0
Research Papers: Design Theory and Methodology

The Relationship Between Design Outcomes and Mental States During Ideation

[+] Author and Article Information
Wan-Lin Hu

School of Mechanical Engineering,
Purdue University,
West Lafayette, IN 47907
e-mail: hu188@purdue.edu

Joran W. Booth

School of Mechanical Engineering,
Purdue University,
West Lafayette, IN 47907
e-mail: boothj@purdue.edu

Tahira Reid

School of Mechanical Engineering,
Purdue University,
West Lafayette, IN 47907
e-mail: tahira@purdue.edu

1Corresponding author.

Contributed by the Design Theory and Methodology Committee of ASME for publication in the JOURNAL OF MECHANICAL DESIGN. Manuscript received September 1, 2016; final manuscript received February 8, 2017; published online March 21, 2017. Assoc. Editor: Harrison M. Kim.

J. Mech. Des 139(5), 051101 (Mar 21, 2017) (16 pages) Paper No: MD-16-1612; doi: 10.1115/1.4036131 History: Received September 01, 2016; Revised February 08, 2017

Using electroencephalography (EEG) to predict design outcomes could be used in many applications as it facilitates the correlation of engagement and cognitive workload with ideation effectiveness. It also establishes a basis for the connection between EEG measurements and common constructs in engineering design research. In this paper, we propose a support vector machine (SVM)-based prediction model for design outcomes using EEG metrics and some demographic factors as predictors. We trained and validated the model with more than 100 concepts, and then evaluated the relationship between EEG data and concept-level measures of novelty, quality, and elaboration. The results characterize the combination of engagement and workload that is correlated with good design outcomes. Findings also suggest that EEG technologies can be used to partially replace or augment traditional ideation metrics and to improve the efficacy of ideation research.

Copyright © 2017 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 1

Experimental setup for discovering the effect of activities on reducing the inhibition to sketch during concept generation

Grahic Jump Location
Fig. 2

A participant wearing the EEG headset at the workstation used to conduct the study

Grahic Jump Location
Fig. 3

The Advanced Brain Monitoring (ABM) cognitive workload metrics describe the workload probability, where higher probability reflects higher workload. This figure qualitatively shows the relative relationship between tasks with different difficulties. Extreme high workload indicates confusion or struggle to process information as presented. Manageable workload is relatively sustainable and optimal for learning.

Grahic Jump Location
Fig. 4

The ABM engagement metrics measure and classify the cognitive state on a continuum from fatigued and inattentive to highly engaged and processing visual stimuli. This figure qualitatively shows the relative relationship between each metric. The metrics describe the probability of a specific moment being classified as a certain state (e.g., high engagement).

Grahic Jump Location
Fig. 5

The decision tree used to determine the novelty of each concept. The tree was adapted from Kershaw et al. [62].

Grahic Jump Location
Fig. 6

The decision tree used to determine the quality of each concept. The tree was adapted from Kershaw et al. [62].

Grahic Jump Location
Fig. 7

The degree of elaboration for each concept was based on a count of “chunks” embedded in it. On the left, concept AK06 shows a motorized wheelchair, which counts as one discrete chunk. On the right, concept AK07 shows an actuated knee brace. The knee brace counts as one chunk, and the actuator counts as another, giving this concept an elaboration score of 2.

Grahic Jump Location
Fig. 8

The definition of concept intervals for mapping EEG outputs to each concept

Grahic Jump Location
Fig. 9

Confusion matrices that summarize the results of testing the SVM models for furtherprediction: (a) novelty (accuracy = 74%), (b) quality (accuracy = 70.7%), and (c) elaboration (accuracy = 64.2%)

Grahic Jump Location
Fig. 10

Predicted concept novelty for ME undergraduate students. The x-, y-, and z-axes represent low engagement (passive attention), high engagement (active attention), and BDS Workload (mental manipulation), respectively. The left to right columns represent low (0–0.3), middle (0.3–0.6), and high (0.6–0.9) distraction. The top to bottom rows represent mid (0.5–0.67), mid-high (0.67–0.83), and high (0.83–1) FBDS Workload (working memory).

Grahic Jump Location
Fig. 11

Predicted concept novelty for non-ME undergraduate students. The x-, y-, and z-axes represent low engagement (passive attention), high engagement (active attention), and BDS Workload (mental manipulation), respectively. The left to right columns represent low (0–0.3), middle (0.3–0.6), and high (0.6–0.9) distraction. The top to bottom rows represent mid (0.5–0.67), mid-high (0.67–0.83), and high (0.83–1) FBDS Workload (working memory).

Grahic Jump Location
Fig. 12

Predicted concept novelty for ME graduate students. The x-, y-, and z-axes represent low engagement (passive attention), high engagement (active attention), and BDS Workload (mental manipulation), respectively. The left to right columns represent low (0–0.3), middle (0.3–0.6), and high (0.6–0.9) distraction. The top to bottom rows represent mid (0.5–0.67), mid-high (0.67–0.83), and high (0.83–1) FBDS Workload (working memory).

Grahic Jump Location
Fig. 13

Predicted concept novelty for non-ME graduate students. The x-, y-, and z-axes represent low engagement (passive attention), high engagement (active attention), and BDS Workload (mental manipulation), respectively. The left to right columns represent low (0–0.3), middle (0.3–0.6), and high (0.6–0.9) distraction. The top to bottom rows represent mid (0.5–0.67), mid-high (0.67–0.83), and high (0.83–1) FBDS Workload (working memory).

Grahic Jump Location
Fig. 14

Predicted concept quality for ME students. The x-, y-, and z-axes represent low engagement (passive attention), high engagement (active attention), and BDS Workload (mental manipulation), respectively. The left to right columns represent short (0–300 s), middle (300–600 s), and long (600–900 s) TimeSpan. The top to bottom rows represent mid (0.5–0.67), mid-high (0.67–0.83), and high (0.83–1) FBDS Workload (working memory).

Grahic Jump Location
Fig. 15

Predicted concept quality for non-ME students. The x-, y-, and z-axes represent low engagement (passive attention), high engagement (active attention), and BDS Workload (mental manipulation), respectively. The left to right columns represent short (0–300 s), middle (300–600 s), and long (600–900 s) TimeSpan. The top to bottom rows represent mid (0.5–0.67), mid-high (0.67–0.83), and high (0.83–1) FBDS Workload (working memory).

Grahic Jump Location
Fig. 16

Predicted concept elaboration for graduate students. The x-, y-, and z-axes represent low engagement (passive attention), high engagement (active attention), and BDS Workload (mental manipulation), respectively. The left to right columns represent short (0–300 s), middle (300–600 s), and long (600–900 s) TimeSpan. The rows indicate the experience of being injured; top: without prior leg injury; bottom: with prior leg injury.

Grahic Jump Location
Fig. 17

Predicted concept elaboration for undergraduate students. The x-, y-, and z-axes represent low engagement (passive attention), high engagement (active attention), and BDS Workload (mental manipulation), respectively. The left to right columns represent short (0–300 s), middle (300–600 s), and long (600–900 s) TimeSpan. The rows indicate the experience of being injured; top: without prior leg injury; bottom: with prior leg injury.

Grahic Jump Location
Fig. 18

An example prediction result of concept novelty with three 2D projection views

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In