Secondary Logo

Journal Logo

Research Reports

A Study of the Effect of Dyad Practice Versus That of Individual Practice on Simulation-Based Complex Skills Learning and of Students’ Perceptions of How and Why Dyad Practice Contributes to Learning

Räder, Sune B.E.W. MD, PhD; Henriksen, Ann-Helen; Butrymovich, Vitalij MD; Sander, Mikael MD, PhD; Jørgensen, Erik MD; Lönn, Lars MD, PhD; Ringsted, Charlotte V. MD, PhD, MHPE

Author Information
doi: 10.1097/ACM.0000000000000373


Effective and efficient clinical skills training is needed given the demands for increased productivity and duty hours restrictions that affect both trainees and mentors.1–4 In addition, advances in technology, such as minimally invasive procedures (e.g., coronary angiography [CA]), have challenged the apprenticeship model for learning.5–7 Evidence supports the effectiveness of virtual reality simulators for teaching these minimally invasive procedures.8,9 Yet, cost and limited time for instruction restrict simulator training in educational institutions.3,5,10 Hence, effective and efficient training programs are needed. Research on motor skills training indicates that practice in pairs, called dyad practice, may benefit simulation-based complex skills learning.11,12

Conceivably, the benefits of dyad practice increase as the complexity of the task being practiced increases.13 Therefore, simulator-based dyad practice could be an efficient method for training future residents in complex skills; for example, CA involves a high degree of complexity, and achieving competence is mandatory for cardiology residents.14,15

Although different aspects of dyad practice have been explored with a moderate degree of complexity,11,12,16,17 studies of specific complex, integrated cognitive and motor skills are sparse. Conceptual frameworks have been proposed in the cognitive and motor skills domains. For example, in the cognitive skills domain, Topping and Ehly18,19 created an explanatory model of collaborative learning. One key element of their model is that both partners are learners, which promotes self-disclosure and fosters communication and dialogue between the pair. The model also emphasizes both partners’ support in managing and modulating information processing, and provides for modeling, support, and scaffolding.

Another conceptual framework is cognitive load theory (CLT).20–23 This framework explains that the advantage of using dyads is the united memory and collaborative information processing of the pair. Recent CLT publications have addressed task complexity and collaboration processes.20,21,24,25 Complex tasks have the risk of overloading the learner’s working memory. By collaborating with a partner, however, this load is divided and thus reduced for the individual learner. Yet, the effort required for the collaboration may in turn increase the learner’s cognitive load. The result, then, is a balance in the individual’s cognitive load. This balance, however, depends on the complexity of the task; when the task is simple22 or the dyad is malfunctioning,25 the balance shifts and the individual’s cognitive load increases.

The mechanisms of dyad practice in the motor skills domain have been extensively studied.11,12,26 In this domain, observation is an important factor contributing to motor skills learning. For example, behavioral and neuroimaging studies reveal that the observer activates similar processes and structures as the performer of the task. Also, the individual’s pauses between steps facilitate his or her partner’s processing of information about his or her movement patterns and strategies. This information processing is often not possible during concomitant practice. Moreover, pauses during complex skills training reduce the individual’s cognitive load.26 However, contrary to studies in the cognitive skills domain, dialogue in motor skills learning has ambiguous results, and a recent study found that dialogue is less important than observation.16

Clinical skills are complex and include a combination of cognitive and motor skills. Although conceptual frameworks from the cognitive and motor skills domains offer several explanations of how dyad practice works, a number of issues regarding learning complex, integrated cognitive and motor skills remain unresolved. In particular, the roles of observation and dialogue are unclear. Moreover, explanations of the mechanisms of dyad practice are based largely on experimental effectiveness studies outside the medical education domain and do not address learners’ perceptions.12,20,25,27

Thus, the aims of our study were (1) to explore the effectiveness of dyad practice compared with individual practice on a simulator for learning a complex clinical skill, and (2) to explore medical students’ perceptions of how and why dyad practice on a simulator contributes to learning complex, integrated cognitive and motor skills.


Study design

In 2011, we conducted a randomized trial of the effect of dyad practice versus individual practice, measured at a two-week follow-up, and a qualitative interview study.


We invited fifth- and sixth-year medical students to participate. We chose medical students instead of cardiology residents to obtain a sufficient sample size. Students at this level have passed the internal medicine exam, which includes cardiology. We advertised the study through the medical students’ newsletter, the medical faculty, and medical students’ organizations, and included participants on a first-come first-serve basis, until we reached our desired sample size.

One of us (S.B.E.W.R.) held information meetings about the study. Immediately after these meetings, interested participants signed an informed consent form and then were randomly assigned to the individual or dyad practice groups. A colleague not involved in the study used lists from for the randomization. See Figure 1 for the study flowchart.

Figure 1:
Flowchart of a study of the effect of dyad practice versus that of individual practice on medical students’ simulation-based complex skills learning, 2011.

Quantitative study

CA training program.

Participants watched a video demonstrating a CA on a real patient and instruction videos demonstrating a CA using the virtual reality simulator (Procedicus VIST, version 8.0.1, Mentice, Gothenburg, Sweden). The instruction videos included a demonstration of how the simulator works and a systematic description of the procedure, and emphasized process goals of performance (e.g., to keep the guidewire in front of the catheter when advancing through the descending aorta).28,29 All of the videos were available during the practice session but not during the assessment. One of us (S.B.E.W.R.) introduced participants to the practice session by giving a short, standardized presentation. Subsequently, the participants spent three and a half hours practicing two CA cases without an instructor present. The dyads were told to change places frequently so that each partner had an equal amount of hands-on time.

Assessment of learning outcomes.

We assessed learning outcomes two weeks after the practice sessions. We encouraged the participants to watch the instruction videos on the day before the assessment. Then, we assessed each individually while he or she performed two CAs, for a maximum of one hour. Thus, we collected two assessments per participant. We video recorded all performances, and two raters (V.B., M.S.), blinded to the participants’ identity and practice group, used a modified version of a previously validated CA rating scale (MOCARS) to assess the participants’ video-recorded performances (see Supplemental Digital Appendix 1 at The video recordings were split-screen and showed the movements of the participants’ hands and the catheter on the fluoroscope screen at the same time. The raters were senior residents in cardiology experienced in performing CA in the cardiac catheterization laboratory. Before the assessment, the raters were carefully instructed on using the MOCARS form. We also collected simulator-generated metrics, including procedure time, fluoroscopy time (i.e., x-ray single-picture time), cineloop time (i.e., x-ray video recording time), and contrast media volume (dye).

Sample size.

Before randomization, we calculated the necessary sample size for our study from the results of a previous study on CA performance curves. That study demonstrated a standard deviation (SD) of nine minutes regarding procedure time in the catheterization laboratory. We used a six-minute drop in procedure time from novice (i.e., no CA experience) to competent CA operator as the minimum required difference (MIREDIF).30 This result indicated that we needed a total sample size of 72 participants. We conducted a second power calculation during the study, on behalf of the first 40 assessments, to determine if we needed additional participants. The MIREDIF was set to 7.5—10% of the maximum MOCARS score. With an SD of 10, we needed a total sample size of 58 participants.

Qualitative study


One of us (A.-H.H.) conducted individual semistructured interviews with the participants in the dyad group immediately after the training program, using the following questions:

  1. What were the advantages and disadvantages of practicing in dyads (pairs)?
  2. What do you consider to be an important difference between dyad and individual practice?
  3. Which type of training would you prefer—dyad or individual practice—and why?

Each interview was recorded and transcribed verbatim.

Analyses of interviews.

We analyzed data according to a theory-driven variant of the immersion–crystallization technique31 using inductive (data-driven) and deductive (theory-driven) analysis in interplay. In this creative process, we restructured the themes that emerged from the data using the conceptual frameworks we identified in the literature and described in the introduction. Three of us (S.B.E.W.R., A.-H.H., C.V.R.) contributed to this iterative analysis process.

First, we independently read all the interview transcripts and then developed an initial coding structure to support the inductive analysis. After the initial coding, we turned to the literature to identify conceptual frameworks supporting the interpretation of our data and the emerging themes. We continuously discussed our interpretation of the data and their relations to the literature and revised the themes and recategorized the data until we reached an agreement. We then developed a new coding scheme, integrating the emergent themes and the frameworks from the literature. Two of us (A.-H.H., C.V.R.) independently used this scheme in coding a sample of the transcripts. Subsequent comparison revealed only minor disagreements that we resolved by consensus. Finally, one of us (A.-H.H.) coded the entire data set and then reviewed the coded data set with another investigator (C.V.R.).

Ethical approval

The study was approved by the regional committee on biomedical research ethics, Capital Region, Denmark (H-C-FSP-2009/17).

Statistical analysis

We present categorical data as numbers or percentages and continuous data as medians (ranges). We compared groups of categorical variables using chi-square tests and, depending on the distribution, continuous variables using the Mann–Whitney U test or the t test. Finally, we compared more than two groups of categorical variables using chi-square tests with gamma coefficients and, depending on the distribution, continuous variables using the Kruskall–Wallis test or the ANOVA test.

MOCARS scores represent the mean of the cases’ and raters’ sum-scores of items as a percentage of the maximum sum-score. We converted normally distributed metrics, a condition for the z score calculation, to a z score of metrics, and present it as a mean of z scores.

We explored aspects of equivalence of the two practice groups by the confidence interval (CI) of the difference between the two groups’ MOCARS scores. We estimated this CI using a corrected t test with regard to the correlation between the dyad pairs.

We explored correlations using R2 values for a linear relationship. To explore the internal consistency of items in the rating scale, we calculated Cronbach alpha (α) for the 15 MOCARS items. To explore interrater reliability, we calculated the intraclass correlation coefficient (ICC) for the two raters.

We considered P < .05 to be statistically significant. We performed power calculations using PS Power and Sample Size Calculations, version 3.0.34 (Dupont and Plummer, Department of Biostatistics, Vanderbilt University, Nashville, Tennessee), and all other statistical analyses using SPSS 18.0 for Windows (SPSS, Inc., Chicago, Illinois).


Study sample

Of the 84 participants who signed up, 72 (86%) completed the study; the remaining 14% did not attend the practice. We found no statistically significant differences in characteristics between the two groups (see Table 1). All assessments but 3 were completed (141 of 144; 98%). The 18 dyads included 8 female dyads (FF), 3 male dyads (MM), and 7 mixed-gender dyads (FM).

Table 1:
Characteristics of Participants in a Study of the Effect of Dyad Practice Versus That of Individual Practice on Simulation-Based Complex Skills Learning, 2011

Quantitative results

Learning outcomes.

The mean MOCARS score was 63% ± 16% for the dyad group and 68% ± 13% for the individual group (see Figure 2). A comparison between the groups that took into account the correlation between individuals in the same dyad yielded P = .18 with an estimated difference between the groups’ MOCARS scores of 3.67 (95% CI: –1.80 to 9.15). The median z scores for procedure time and fluoroscopy time were 0.23 (range –1.12 to 2.32) for the dyad group and –0.30 (range –1.02 to 4.32) for the individual group (P = .92) (see Figure 3). We did not calculate z scores for cineloop time and contrast media volume because they were not normally distributed.

Figure 2:
Box and whisker plot of the scores on a modified coronary angiography rating scale (MOCARS) of medical students in dyad practice versus those in individual practice, 2011.
Figure 3:
Box and whisker plot of the composite z score of procedure time and fluoroscopy time of medical students in dyad practice versus those in individual practice, 2011. Dots represent outliers.


The raters’ rankings of the participants were similar in approximately 70% of the assessments (MOCARS scores: ICC = 0.71; overall scores: ICC = 0.68). MOCARS items seemed to measure the same construct—CA competence (first case: α = 0.88 and α = 0.89 per rater; second case: α = 0.92 and α = 0.93 per rater). Correlations between MOCARS scores and overall performance scores were high (R2 = 0.92). A poor inverse correlation existed between the simulator-derived metrics and the performance scores given by the raters (MOCARS scores: R2 = 0.05).

Qualitative results

In general, we reached saturation of themes after 10 interviews. However, we interviewed all 36 students who participated in a dyad to ensure representativeness of dyad compositions regarding gender mix.

The mixed inductive/deductive analysis resulted in the following explanatory model: The shared memory of the instruction materials and cooperation in dealing with the complex task contributed to a reduction in cognitive load, particularly at the beginning of the practice. Alternate practicing (one participant observing while the other practiced, and then switching places) further reduced the cognitive load by offering participants breaks in concentration. In addition, participants learned from being observed as well as by observing their partner’s performance, errors, and coping strategies. Observational learning, in turn, mediated overt communication, which fostered individual reflection during performance and observation of the action, contributing to the co-construction of cognitive schema for problem solving.

However, this learning process was dependent on two equal-level novices and on the social interaction between them. The fact that both partners were novices promoted self-disclosure in performance and communication. Moreover, the social component of the dyad affected the participants’ motivation and encouragement. Finally, their dialogue and development over time prompted meta-cognitive reflections, resulting in scaffolding regarding feedback and comments when they became more competent (see Figure 4).

Figure 4:
Model of the interplay between the five theme categories derived from interviews of medical students who participated in a study of the effect of dyad practice with two equal-level novices on simulation-based complex skills learning. The five theme categories are (1) reduction of cognitive load, (2) observational learning, (3) learning from overt communication and dialogue, (4) social aspects and motivation, and (5) meta-cognition.

From the interviews, we extracted five theme categories: (1) reduction of cognitive load, (2) observational learning, (3) learning from overt communication and dialogue, (4) social aspects and motivation, and (5) meta-cognition. We identify the quotations below using a number indicating the specific dyad and letters indicating the gender composition of the dyad. In same-gender dyads, the second number indicates the participant. Data from the three dyad gender compositions equally reflected the five theme categories.

Reduction of cognitive load.

In several ways, participants expressed that the dyad contributed to reducing the cognitive load of learning the complex task. Partners supplemented each other and benefited from a united memory of the instruction prior to the practice. Moreover, they practiced a split attention allowing the operator to focus on the procedure while the observer provided oral information from the instruction material. Participants described this process as efficient. They also noted that having breaks from performing the procedure left time to process information and to rest from concentrating on the task, both advantages of dyad practice.

I think it is very nice that you can supplement each other. While you practice, your partner can read the instructions or watch the video, and then you can get the spoken information and don’t have to stop practicing. (5FM/F)

Observational learning.

Participants noted that they benefited both from observing their partner perform the procedure and from being observed. They learned from their partner’s performance and errors, stimulating their reflection on the action and contributing to their planning of their own performance. During observations, they pictured themselves completing the procedure. From being observed, participants received encouragement, advice, and corrections from their partner. A potential disadvantage, however, was that participants became dependent on being corrected, and hence less attentive in avoiding errors.

You observe things done, and you have time to reflect on it, and perhaps be aware of things you would not otherwise be aware of. (11FF/1F)

Learning from overt communication and dialogue.

Participants emphasized the advantage of overt communication and dialogue in fostering a cognitive coproduction. They described that being paired up forced them to talk and think aloud, which facilitated both collaborative and individual reflection during performance and observation of the action. However, participants also speculated that practicing with a more skilled partner or a tutor would not foster this kind of reflection and problem solving.

It is a huge advantage to think aloud and discuss the next step compared to what worked earlier and the errors we made. (15MM/1M)

Social aspects and motivation.

Participants noted that training with another novice gave them a sense of security. They were able to support each other and have fun while practicing. Again, the fact that both partners were novices promoted self-disclosure in performance and communication. Participants speculated that a potential disadvantage to dyad practice could be malfunctioning cooperation due to either domination by one partner or other personality issues. However, only 1 participant described a concrete example. Despite participants indicating that they would prefer to choose their own partner, 30 of the 36 preferred to practice in dyads, even if their partner was chosen for them. They indicated, though, the importance of partners being at the same experience level.

We had so much fun together and time went very fast.… You don’t feel stupid when you are two, because your training partner is equally as “stupid” as you. (9FM/F)

We didn’t cooperate properly. I used too much energy trying to find out how to cope with the situation. (16FM/F)


Participants reported reflecting on the learning process and developing learning strategies. In particular, the overt communication and dialogue fostered reflection during performance and observation of the action and contributed to a deeper understanding and better memory. Participants described how they worked out learning strategies including learning aids. In addition, after they had overcome the first challenges together, they practiced scaffolding to ensure that they practiced without the interference of their partner. Participants also reflected on the advantage of being equals and novices, rather than having a more capable partner or expert as a tutor. Finally, they indicated that the advantages of being a pair were more important than the disadvantages of having less hands-on practice time. Yet, participants suggested that future training protocols should include time for clarifying conditions regarding the collaboration and instruction in scaffolding.

By explaining things to each other you get a deeper understanding because you are forced to think about it, and together we created some algorithms. (20MM/1M)

It would have been so much different, if your partner was an expert regarding the procedure, because then you would be handed the solution to the problem instead of having to think it through together. (17FM/M)


We found that dyad practice was as effective as individual practice to learn a complex clinical skill. Thus, dyad practice is more efficient because pairs of students learn the same skill in less practice time.

Several factors support the robustness of our results. First, to standardize the quality and length of time of the instruction, we used instruction videos to avoid bias. Second, we calculated the necessary sample size before and during the study; our findings demonstrated a statistical power of more than 80% with the number of participants who completed the study. Hence, the risk of a type II error is minimal. Next, the high internal consistency of the MOCARS items, the small difference between MOCARS scores and overall scores, and the acceptable interrater reliability all contribute to the validity of the assessments.32 However, better rater training and including more raters or more cases may have increased the reliability in our study.32,33 Fourth, raters were blinded to participants’ training modality. Finally, the two groups (individual and dyad) were well matched in terms of participant characteristics.17,34–36

Both groups received directed self-guided training using instruction videos and focusing on process goals.28,37,38 Our findings demonstrate that this type of directed self-guided training is cost-effective, as both groups learned the CA skills in the absence of an instructor. However, unsupervised training is problematic because learners may develop bad habits and form incorrect conclusions about their learning.37 Still, the Dreyfus developmental model of skill acquisition shows that learners acquire the procedural steps of a new process first.39 Therefore, directed self-guided training likely is only possible during the initial phase of learning a new procedure, and a supervisor is needed after that to diagnose the trainee’s learning needs and focus her or his learning efforts more effectively.

The interviews we conducted explored students’ perceptions of how and why dyad practice contributes to learning complex, integrated cognitive and motor skills. We could have predicted that they would report a reduction in cognitive load, given the complexity of the task and the level of the learners.21,24 However, other studies carried out with more advanced or experienced learners or with simpler tasks show quite different results.22,26,40

Our finding that students learned from both observing and being observed is in line with the findings of prior studies on learning motor skills.16,40 However, the students’ emphasis of the advantage of both partners being novices was new. Recent studies indicate that the observation of the learning of a task rather than its performance is important for the implicit engagement of the observer’s neural systems for movement strategies.41 This argument supports the students’ claims about the advantage of same-level learners. Moreover, students’ conclusions about using imagery during observation to learn are supported by a recent study demonstrating that motor imagery improves performance more than motor practice alone, another effect that is related to complexity of the task.42 In addition, students emphasized that they learned from observing their partner’s errors. A recent review of cognitive and social-cognitive research on the mechanism of example-based learning indicates that observing a model that demonstrates erroneous performance and learning to cope with those problems is more powerful than observing a faultless model.43 This finding contributes to the explanation of why dyad practice is more efficient for novices than individual practice to learn a complex skill.

Our findings indicate that observation acted as a mediator of overt communication, dialogue, and active problem solving. This finding is in accordance with those of Crook and Beier,44 who concluded that the interaction and communication between high-quality dyad participants will prompt meta-cognitive processes in their partners. However, in contrast to recent results from the motor skills domain, students in our study noted the benefit of being forced to think and talk aloud.16 In the cognitive skills domain, however, no evidence supports the assertion that explaining to others is more advantageous than explaining to oneself.45 Yet, self-explaining and co-construction may produce more learning than listening to someone else explain.46,47 Although self-explaining is a powerful constructive activity, interactive joint explaining, prompted by practicing with a partner, may be even more effective.48 Hence, overt self-explaining and interactive communication seem to be important factors in complex, integrated cognitive and motor skills learning.

The meta-cognitive aspects of our study were rather elaborate, and, interestingly, students reflected on the timing of feedback as well as of scaffolding. Although the strategies they described may inspire changes to the structure and format of future training protocols, there is already an increasing awareness in the literature of the benefits of self-guided learning.25,37 For example, learning from a partner’s errors and the overt communication that takes place within a dyad are unique to simulation-based training and are not readily transferable to clinical training because of patient safety concerns. However, both may contribute to more efficient learning practices and in part explain why students learn more quickly from simulation training than from clinical training.49

Our study has several limitations. First, we included medical students instead of cardiology residents. However, the training program and the assessment instrument primarily focused on the initial learning of a clinical skill, which is similar for medical students and junior residents. Furthermore, high-achieving students are more likely to sign up for this kind of experiment. Thus, several of our findings may be applicable only to such learners. Moreover, learning CA was not part of the curriculum for medical students, so the aspect of having fun together could be unique to this voluntary setting. Finally, we designed our study to show a difference—not equivalence—between the two groups.50 The CI of the difference between the two groups’ MOCARS scores included the MIREDIF of 7.5 between the two groups. However, the CI was inconclusive regarding aspects of equivalence between the two groups because it included zero.

Our findings indicate that dyad practice may be beneficial in training protocols for other complex procedures, such as simulation-based laparoscopic surgery, and for low-risk procedures in the clinical setting, such as sonography.51 Future research should explore combinations of supervised and directed self-guided dyad practice. Further research also should explore combinations of dyad practice and individual practice and determine the optimal amount of dyad practice before subsequent individual practice.


The quality of the complex skills that students learned from dyad practice on a CA simulator was equal to that of the skills learned from individual practice. As dyad trainees learned as much as individual trainees during the same period, dyad practice is more cost-effective and can be employed to efficiently use costly virtual reality simulator equipment to teach complex skills.

Novices’ perceptions of how and why they learned complex, integrated cognitive and motor skills from dyad practice relate to five key themes: (1) reduction of cognitive load, (2) observational learning, (3) learning from overt communication and dialogue, (4) social aspects and motivation, and (5) meta-cognition. These findings may inspire the design of future simulation-based training protocols. However, educators should take caution as they may not apply to clinical training involving real patients, because learning from errors and overt communication, both keys to dyad practice, do not transfer to clinical practice.

Acknowledgments: The authors gratefully acknowledge the helpful advice regarding statistical analysis from Lene Theil Skovgaard (University of Copenhagen, Denmark). They also thank the participants for giving their time.


1. Ahmed K, Keeling AN, Fakhry M, et al. Role of virtual reality simulation in teaching and assessing technical skills in endovascular intervention. J Vasc Interv Radiol. 2010;21:55–66
2. Curet MJ. Resident work hour restrictions: Where are we now? J Am Coll Surg. 2008;207:767–776
3. DeMaria AN. Medicine, aviation, and simulation. J Am Coll Cardiol. 2011;57:1328–1329
4. Desender LM, Van Herzeele I, Aggarwal R, Vermassen FE, Cheshire NJ. Training with simulation versus operative room attendance. J Cardiovasc Surg (Torino). 2011;52:17–37
5. Dawson DL, Lee ES, Hedayati N, Pevec WC. Four-year experience with a regional program providing simulation-based endovascular training for vascular surgery fellows. J Surg Educ. 2009;66:330–335
6. Willaert WI, Aggarwal R, Nestel DF, et al.European Virtual Reality Endovascular Research Team, EVEResT. Patient-specific simulation for endovascular procedures: Qualitative evaluation of the development process. Int J Med Robot. 2010;6:202–210
7. Windsor JA. Role of simulation in surgical education and training. ANZ J Surg. 2009;79:127–132
8. Larsen CR, Soerensen JL, Grantcharov TP, et al. Effect of virtual reality training on laparoscopic surgery: Randomised controlled trial. BMJ. 2009;338:b1802
9. Sturm LP, Windsor JA, Cosman PH, Cregan P, Hewett PJ, Maddern GJ. A systematic review of skills transfer after surgical simulation training. Ann Surg. 2008;248:166–179
10. Coates PJ, Zealley IA, Chakraverty S. Endovascular simulator is of benefit in the acquisition of basic skills by novice operators. J Vasc Interv Radiol. 2010;21:130–134
11. Shea CH, Wulf G, Whitacre C. Enhancing training efficiency and effectiveness through the use of dyad training. J Mot Behav. 1999;31:119–125
12. Wulf G, Shea C, Lewthwaite R. Motor skill learning and performance: A review of influential factors. Med Educ. 2010;44:75–84
13. McNevin NH, Wulf G, Carlson C. Effects of attentional focus, self-control, and dyad training on motor learning: Implications for physical rehabilitation. Phys Ther. 2000;80:373–385
14. Gillebert TC, Brooks N, Fontes-Carvalho R, et al. ESC core curriculum for the general cardiologist (2013). Eur Heart J. 2013;34:2381–2411
15. Jacobs AK, Babb JD, Hirshfeld JW Jr, Holmes DR JrSociety for Cardiovascular Angiography and Interventions. . Task force 3: Training in diagnostic and interventional cardiac catheterization endorsed by the Society for Cardiovascular Angiography and Interventions. J Am Coll Cardiol. 2008;51:355–361
16. Granados C, Wulf G. Enhancing motor learning through dyad practice: Contributions of observation and dialogue. Res Q Exerc Sport. 2007;78:197–203
17. Sanchez-Ku ML, Arthur W Jr. A dyadic protocol for training complex skills: A replication using female participants. Hum Factors. 2000;42:512–520
18. Topping KJ, Ehly SW. Peer assisted learning: A framework for consultation. J Educ Psychol Consult. 2001;12:113–132
19. Topping KJ. Trends in peer learning. Educ Psychol. 2005;25:631–645
20. Janssen J, Kirschner F, Erkens G, Kirschner PA, Paas F. Making the black box of collaborative learning transparent: Combining process-oriented and cognitive load approaches. Educ Psychol Rev. 2010;22:139–154
21. Kirschner F, Paas F, Kirschner PA. A cognitive load approach to collaborative learning: United brains for complex tasks. Educ Psychol Rev. 2009;21:31–42
22. Kirschner F, Paas F, Kirschner PA. Task complexity as a driver for collaborative learning efficiency: The collective working-memory effect. Appl Cogn Psychol. 2011;25:615–624
23. van Merriënboer JJG, Sweller J. Cognitive load theory and complex learning: Recent developments and future directions. Educ Psychol Rev. 2005;17:147–177
24. Kirschner F, Paas F, Kirschner PA. Superiority of collaborative learning with complex tasks: A research note on an alternative affective explanation. Comput Hum Behav. 2011;27:53–57
25. Paas F, Tuovinen JE, van Merriënboer JJG, Darabi AA. A motivational perspective on the relation between mental effort and performance: Optimizing learner involvement in instruction. Educ Technol Res Dev. 2005;53:25–34
26. Magill RA Motor Learning and Control: Concepts and Applications. 200711th ed New York, NY McGraw-Hill
27. Kirschner PA, Ayres P, Chandler P. Contemporary cognitive load theory research: The good, the bad and the ugly. Comput Hum Behav. 2011;27:99–105
28. Brydges R, Carnahan H, Safir O, Dubrowski A. How effective is self-guided learning of clinical technical skills? It’s all about process. Med Educ. 2009;43:507–515
29. Mitchell A, West N, Lesson P, Banning A Cardiac Catheterization and Coronary Intervention. 2008 Oxford, UK Oxford University Press
30. Räder SB, Jørgensen E, Bech B, Lönn L, Ringsted CV. Use of performance curves in estimating number of procedures required to achieve proficiency in coronary angiography. Catheter Cardiovasc Interv. 2011;78:387–393
31. Borkan JCrabtree BF, Miller WL. Immersion/crystallization. Doing Qualitative Research. 19992nd ed Thousand Oaks, Calif Sage Publications
32. Downing SM, Yudkowsky R Assessment in Health Professions Education. 2009 New York, NY Routledge
33. Lipner RS, Messenger JC, Kangilaski R, et al. A technical and cognitive skills evaluation of performance in interventional cardiology procedures using medical simulation. Simul Healthc. 2010;5:65–74
34. Lynch J, Aughwane P, Hammond TM. Video games and surgical ability: A literature review. J Surg Educ. 2010;67:184–189
35. Nomura T, Miyashita M, Shrestha S, et al. Can interview prior to laparoscopic simulator training predict a trainee’s skills? J Surg Educ. 2008;65:335–339
36. Van Hove C, Perry KA, Spight DH, et al. Predictors of technical skill acquisition among resident trainees in a laparoscopic skills education program. World J Surg. 2008;32:1917–1921
37. Brydges R, Dubrowski A, Regehr G. A new concept of unsupervised learning: Directed self-guided learning in the health professions. Acad Med. 2010;85(10 suppl):S49–S55
38. Snyder CW, Vandromme MJ, Tyra SL, Porterfield JR Jr, Clements RH, Hawn MT. Effects of virtual reality simulator training method and observational learning on surgical performance. World J Surg. 2011;35:245–252
39. Carraccio CL, Benson BJ, Nixon LJ, Derstine PL. From the educational bench to the clinical bedside: Translating the Dreyfus developmental model to the learning of clinical skills. Acad Med. 2008;83:761–767
40. Wulf G, Shea CH. Principles derived from the study of simple skills do not generalize to complex skill learning. Psychon Bull Rev. 2002;9:185–211
41. Mattar AA, Gribble PL. Motor learning by observing. Neuron. 2005;46:153–160
42. Avanzino L, Giannini A, Tacchino A, Pelosin E, Ruggeri P, Bove M. Motor imagery influences the execution of repetitive finger opposition movements. Neurosci Lett. 2009;466:11–15
43. van Gog T, Rummel N. Example-based learning: Integrating cognitive and social-cognitive research perspectives. Educ Psychol Rev. 2010;22:155–174
44. Crook AE, Beier ME. When training with a partner is inferior to training alone: The importance of dyad type and interaction quality. J Exp Psychol Appl. 2010;16:335–348
45. Ploetzner R, Dillenbourg P, Preier M, Traum DDillenbourg P. Learning by explaining to oneself and to others. Collaborative Learning: Cognitive and Computational Approaches. 1999 Oxford, UK Elsevier
46. Hausmann RGM, Chi MTH, Roy M. Learning from collaborative problem solving: An analysis of three hypnotized mechanisms. Proceedings of the 26th Annual Conference of the Cognitive Science Society. 2004:547–552
47. Hausmann RGM, van de Sande B, van de sande C, VanLehn K. Productive dialog during collaborative problem solving. Proceedings of the 8th International Conference for the Learning Sciences. 2008:327–334
48. Chi MTH. Active–constructive–interactive: A conceptual framework for differentiating learning activities. Top Cogn Sci. 2009;1:73–105
49. De Ponti R, Marazzi R, Ghiringhelli S, Salerno-Uriarte JA, Calkins H, Cheng A. Superiority of simulator-based training compared with conventional training methodologies in the performance of transseptal catheterization. J Am Coll Cardiol. 2011;58:359–363
50. Piaggio G, Elbourne DR, Altman DG, Pocock SJ, Evans SJCONSORT Group. . Reporting of noninferiority and equivalence randomized trials: An extension of the CONSORT statement. JAMA. 2006;295:1152–1160
51. Norman G, Dore K, Grierson L. The minimal relationship between simulation fidelity and transfer of learning. Med Educ. 2012;46:636–647
© 2014 by the Association of American Medical Colleges