VR Safety Training Efficacy Research: The Central Washington University Study

Last Updated: September 2026

An independent study at Central Washington University found that adding about 20
minutes of VR safety training on top of standard classroom and hands-on instruction improved
comprehension and 30-day knowledge retention. The VR group scored 6% higher in aggregate than
the control group, with gaps up to 30% on individual confined space questions. No VR
participant scored worse after 30 days, while some control participants declined by 8–31%.
The study was peer reviewed and published in the ASCE Computing in Civil Engineering 2023
proceedings under DOI 10.1061/9780784485231.004, and has been cited 10 times in Scopus-indexed
literature. Humulo supplied the two VR modules used and had no other role.

Citation

Dang, H., Serne, J., & Tafazzoli, M. (2023). “Virtual Reality Safety
Training Assessment in Construction Management and Safety and Health Management Programs.”
Computing in Civil Engineering 2023: Visualization, Information Modeling, and Simulation
(pp. 28–35). Reston, VA: American Society of Civil Engineers. ISBN 9780784485231.
DOI: 10.1061/9780784485231.004

Abstract, as published

The paper itself sits behind ASCE’s paywall. There is no open-access copy of it anywhere
— Unpaywall, OpenAlex, and Semantic Scholar all return no free full text, and OpenAlex records
its access status as closed. So the authors’ own abstract is reproduced below in full, with
attribution, because this page is the only place on the open web where you can read what the study
actually says without an ASCE subscription. Everything below this block is Humulo’s summary;
this block is the researchers’ words.

Ensuring accessible and high-quality safety training is critical for preventing injuries and fatalities on construction job sites, especially considering the potential impact on the staggering number of over seven million construction workers in the United States. Construction workers accounted for approximately 20% of worker fatalities, equating to one in every five deaths annually. Although relatively new, virtual reality (VR) safety training has significant potential for improving safety training with immersive experiences and interactive environments. However, the effectiveness of VR safety training remains unknown, and limited studies are available. To investigate VR safety training efficacy, the authors compared pre- and post-training between a control and an experimental group to evaluate knowledge comprehension and retention. The control group participated in traditional school-based and physical simulation learning, while the experimental group completed both the VR safety training and the same training as the control group. The study focused on two safety topics: fire safety and confined space safety. The results demonstrate that VR safety training significantly improves learner comprehension and retention of safety knowledge, supporting its potential as an effective training tool for construction workers. The study also revealed that the immersive VR experience was user-friendly and contributed to increased self-confidence in understanding safety procedures, knowledge, and practices.

— Dang, Serne & Tafazzoli (2024), abstract as
registered with the publisher. Retrieved from OpenAlex record
W4392456292.

Independent citation record

You do not have to take Humulo’s word for any of this. The paper has a DOI, and three
independent bibliographic indexes track it. Anyone can check the numbers below against the live
APIs. Counts differ between indexes because each one crawls a different slice of the literature,
which is normal and not a discrepancy to be explained away.

Two things worth being straight about. First, seven to ten citations is a modest but real
footprint for a conference paper three years out; it is not a landmark study and nobody should cite
it as one. Second, the paper does not name Humulo anywhere in it, so none of those citing authors
were citing us. What the record establishes is narrower and more useful than a marketing claim: an
independent university team put VR safety training on fire safety and confined space entry through a
controlled pre/post design, published the result through ASCE peer review, and other researchers
have since built on it.

Investigators

The study was designed, administered, and analyzed entirely by the university research team.
Humulo had no involvement in any of it.

• Dr. Hongtao Dang, Ph.D. — Assistant Professor, Washington State University.
• Dr. Jennifer Serne, Ph.D. — Assistant Professor, Department of Engineering Technologies, Safety, and Construction, Central Washington University.
• Dr. Mohammadsoroush Tafazzoli, Ph.D. — Associate Professor, Civil Engineering and Construction, Washington State University.

Humulo’s role in the study

Humulo provided off-the-shelf VR headsets loaded with two modules from its training library,
Fire Safety and Confined Space Entry. That was the entire extent of the company’s participation.
Humulo did not design the study, fund it, administer any session, collect or analyze data, or
contribute to the paper. The published paper reports the research design and outcomes and does
not name the VR vendor whose modules were used; that attribution comes from Humulo and from
CWU’s selection of the modules. We spell this out because a study’s value depends on who paid
for it, and the answer here is nobody at Humulo.

Study Design

The design is what makes the numbers readable. The experimental group did not swap classroom
training for VR. They got everything the control group got and added about 20 minutes of VR per
topic. Whatever difference shows up is the marginal effect of that addition, measured against
training that already included a live propane fire and a physical confined space exercise. This
is a harder test than comparing VR against nothing.

Quantitative Results

The headline 6% is the least interesting number here. The floor is the finding worth carrying.
Nobody in the VR group came back 30 days later scoring worse than they started, while part of the
control group lost as much as 31%. That pattern points at decay rather than peak performance:
twenty minutes of repeatable, scored practice seems to keep safety knowledge from draining away,
which is the actual problem most EHS managers have with annual refresher training.

Qualitative Results

All 17 members of the experimental group said the VR training improved their comprehension of
the fire extinguishment and confined space material. Participants were asked in writing: “Do
you feel that taking the VR training helped your comprehension of the course material related to
fire extinguishment and confined spaces?” Written responses pointed repeatedly at forced
sequencing and embodied practice as the reason, with participants noting that the module required
completing each step before moving on, and that recall during testing came back as a memory of
having physically done the task. Several reported performing better on the subsequent live fire
exercise. These are self-reports, and should be read as such.

What This Study Does Not Establish

Anyone citing this work should carry these limits alongside the numbers.

• Small sample. 37 completions at a single university. These are directional
  phase 1 results, not a population estimate.
• Students, not incumbent workers. Participants were safety and health majors.
  Closer to the target population than a general student sample, but not plant or warehouse employees.
• VR was a supplement, not a replacement. The study does not show VR can stand in
  for live fire training or a physical confined space exercise, and Humulo does not claim it can.
  Hands-on performance evaluation remains the employer’s obligation under the applicable OSHA standards.
• Retention was measured at 30 days. The data say nothing about six-month or
  annual retention.
• Qualitative findings are self-reported. The 100% figure reflects what
  participants said about their own comprehension, not an independent measure of it.
• Two modules, one vendor. Results describe these two modules and are not a
  general claim about every VR safety training product on the market.

Frequently Asked Questions

Is there peer-reviewed research proving VR safety training works?

Yes. The Central Washington University study was published as a peer-reviewed paper in the ASCE
Computing in Civil Engineering 2023 proceedings (DOI 10.1061/9780784485231.004) and has been cited
10 times in Scopus-indexed literature. It found that VR safety training significantly improved
comprehension and 30-day retention of safety knowledge compared with classroom and physical
simulation alone.

How much does VR improve safety training retention?

In the CWU study the VR group scored 6% higher in aggregate than the control group on a 30-day
delayed retention test, with advantages up to 30% on individual confined space questions. No VR
participant scored worse after 30 days, while some control participants dropped by 8–31%.

Who conducted the study and who funded it?

It was designed and carried out by Dr. Hongtao Dang, Dr. Jennifer Serne, and Dr. Mohammadsoroush
Tafazzoli in 2022, with fieldwork at Central Washington University. Humulo provided headsets and two
training modules and nothing else. Humulo did not fund, administer, or analyze the study and had no
role in the paper.

How long was the VR training in the study?

About 20 minutes per module, covering fire safety and confined space entry. Participants could
repeat modules, and scoring was embedded in the module. The measured gains came from that 20-minute
addition on top of standard training, tested 30 days later.

How many people took part in the study?

37 participants completed it: 20 in the control group and 17 in the experimental group that used
VR. All were students in CWU’s Safety and Health Management Program.

Can VR replace hands-on fire extinguisher or confined space training?

Not based on this study, and Humulo does not claim it can. Every participant in the experimental
group also completed the live propane fire exercise and the physical confined space simulation. The
study measured VR as a supplement. Employer obligations for hands-on performance evaluation under
OSHA standards are unchanged.

Related Reading

• peer-reviewed research on VR safety training — the systematic reviews and meta-analyses this study sits alongside.
• VR Safety Training Statistics & ROI Data — OSHA injury data, training costs, and wider efficacy research.
• VR safety training — how Humulo’s program is built and deployed.
• VR fire extinguisher training — the fire module used in this study.
• VR confined space training — the confined space module used in this study.
• Enterprise VR training — deployment, pricing, and scheduling a demo.