Journal Logo

Empirical Investigations

The Effect of an Olfactory and Visual Cue on Realism and Engagement in a Health Care Simulation Experience

Nanji, Karen C. MD, MPH; Baca, Kirsten MD; Raemer, Daniel B. PhD

Author Information
Simulation in Healthcare: The Journal of the Society for Simulation in Healthcare: June 2013 - Volume 8 - Issue 3 - p 143-147
doi: 10.1097/SIH.0b013e31827d27f9
  • Free


Health care simulation is an educational technique that has gained popularity during the past 20 years. It is widely believed that subjects’ perceptions of realism in simulation settings can have a large impact on their experiences and learning.1,2 Despite this widespread agreement, there is little empirical evidence regarding what constitutes optimal fidelity for health care simulation exercises.

Various theories of realism and learning engagement have been proposed to explain what might affect the subject’s experiences. Beaubien and Baker3 were the first to propose that health care simulation realism is a multidimensional phenomenon. They proposed a model adapted from aviation simulation that describes environment fidelity, equipment fidelity, and psychological fidelity. Dieckmann et al,4 interpreting theory from the social sciences and advancing the theory of suspension of disbelief, have proposed 3 aspects of simulation realism (physical, semantical, and phenomenal) as having strong bearing on the learner’s experience.5 Rudolph et al6 suggested using the terms physical, conceptual, and emotional/experiential, to be more descriptive. We elected to expand on these theories as a basis for exploring realism and learning engagement in this study.

We propose that health care simulations posses the attributes of 3 types of fidelity, where fidelity describes how accurately reality is represented. Physical fidelity is the degree to which the simulation elements are sensed as approximating reality. Properties of physical fidelity include visual, tactile, auditory, olfactory, and orientation of the simulation elements. Conceptual fidelity is the degree to which the simulation proceeds in a plausible manner. The physiologic or pharmacologic response of the simulated patient is an example property of conceptual fidelity. Emotional/experiential fidelity is the degree to which the simulation generates feelings of reality. Realistic time pressure, consequences, and interactions would be a property of emotional/experiential fidelity.

For a subject in a simulation exercise, the 3 kinds of fidelity combine to produce a perception of realism for that individual. Thus, realism is a property of the individual rather than of the simulation itself. With a given simulation fidelity, one individual may perceive a certain degree of realism, whereas another may experience a very different perception of realism.

Considering the perceived realism, an individual must decide if he or she can engage in a meaningful learning experience. This decision is often affected by the fiction contract between that individual and the educator conducting the educational experience. A fiction contract is the implicit or explicit agreement between the parties about how the subject is expected to interact with the simulated situation and how the educators will treat that interaction.5

Using this theory of realism and learning engagement, we are interested in the impact of simulation fidelity. This study investigates the physical component of fidelity in an effort to understand what contribution a single physical element may have on a subject’s perception of realism in a simulation and on their subsequent engagement in the learning experience. Specifically, the purpose of this study was to determine whether an isolated visual and olfactory sensory change in the simulation environment affects the subjects’ perceptions of realism during simulation cases.


This study was conducted as part of regularly scheduled full-daylong simulation-based crisis resource management courses at the Center for Medical Simulation in Cambridge, MA. Subjects in consecutive courses of 2 types from June 14, 2011, to December 13, 2011, were included in the study. One course type was for board-certified and/or board-eligible anesthesiologists from 6 different institutions. and the other course type was for anesthesiology resident trainees in their second to fourth postgraduate years at 4 different teaching hospitals. Three to 6 subjects were assigned to each simulation course by their institutions on an availability basis. Each course ran approximately 7 hours, during which 3 to 6 scenarios were enacted. All of the scenarios took place in the operating room (OR) during a surgical procedure. However, none were intended to focus on electrosurgery in any manner. With institutional review board approval, 103 subjects from 22 courses were invited to participate, and all chose to do so. Before the study, a randomization table was used to prospectively designate each course day as intervention or control.

The same instructors and staff conducted the intervention and control simulations during routine crisis resource management courses. These courses for practicing anesthesiologists and anesthesiology trainees had occurred approximately weekly for 9 years and 16 years, respectively. Each course consisted of an introduction followed by 3 to 6 simulated cases and debriefings. The introduction for each course included a 4-page document that described the simulated environment, the mannequins, and any special equipment the subjects were expected to use during the course. Known limitations in the environment were also described in this document. In addition, the instructor presented a fiction contract to the subjects, explaining that the realism is limited by the available technology and asking that the subjects try to solve the problems presented to them as if they were encountering them in an actual clinical environment. The cases presented for each course type did not change from day to day. The simulation staff and instructors were blind to whether a course day was in the control or intervention group until the first simulation case was set up and about to commence.

An OR environment designed for simulation was used throughout. The physical space was similar in size, shape, and general appearance to a typical OR. The OR was outfitted with a standard operating table, an anesthesia machine and stock cart, and various devices that would be expected in an OR including a suture and equipment shelf, a radio tuned to a local FM radio station, surgical lights, suction canisters, laparoscopic equipment towers, and Mayo stands. Close inspection would reveal that, unlike a real OR, the floor had office quality linoleum tiles instead of a seamless poured surface, the ceiling tiles were acoustic absorbing instead of low particulate tiles, certain stainless steel cabinetry was missing, and 2 mirrored windows were present.

To simulate a patient, a Laerdal 3G mannequin (Laerdal, Stavanger, Norway) was placed on the operating table. The mannequin was draped appropriately for each case, and the scrub table was equipped with the proper surgical instruments. The anesthesia station was appropriately prepared for each case with airway equipment, medications in labeled syringes, intravenous lines with the appropriate administration sets, and other ancillary equipment readily available. Certain elements were not provided, such as volatile agents in the vaporizers, a twitch monitor, and some specialized airway management devices. Vital signs generated by the mannequin patient software were presented in real-time on a computer monitor placed in a typical location on the anesthesia machine. Actors who played the various roles (including surgeon, scrub technician, and circulating nurse) wore surgical caps, masks, and, where appropriate, surgical gloves and gowns. The actors were all experienced health care workers rather than professional actors. The anesthesia provider was the course subject and was also asked to wear scrub attire, a surgical cap, and a mask.

Two custom-made silicone devices, as shown in Figure 1, were constructed to simulate a surgical wounds in the abdomen and neck. The simulated surgical wounds were designed to allow for the actor surgeon to use electrosurgery to produce the characteristic appearance of smoke and odor of burning tissue. The design concept of the simulated electrosurgical unit (ESU) is shown in Figure 2.

Simulated abdominal surgical wound.
Design concept of the simulated ESU.

Before each case, a 3- to 5-oz strip of bovine muscle tissue was placed within the simulated surgical wound device on an aluminum foil sheet that was, in turn, connected to the return wire of the ESU. The surgical wound devices were prepared identically on intervention and control days. On intervention days, an electrosurgical operating tip and the return wire from the surgical wound device were connected to a working ESU. On control days, the electrosurgical operating tip and return wire were connected to an ESU that had been modified such that it did not produce electrical current but did produce the typical operating sound of that device. On both intervention and control days, the actor surgeon performed realistic surgical movements with the electrosurgical tip. On intervention days, the bovine tissue burned and produced smoke and a characteristic odor. Before beginning the study, several surgeons, anesthesiologists, as well as OR nurses and technicians not involved in the experiment were informally asked if the smell and sight of the ESU simulation were accurate and all agreed that they were.

At the end of each simulation course day, subjects completed a 7-question survey soliciting their assessment of fidelity of the simulation, perceived realism, and their learning engagement. The first 3 questions were intended to assess level of training, specialty, and simulation experience. The next 3 questions were intended to assess the perception of physical, conceptual, and emotional/experiential fidelity. The final question was intended to assess whether the perception of realism enabled learning engagement.

The ordinal rating data obtained from the survey were analyzed using a nonparametric 2-sample independent Wilcoxon rank sum test (Statistics Online Computational Resource;


The experiment was conducted during 23 consecutive courses with 107 subjects serving as experimental subjects. The survey data from 1 course day was excluded owing to external disruption. No other exclusions were made. The intervention (n = 52) and control (n = 51) groups were not statistically different with respect to demographic characteristics, which are summarized in Table 1.

Subject Demographic Data

Figures 3, 4, and 5 show that the subjects’ assessment of the simulation’s physical (P = 0.73), conceptual (P = 0.34), and emotional/experiential (P = 0.12) fidelity, respectively, did not differ between the intervention and control groups. Furthermore, Figure 6 shows that the subjects’ perceived realism leading to learning engagement of the simulation exercise did not differ between the 2 groups (P = 0.71). When looking at subgroups of trainee or nontrainee physicians alone, no differences were noted between the intervention and control groups.

Response frequency to question 4 on the survey: The simulation looked, felt, sounded, and smelled like an actual clinical situation. This question was used to measure physical fidelity on a 7-point Likert scale. The responses between the intervention and control groups were not statistically different (P = 0.73)
Response frequency to question 5 on the survey: Things proceeded in the simulation like in an actual clinical environment. This question was used to measure conceptual fidelity on a 7-point Likert scale. The responses between the intervention and control group were not statistically different (P = 0.34).
Response frequency to question 6 on the survey: I felt like I had an actual clinical experience. This question was used to measure emotional fidelity on a 7-point Likert scale. The responses between the intervention and control groups were not statistically different (P = 0.12).
Response frequency to question 7 on the survey: I found the simulation realistic enough to allow me to have an engaging learning experience. This question was used to measure the subjects’ perceived realism on a 7-point Likert scale. The responses between the intervention and control groups were not statistically different (P = 0.71).


In our study, when an isolated component of physical fidelity was changed in a series of simulated OR cases during a crisis resource management course, we found no statistically significant difference in subjects’ perceptions of the realism of the simulation. Our subgroup analysis on trainee or nontrainee physicians alone also showed no difference in subjects’ perceptions of realism, despite presumed better situational awareness among nontrainee physicians. This is contrary to existing literature, which suggests that the physical fidelity of simulations is an important factor in perceived realism and subject engagement.1,2,7

There are several potential explanations for our finding. First, the simulation environment used may have had a relatively high preexisting level of fidelity such that the perception of realism was high for the subjects at baseline. Sixty-nine percent of subjects rated 7 on a 7-point Likert scale that the simulation was realistic enough for them to have an engaging learning experience and 96% rated 6 or 7, leaving little room for improvement. These ratings are noticeably higher than the assessments of the 3 types of fidelity. This suggests that the fiction contract or educational content was compelling despite some doubts about the fidelity and may have resulted in a ceiling effect in this experiment. The incremental benefit gained from the enhancement to physical fidelity in the intervention group may not have had as much impact as it could have in an overall lower-fidelity simulation. Second, our sample size may not have been large enough to show a very small magnitude effect of improving only one of the several potential aspects of fidelity. Even with a small effect detected, we would not know whether the value of such an improvement in fidelity would be worth the expense and effort. Third, our survey instrument may not have been sensitive enough to detect a change in perceived realism. Finally, the particular choice of primarily an olfactory and visual cue as the stimulus in this experiment might be too subtle to come to the subject’s conscious level as they completed the survey instrument. Furthermore, the olfactory and visual stimuli of burning tissue may be subconsciously blocked by anesthesiologists because they are generally not responsible for these odors in their environment. A more conspicuous stimulus or one that was in the anesthesiologist’s central workspace may have produced a larger effect. Conversely, this stimulus may have produced a larger effect if the simulation subjects were surgeons because the ESU is in the central workspace of surgeons.8 Future research can be directed at understanding how simulation subjects respond to cues in various circles of focus within their workspaces.

Although we did not find an effect of improving one aspect of physical fidelity, we are encouraged to study other elements that contribute to perceived realism using this methodology. The choices are numerous as we can vary many different cues to map the most compelling aspects of the simulation fidelity.

In summary, we have proposed a model for the relationship of different kinds of fidelity to perceived realism and learning engagement in a health care simulation. An experimental method to test the effect of fidelity on perceived realism and learning engagement was devised and conducted for a small improvement to physical fidelity. The addition of a high-fidelity surgically generated odor and visual cue to an anesthesia crisis resource management course for trainees and for practicing anesthesiologists did not affect their overall perceptions of realism and engagement in the learning experience.


The authors thank the simulation participants as well as the staff at the Center for Medical Simulation in Boston, MA, especially Anthony Dancel, Christina Valle, Laura Gay Majerus, Robert Nadelberg, MD, and Rhonda Young.


1. Saus ER, Johnsen BH, Eid J. Perceived learning outcome: the relationship between experience, realism, and situation awareness during simulator training. Int Marit Health 2010; 64: 258–264.
2. Issenberg SB, McGaghie WC, Petrusa ER, Lee Gordon D, Scalese RJ. Features and uses of high-fidelity medical simulations that lead to effective learning: a beme systematic review. Med Teach 2005; 27: 10–28.
3. Beaubien JM, Baker DP. The use of simulation for training teamwork skills in health care: how low can you go? Qual Saf Health Care 2004; 13 (Suppl 1): i51–i56.
4. Coleridge S. Lyrical ballads. In: Engell J, Bate J, eds. The Collected Works of Samuel Taylor Coleridge. Princeton, NJ: Princeton University Press; 1801: 5–18.
5. Dieckmann P, Gaba D, Rall M. Deepening the theoretical foundations of patient simulation as social practice. Simul Healthc 2007; 2: 183–193.
6. Rudolph JW, Simon R, Raemer DB. Which reality matters? Questions on the path to high engagement in healthcare simulation. Simul Healthc 2007; 2: 161–163.
7. Weaver A. High-fidelity patient simulation in nursing education: An integrative review. Nurs Educ Perspect 2011; 32: 37–40.
8. Kneebone R. Simulation, safety and surgery. Qual Saf Health Care 2010; 19 (Suppl 3): i47–i52.

Simulation fidelity; Realism

© 2013 Society for Simulation in Healthcare