Intrinsic motivation of preclinical medical students participating in high-fidelity mannequin simulation

Abstract

Introduction

While medical schools strive to foster students’ lifelong learning, motivational theories have not played an explicit role in curricular design. Self-determination Theory is a prominent motivational theory. It posits that perceived autonomy, competence and relatedness foster intrinsic motivation. This study explores the effects of autonomy on intrinsic motivation in medical students participating in high-fidelity mannequin simulation.

Methods

A non-randomised crossover trial compared first-year medical students participating in (1) required simulation sessions with predetermined learning objectives and (2) extracurricular simulation sessions with student-directed learning objectives. An adapted Intrinsic Motivation Inventory (IMI) was used to assess intrinsic motivation, perceived autonomy, competence and relatedness. Each participant completed the IMI survey after each type of session. Variables were compared with signed-rank tests.

Results

All 22 participants completed the IMI after both types of session. Perceived autonomy was significantly higher during extracurricular simulation (p<0.001), but intrinsic motivation, competence and relatedness were not. Intrinsic motivation correlated with autonomy (RS=0.57 and extracurricular simulation, ES=0.52), competence (RS=0.46 and ES=0.15) and relatedness (RS=0.51 and ES=0.64). The IMI subscales had good internal consistency (Cronbach's α=0.84, 0.90, 0.90 and 0.76 for intrinsic motivation, autonomy, competence and relatedness, respectively).

Conclusions

Extracurricular sessions increased students’ perceived autonomy, but they were highly intrinsically motivated in both settings. Further study is needed to understand the relationship between perceived autonomy and intrinsic motivation in medical education learning activities. The IMI shows promise as a measurement tool for this work.

Introduction

A shift towards student-centred medical education has resulted in a proliferation of educational methods such as problem-based learning, team-based learning and outcome-based education.^{1 2} These innovations were also intended to develop students into lifelong learners.³ While lifelong learners are, by definition, self-motivated to continue learning, motivational theories have not historically played a prominent role in the design and evaluation of medical curricula.⁴ This is particularly relevant given that a medical student's motivation may have more of an effect on learning outcomes than formal medical school curricula.⁵ This conclusion is supported by evidence that different educational approaches⁴ and curricula⁶ result in remarkably similar outcomes.

Self-determination Theory (SDT) is a prominent theory in the psychology of motivation and there have been calls to further incorporate its tenets into medical education.^{4 5 7–10} SDT posits that individuals are naturally inclined to seek personal growth and self-actualisation, and that support of three basic psychological needs—autonomy, competence and relatedness—will facilitate such growth. Autonomy refers to an individual's beliefs about the control and responsibility they have over their behaviour. Competence refers to an individual's perception of their capability of achieving a desired outcome. Relatedness refers to an individual's feelings of connectedness with others. Learning and learner self-directedness is believed to be dependent on meeting these three basic needs.¹¹

High-fidelity mannequin simulation is an educational modality that is well supported by the tenets of SDT. Such simulation can foster competence by matching case difficulty with learner ability and providing detailed feedback.^{12 13} Relatedness is generally high because simulation instructors are trained to be particularly attentive, caring and supportive.^{12 13} While students are given autonomy to care for a patient during simulation, their learning objectives, session timing and peers are often selected for them based on predetermined curricular objectives. This is relevant because in other contexts students experienced sustained learning¹⁴ and greater academic success¹⁵ when their instructors were perceived to be supportive of participant autonomy. The level of autonomy that medical students perceive during simulation exercises and how it affects their intrinsic motivation is not well understood. The purpose of this study was to determine whether first-year medical students reported greater intrinsic motivation when participating in higher autonomy simulation sessions as compared with lower autonomy sessions. We hypothesised that increasing autonomy would increase intrinsic motivation.

Methods

Study design

A non-randomised crossover design (figure 1) was used to compare the motivational profiles of medical students participating in required simulation with predetermined learning objectives and the motivational profiles of the same medical students participating in extracurricular simulation with student-directed learning objectives (figure 2). Each student served as his/her own control. The primary outcome was the difference in the intrinsic motivation of each student during these sessions. Secondary outcomes included changes in perceived autonomy, competence and relatedness. The study was approved by the Harvard Medical School Institutional Review Board. Results were reported in compliance with the TREND statement for non-randomised trials.¹⁶

Figure 1.

The study participation, facilitator and session order of medical students in the elective simulation programme.

Figure 2.

Comparison of required and extracurricular simulation sessions.

Study setting and sample

All simulation sessions occurred in simulation laboratories on the medical school campus. An elective simulation programme offered extracurricular sessions to enrolled preclinical medical students throughout the year. Each participating student selected a 1 h session with 3–4 other students at a pre-specified date and time every 2 weeks for the full term. The required sessions occurred during various courses throughout the medical school's preclinical curriculum over the course of the year. There were typically 7–10 students in each of the required sessions.

Sixty of the 165 first-year medical students participated in the elective simulation programme during the 2014 winter semester. A convenience sample of these 60 students was the cohort for this study.

The facilitators for the study simulation sessions were two physicians with residency training in emergency medicine, as well as simulation fellowship training (BT and NW). Both regularly facilitated the extracurricular and required sessions throughout 2013–2014.

Selection and assignment of participants

Mandatory sessions during the students’ first-year physiology course that were held during a single day were used as the required session for this study. Elective sessions held during a 2-week period before and after this required session were used as the extracurricular sessions. This time frame was selected so that the sessions would be temporally related and the order effect could be mitigated.

A convenience sample of the students in the elective simulation programme was scheduled to participate in a required session with the same facilitator by course managers unfamiliar with other aspects of the study. Students who either did not participate in extracurricular simulation or who could not be paired with the same facilitator in both sessions were excluded.

Simulation sessions

Figure 2 compares the characteristics of the required and extracurricular sessions. In both types, students were presented with a patient, as represented by a high-fidelity mannequin (Laerdal SimMan 3G Essential), who had a medical condition that required active management. A single clinician-facilitator (BT or NW) presented the scripted case, provided the voice for the mannequin, clarified physical exam findings, adjusted the mannequin's vital signs and other parameters based on student actions, and conducted the debriefing session. Imaging, laboratories and ECGs were displayed on a large LCD screen within the room. Consultant physicians and a nurse were also voiced overhead by the facilitator. The facilitators and students were not blinded to whether the session was required or extracurricular, but the students were unaware of the specific elements of the study hypothesis.

The case for the required session had previously been developed and used as part of the formal physiology course curriculum. It involved a patient presenting to the emergency department with chest pain and aortic stenosis who subsequently develops rapid atrial fibrillation and hypotension after being admitted to the hospital. This required simulation was held during class time, attendance was obligatory and the learning objectives of the debrief discussion were predetermined by the course directors. Half of the students participating in each session (about 3–5 students) cared for the patient at the bedside during the emergency department portion of the case, while the other half watched. These roles were reversed when the patient was transferred to a new team on the floor, at which point the second team assumed care at the bedside. Following the case, the students were guided through learning objectives predetermined by the course director. This is the norm for required simulation sessions at our institution and ensures that the predetermined learning objectives for each case are met by all groups.

The case for the extracurricular session was developed by the session facilitators (BT and NW) with the intention of paralleling the core clinical elements of the required case (ie, shock), but with a different disease entity. It involved a patient with hypovolemic shock secondary to intraperitoneal haemorrhage following a trauma. It was held outside of class time, attendance was optional, and the learning objectives and debrief discussion were determined by the students. Following the case, the students were asked what they would like to discuss using various questions (ie, What did you struggle with? What do you want to learn about?). These questions were designed to identify student-directed learning objectives that were then discussed during the debrief component of the session. This is the norm for the extracurricular programme at our institution and can result in discussions that vary dramatically between groups.

Study procedures

An online survey was created using Qualtrics (http://www.qualtrics.com/), a secure online survey tool. The survey consisted of demographic questions and items of the selected Intrinsic Motivation Inventory (IMI) subscales arranged in random order.¹⁷ Survey completion was optional and anonymous.

The study participants were asked to complete the IMI following both their required and extracurricular sessions. As outlined in figure 1, approximately half of the students were scheduled to complete their extracurricular session first with facilitator A, while the other half were scheduled to complete their required session first with facilitator B. This allowed the order effect to be mitigated and assessed. Each eligible participant received an email that outlined the study prior to their first study session and an email with a link to the survey following their session. Two reminder emails were sent over the following week. The survey was closed when responses from all students in each session had been received or 1 week from the end of the session, whichever occurred first.

Measurements

The IMI instrument was used to assess four constructs of SDT within the simulation sessions.¹⁷

Each participant's motivational profile during the extracurricular and required sessions was examined using four of the IMI's subscales: intrinsic motivation, autonomy, competence and relatedness. In addition to the intrinsic motivation subscale, the three additional subscales on autonomy, competence and relatedness were selected because they represent the innate psychological needs of the learner.¹¹

Autonomy was the independent variable. Efforts were made to minimise comparative variation in competence (the elective and required case materials were designed to be of similar difficulty) and relatedness (the students had the same facilitator for both the elective and required sessions); therefore, these subscales were not expected to be significantly different between the sessions. The version of the IMI used contained all 28 statements from the selected subscales. Each was scored on a seven-point Likert scale (see online supplementary appendix 1).

Validity evidence for the IMI

Five forms of validity evidence in the context of simulation are described by Messick,¹⁸ and several have been assessed in other studies that used the IMI.¹⁹ The IMI has content evidence because it was developed using a well-known framework by experts in the field and its components have demonstrated validity in other studies.^{17 20–22} While it has not been used in the context of simulation or medical education, it has been used to assess the motivational aspects of competitive sports,²⁰ fitness classes,^{23 24} tests of endurance,²² a psychology class,²⁵ memory,²¹ homework²⁶ and other activities¹⁷ and found to have relationships with other variables that are consistent with its theoretical framework. Internal structure evidence for the intrinsic motivation and competence subscales has been demonstrated with internal consistency^{20 22 23} and temporal stability²² while the autonomy²⁶ and relatedness^{17 25} subscales have rarely been assessed. Consequence evidence requires long-term follow-up and has not been assessed for the IMI while response process evidence must be assessed within the context of individual studies.

Statistical analysis

The median score (scale=1–7) for each component of the IMI was calculated.¹⁷ Wilcoxon signed-ranks tests were used to compare intrinsic motivation, autonomy, competence and relatedness between the extracurricular and required sessions for each group.

A power calculation was conducted a priori. With an α of 0.05, a β of 0.2 and an SD of 1.3,^{20 22 23} 16 participants were required to complete both sessions to find a one-point change on the Likert scale for the primary outcome (a magnitude of change we feel would be clinically significant). Since comparisons were planned between each of the four subscales, the α was corrected to 0.0125 with the Bonferroni correction,²⁷ which increased the number of participants required to 22.

As the IMI had not previously been used in medical simulation, its validity evidence was assessed in accordance with Messick's¹⁸ framework when possible.¹⁹ The internal consistency of each subscale of the IMI was assessed by calculating Cronbach's α,²⁸ while the relationships with other variables were assessed by calculating Spearman's r for the IMI subscales in both groups to determine whether they correlated in a manner consistent with SDT.

Results

Participants

Table 1 outlines the demographics of the study participants. Figure 1 outlines the student participation. Of the 60 potential participants who were already participating in the extracurricular simulation sessions, 22 (36.7%) attended the required and extracurricular sessions with the same study facilitator and were eligible to participate in the study. The median number of students in the required and extracurricular sessions was 9.5 and 3, respectively. All study participants completed the IMI following both sessions (100% response rate).

Table 1.

Demographic characteristics of learners

	n (%)
Facilitator
A	9 (41)
B	13 (59)
Session order
Extracurricular first	9 (41)
Required first	13 (59)
Gender
Male	10 (45)
Female	12 (55)
Age
21–25	16 (73)
26–30	4 (18)
31–35	2 (9)

Motivational profile

The IMI median (IQR) scores of the learners in the extracurricular and required sessions are presented and compared in table 2. Autonomy was significantly higher during the extracurricular session (median 7.0 vs 5.1, p<0.001), but intrinsic motivation, competence and relatedness were not. Intrinsic motivation was scored highly in the extracurricular and required sessions (median 6.4 vs 6.3).

Table 2.

Motivational profile of medical students during extracurricular and required simulation sessions

IMI subscale	Extracurricular median (IQR)	Required median (IQR)	p Signed-ranks test
Intrinsic Motivation	6.4 (5.9–6.7)	6.3 (5.9–6.6)	0.81
Autonomy	7.0 (6.6–7.0)	5.1 (4.1–6.0)	<0.001
Competence	3.8 (3.3–4.6)	4.1 (3.3–4.5)	0.92
Relatedness	5.8 (5.5–6.4)	5.7 (5.3–6.1)	0.17

Validity evidence for the IMI

In the extracurricular and required sessions, Spearman's r indicated a modest positive correlation between intrinsic motivation and measures of perceived autonomy (r=0.52 and 0.57), competence (r=0.15 and 0.46) and relatedness (r=0.64 and 0.51). High levels of internal consistency were found among the items of the IMI subscales for intrinsic motivation (α=0.84), autonomy (α=0.90), competence (α=0.90) and relatedness (α=0.76).

Discussion

This study examined how varying autonomy during required and extracurricular medical student simulation sessions related to intrinsic motivation. As expected, participants scored significantly higher on the perceived autonomy subscale during extracurricular sessions. Also as expected, competence and relatedness scores were not significantly different between the two study conditions, suggesting that the simulation case difficulty and participants’ feelings about the facilitator did not confound the experiment. However, we did not find the hypothesised difference in intrinsic motivation between the sessions.

This hypothesis was based on SDT, which suggests that when students perceive increased autonomy, they will report higher levels of intrinsic motivation. This finding would have provided preliminary support for increasing the amount of student-directed activities in the medical school curriculum with the goal of enhancing lifelong learning. However, our results suggest that medical students participating in these study sessions were highly intrinsically motivated in general, regardless of differences in autonomy conditions.

While this is not the result we expected, these findings can be reconciled within SDT. SDT posits that motivation exists on a continuum from amotivation (absence of motivation) through extrinsic motivation to intrinsic motivation (engaging in an activity out of the inherent satisfaction it provides).⁴ Integrated regulation is the portion of the spectrum closest to intrinsic motivation where an individual is prompted to act based on an external regulation, but fully endorses the activity because it is congruent with their goals and values.¹¹ Thus, if the requirements of the required sessions were consistent with the students’ goals and values, they would attain a highly integrated, autonomous form of extrinsic motivation despite lower perceived autonomy.¹¹ This was most likely the case in our population of students because all of them elected to participate in extracurricular simulation, suggesting that learning through this methodology was consistent with their goals and values.

From a psychometric standpoint, another explanation for the results is that the high values obtained for the intrinsic motivation subscale during the required sessions resulted in a ceiling effect²⁹ that obscured any change in motivation that resulted from the increase in autonomy. Previous studies in various settings have recorded much lower scores for this subscale of the IMI (mean=5.69,²³ 4.2,²² and 3.25–3.5²⁶), suggesting that medical students who elected to participate in simulation during their free time have a relatively high level of intrinsic motivation for this activity relative to the populations and contexts previously examined using the IMI. This is an interesting finding suggesting that these medical students had a uniquely high intrinsic motivation when participating in simulation. If other populations of medical students score as highly on this subscale, it may need to be altered or redesigned for use in medical simulation.

The results of this study suggest that preclinical medical students who volunteer to participate in a learning activity like simulation have high levels of intrinsic motivation that are not affected by their level of perceived autonomy. As this study assessed a select, highly motivated cohort of medical students who volunteered for extracurricular simulation, its results cannot be applied broadly. However, further research characterising the importance of student autonomy in non-simulation learning activities and in students who did not volunteer for an elective programme would provide further insight into the impact of learning activities on medical student motivation.

The IMI may be a viable instrument for use in this research and its use in medical education should be guided by these findings. Consistent with Messick's¹⁸ framework,¹⁹ this study added to the IMI's pre-existing validity evidence by demonstrating internal structure evidence (high Cronbach's αs for each of the studied subscales) and appropriate relationships with other variables (Spearman's correlations were consistent with the theoretical relationships between the constructs) while highlighting a potential problem (the ceiling effect in the intrinsic motivation subscale).

Limitations

This study has several limitations. A low proportion (36.7%) of the study population and the first-year class (13.3%) was involved in the study. Since study eligibility was determined by administrative scheduling logistics, sampling bias was unlikely within the study population but a larger population of students would have increased the generalisability of the results. Additionally, selection bias is a potentially major confounder as students in the first-year class who elected to participate in additional simulation were likely to be more intrinsically motivated for any simulation activity than their peers.

The session order did not vary between the facilitators and more students completed their sessions with Facilitator B (13; required first) than Facilitator A (9; extracurricular first). Differences between the facilitators could have impacted the results because the larger number of students who worked with Facilitator B weighed more strongly in the results. However, the magnitude of any potential difference is likely to be negligible given the relatively small variation in cohort size and similarity in teaching style between the two instructors (both were fellows in the same teaching programme). No significant differences between the assessed variables were found in groups taught by either instructor, although the study was not powered to assess this.

Although self-report data are generally considered the lowest form of evidence³⁰ and were the only type collected in this study, the constructs of SDT are necessarily intrinsic to the individual, making it difficult to assess them in a more objective fashion (eg, it is not whether a learner appears competent to an observer that matters, but whether the learner perceives that they are competent).

Both facilitators were aware of the group assignment. The students, while unaware of the purpose of the study, also knew when they were participating in required and extracurricular sessions. This lack of blinding may have affected the actions of the facilitators and responses of the students.

Conclusions

We hypothesised that extracurricular sessions taught with student-directed objectives would increase the self-reported measure of autonomy and intrinsic motivation on the IMI. Our results suggest that the extracurricular sessions increase the students’ perceived autonomy compared with the required sessions with predetermined objectives, but that the extracurricular sessions did not increase intrinsic motivation as compared with the required sessions. The students reported levels of intrinsic motivation in both groups that were substantially higher than those found previously. More research is needed to understand the relationship between perceived autonomy and intrinsic motivation among medical students both within simulation and other learning activities. The IMI shows promise as a measurement tool for such work.