problem the designer adopted (P-codes) and what type of solution they generated (S-codes). The P and S variables were then used to compute two measures of ideation performance: (i) breadth of exploration (BR); and (ii) solution novelty (SN).


The dataset is organised into folders and files as outlined below.

Folder 1 – Participant data and tasks:

“Task descriptions”: a subfolder containing .txt files with descriptions of all creative and innovative design ideation tasks (DT; ‘a’ denotes creative and ‘b’ denotes innovative), manipulate tasks (MT), imagine tasks (IT), and the baseline task (B) used in the study. The task ID numbers are used to refer to the tasks throughout the other files.

“Participant data”: a .csv file containing the age, gender, and years of design experience for each of the 29 participants, and the ID number used to refer to them throughout the other files.

Folder 2 – Scanning parameters and response data:

“fMRI scanning parameters”: a PDF file containing the MRI scanning parameters applied.

“Response data”: subfolder containing a .csv file for each participant that includes: (1) what tasks they completed in what order; and (2) response time data for each task. Each file is named with the participant ID.

“Key to response data variables”: a .csv file containing definitions of the variables included in each column of the “Response data” files for participants.

Folders 3, 4, and 5 – fMRI data:

Subfolders containing the raw fMRI data collected for each participant in DICOM format. Folders are named with the participant ID number.

Folder 6 – Sketches:

A set of .jpeg images of the sketches produced by each participant for each ideation task. Sketches are firstly organised into subfolders named with the participant ID number, and then further subfolders named with the ideation task order (i.e. Task 1, 2, … 20). Each image file name follows this convention: CAAA-B-C, where AAA is the participant ID number (e.g. 001), B is the ideation task order, and C is the idea number (e.g. 1st, 2nd, or 3rd idea).

Folder 7 – Sketch coding data:

“Coding schemes”: subfolder containing the schemes of P (problem) and S (solution) codes defined for sketches generated in each ideation task. Each set is in .csv format, and named with the ideation task ID number (DTXX) from the “Task descriptions” file.

“Sketch coding”: subfolder containing .csv files showing the P- and S-codes applied to each sketch produced by each participant for each ideation task. Separate .csv files are included for creative and innovative ideation tasks, and an additional file shows ‘other’ codes applied in cases where a sketch could not be coded with P and S. Throughout, sketches are denoted using the .jpeg file names from the “Sketches” folder above.

“SN and BR scores”: subfolder containing .csv files with the breadth of exploration (BR) and solution novelty (SN) scores for each sketch and/or participant. The variables (including definitions) and formulae used to calculate these scores are also included. There are separate .csv files for creative and innovative ideation task scores. Throughout, sketches are denoted using the .jpeg file names from the “Sketches” folder above.


All participant data in this dataset has been fully anonymised. All participants gave written informed consent prior to participation.


The dataset is owned by the University of Strathclyde.

Data embargo until 01/10/19
Date made available30 Aug 2019
PublisherUniversity of Strathclyde
Date of data production2017 - 2019

Cite this