Emotional Style Dimensions in Human Capital with Mind Training

We hypothesized that a mindfulness-based training program designed for engineering graduate students would improve emotional well-being and, secondarily, enhance research capacity. An initial pilot study was conducted at a single institution (Phase 1), followed by a larger study conducted at both the original and a second institution (Phase 2) to gather additional data and show the program’s transferability. The program comprised eight weekly mindfulness training sessions. Individuals in the study were randomly assigned to either an intervention group or wait-list control group. We administered pre- and post-test surveys with quantitative measures designed to assess emotional and physical well-being, as well as creativity, research satisfaction, and desire to contribute to the betterment of society. Participants also completed a summative survey to evaluate the impact of the program on their well-being and research. Analysis revealed statistically significant findings: improved emotional health, decreased neuroticism, increased positive affect, decreased negative affect, and increased mindfulness in the intervention groups compared to the control groups. Intervention groups in Phase 2 also reported statistically significant improvement in satisfaction with their research. Our findings suggest that mindfulness training has the potential to play a vital professional and personal development role in graduate engineering education.

Education and Social/Behavioral Science Institutional Review Board (IRB) at the University of Wisconsin-Madison (ID#2016–0917 for Phase 1 and ID#2018–1055 for Phase 2).

2019-08-01

2020-05-15

Emotional well-being, operationalized through the following validated measures administered pre- and post-intervention: (1) overall emotional health and its six subscales via the 48-item Emotional Style Questionnaire (ESQ: Attention, Self-Awareness, Resilience, Outlook, Social Intuition, Sensitivity to Context); (2) Neuroticism via the Ten Item Personality Inventory (TIPI); (3) positive and negative affect via the PANAS; and (4) trait mindfulness via the Mindful Attention and Awareness Scale (MAAS) and the 24-item Five Facet Mindfulness Questionnaire–Short Form (FFMQ-SF).

The ESQ overall score is the average of all 48 items (scored 1–7) and serves as an indicator of general emotional health; six subscale scores are each computed as the mean of their eight constituent items. PANAS Positive and Negative Affect are the means of their respective 10 emotion-word items (scored 1–5). MAAS is the mean of 15 items (scored 1–6). FFMQ-SF yields an overall score plus five subscale scores (Observe, Describe, Act with Awareness, Non-judge, Non-react). Neuroticism is the relevant two-item TIPI dimension. Group differences are assessed via Time × Group interactions in repeated-measures ANOVAs, with Cohen's d effect sizes.

Research satisfaction and research capacity, measured by an eight-item Research Satisfaction Scale (perceptions of progress, creativity/innovation, and persistence through obstacles). Secondary outcomes also include physical well-being (Cohen-Hoberman Inventory of Physical Symptoms, CHIPS) and desire to contribute to the betterment of society (a five-item Contributive Desire scale).

Research Satisfaction is the mean of eight items (1–7). Contributive Desire is the mean of five items (1–7). CHIPS is a count/severity index across 39 common physical symptoms. All secondary outcomes were analyzed with the same repeated-measures ANOVA approach as the primary outcomes. Qualitative secondary data came from open-ended summative-survey questions on the training's impact on professional work and personal life.

Randomized, two-arm waitlist-controlled design. Participants were randomly assigned to either an intervention group, which received the eight-week mindfulness training immediately, or a wait-list control group, which received the same training after a delay. Both groups completed pre-test and post-test surveys at the same points in the academic calendar (beginning and end of the fall semester), so that between-group differences could be attributed to the training rather than to seasonal or academic-stress variation. The study ran in two phases across three cohort-years: Phase 1 (one cohort at the University of Wisconsin-Madison) and Phase 2 (two cohorts at the University of Wisconsin-Madison and University of Virginia).

The intervention was the "Cultivating Transformative Research Through Mindfulness" (Mindful Engineer) curriculum: a skill-based, eight-week program of weekly instructor-led sessions (50–60 minutes each) covering the six dimensions of emotional style, brain neuroplasticity, mindfulness meditation and contemplative/cognitive practices, and strategies for healthy mental habits. Participants were encouraged to practice roughly 10 minutes daily; reported practice averaged about an hour to an hour-and-a-quarter per week. Small practice teams provided peer support between sessions.

Phase 1 was conducted at the University of Wisconsin-Madison, with the intervention group enrolled in Fall 2016 and the control group in Spring 2017. Phase 2 ran in parallel at the University of Wisconsin-Madison and the University of Virginia across two cohorts: Year 1 intervention in Fall 2018 / control Spring 2019, and Year 2 intervention in Fall 2019 / control Spring 2020. Instruction at Wisconsin was delivered by two experienced trainers across both phases; the Virginia site in Phase 2 was led by a separate experienced instructor who received in-person curriculum training and a detailed manual to ensure consistency across sites. All pre- and post-tests were collected in the fall terms, before the COVID-19 pandemic; the Year 2 control group's training was interrupted after the fifth in-person session and completed remotely after a two-week delay and IRB approval of the modification.

Eligible participants were M.S. and Ph.D. students engaged in research in engineering (and computer science in Phase 2) at two universities classified as research-doctoral with very high research activity. Students with a history of schizophrenia-spectrum, bipolar, or other psychotic disorders were excluded. Because sample sizes per university were relatively low, Phase 2 data from the two institutions were combined and analyzed together; this was justified by the absence of significant Group × Time × University interactions, and no significant differences were found between institutions. Surveys were administered via Qualtrics and included the ESQ, TIPI, PANAS, CHIPS, MAAS, FFMQ-SF, and custom research-satisfaction and contributive-desire scales, plus open-ended summative surveys. Analyses used repeated-measures ANOVAs (Time × Group), with Cohen's d effect sizes; a final summative survey was collected from Phase 2 Year 1 participants 18–24 months post-training.

Randomization was done digitally by a computer.

The individual participant (graduate student) was the unit of randomization; there was a single level of randomization.

No

Not clustered; the unit of randomization is the individual participant. Number of clusters equals number of observations (215 participants).

215 participants (individuals) analyzed across both phases: 58 in Phase 1 and 157 in Phase 2. (Phase 1 initial enrollment was 73 consented participants; the study did not set a separate a priori enrollment target.)

Phase 1: 24 intervention, 34 waitlist control (58 total).
Phase 2 (Year 1 and Year 2 combined): 73 intervention, 84 waitlist control (157 total).
Overall: 97 intervention, 118 waitlist control (215 total).

Randomization was at the individual level within each university. The two sites (University of Wisconsin-Madison and University of Virginia) serve as strata rather than clusters, since both treatment and control participants were present at each site. This is a post hoc calculation; no a priori power analysis was conducted. Under individual-level randomization, for a two-sample comparison at α = 0.05 and power = 0.80, the minimum detectable effect size is approximately d = 0.45 for the largest cohort grouping (Phase 2, Treatment = 73 vs. Waitlist Control = 84). On the ESQ overall emotional-health score (7-point scale, pre-test SD ≈ 0.7), this corresponds to roughly 0.3 scale points. For the smaller Phase 1 sample (24 vs. 34), the minimum detectable effect is larger, approximately d = 0.75.

The mental health crisis in graduate education combined with low treatment rates among engineering graduate students underscores the need for engineering graduate programs to provide effective methods to promote well-being. There is an extensive body of neuroscience research showing that contemplative practices, such as mindfulness, produce measurable effects on brain function and overall well-being. We hypothesized that a mindfulness-based training program designed for engineering graduate students would improve emotional well-being and, secondarily, enhance research capacity. An initial pilot study was conducted at a single institution (Phase 1), followed by a larger study conducted at both the original and a second institution (Phase 2) to gather additional data and show the program’s transferability. The program comprised eight weekly mindfulness training sessions. Individuals in the study were randomly assigned to either an intervention group or wait-list control group. We administered pre- and post-test surveys with quantitative measures designed to assess emotional and physical well-being, as well as creativity, research satisfaction, and desire to contribute to the betterment of society. Participants also completed a summative survey to evaluate the impact of the program on their well-being and research. Analysis revealed statistically significant findings: improved emotional health, decreased neuroticism, increased positive affect, decreased negative affect, and increased mindfulness in the intervention groups compared to the control groups. Intervention groups in Phase 2 also reported statistically significant improvement in satisfaction with their research. Our findings suggest that mindfulness training has the potential to play a vital professional and personal development role in graduate engineering education.

Crone WC, Kesebir P, Hays B, Mirgain SA, Davidson RJ, Hagness SC (2023) Cultivating well-being in engineering graduate students through mindfulness training. PLoS ONE 18(3): e0281994. https://doi.org/10.1371/journal.pone.0281994

https://doi.org/10.1371/journal.pone.0281994