Trainees in the health professions are increasingly being exposed to a variety of unsupervised learning contexts throughout their training. While some opportunities (such as learning history-taking from a textbook or practicing suturing on an orange) have always been available, several recent factors have conspired to increase both the number of these opportunities and the sophistication of the skills and knowledge that can be learned in the unsupervised context. For example, the advent of increasingly elaborate educational technologies such as simulation, virtual patients, and various asynchronous learning tools has allowed trainees the convenience of learning on their own schedules rather than being restricted by the availability of clinical teachers. This affordance has been seen as a great advantage not only for the trainees themselves but also from a resource perspective, as it has helped to reduce the demands on clinical educators.1 Further, from an ethical perspective, these technologies have removed the need to place patients at risk when trainees are learning clinical skills, thereby decreasing the pressure on supervisors to provide direct clinical oversight2 of the learning experience. The loss of this ethical rationale for supervised learning in the traditional clinical context has allowed resource issues to become the prominent driving factor in the decision to allow trainees to learn without expert guidance. Two additional pressures arise from the theoretical domain. First, from the perspective of “learning how to learn,” an increased use of unsupervised learning may be justified on the premise that allowing trainees to “figure it out themselves” will give them practice with the lifelong learning skills they will need to maintain competence in later practice. Second, there is ample evidence in the education literature suggesting that a trainee who is involved actively in her education will benefit more than when an educator excessively controls the process.3,4 Thus, interestingly, as patient safety issues have placed increasingly heavy pressure to decrease trainee autonomy in the clinical context,2 the confluence of the various factors described above has resulted in more unsupervised, self-guided learning opportunities for trainees than ever before in the nonclinical context.
While these various pressures provide a sensible logic for a shift toward unsupervised learning in the nonclinical context, we must be cautious to ensure that this shift is not simply an unreflective drift away from supervised, educator-guided learning. As Cook5 has suggested educational research activities have long lagged behind advances in technology. As a result, technologies, rather than innovations in instruction and learning theory, often drive educational change,6,7 and this may be more true for the unsupervised, self-guided learning context than any other area in health professions education. Further, proponents of several learning theories8–10 would argue that excessive unsupervised learning is problematic and, therefore, that some level of supervision must be maintained. Indeed, there are many reasons for educators to supervise trainee learning. For example, it has been demonstrated that educators can diagnose the trainee's learning needs and focus her learning efforts more effectively.11,12 In addition, without informed corrective feedback trainees may form bad habits and incorrect judgments about their learning. Thus, while there may be good reasons to increase our reliance on unsupervised, self-guided learning for our trainees in the nonclinical context, if we are not reflective and strategic in this shift we run the risk, in the words of Lee Brooks (personal communication to GR, July 1992) of turning our curricula into the equivalent of arranging a swimming lesson by sinking a boat: Those who survive learn how to keep their heads above water, not how to swim effectively.
In an effort to become more intentional, informed, and systematic in our enactment of self-guided learning in the increasingly complex, unsupervised learning contexts to which we are exposing our trainees, this paper will review and reconsider several literatures that speak to the strengths and weaknesses of both supervisor-supported and self-guided learning. Consulting these various literatures may allow us to understand how we can embed the benefits of each into the structure of trainees' unsupervised learning activities. That is, rather than relying on the technology or the trainee to get it right, a more strategic approach to self-guided learning may be to create conditions so that, even in unsupervised settings, the educator is present through the design and structure of the learning setting. Such a process has been described as directed self-guided learning, and it requires a knowledgeable educator to design practice conditions using validated learning principles.13 A trainee then steps into this structured setting and is given limited control of a specific aspect of practice and, therefore, is metacognitively, behaviorally, and motivationally active in her learning while also receiving support in her learning decisions.3
Further research and reflection are necessary before the concept of directed self-guided learning can be considered in medical education. To this end, this targeted, nonsystematic review will focus on literatures that might inform how we can achieve a positive blend between unsupervised, self-guided learning and supervised learning. We begin by reviewing the educational psychology literature which speaks to the advantages of self-guided learning. Specifically, we explore evidence from social cognitive theory, which defines human learning as an interaction of personal, behavioral, and environmental factors14—a perspective that may find application in the medical training context. Next, the metacognition literature will be explored in an effort to understand what trainees do spontaneously when self-guiding their learning, where it will become obvious that both trainees and many current approaches to training have flaws that must be managed in order to enhance unsupervised learning. Here, metacognition researchers who have concentrated on relating metacognitive strategies to how humans learn independently will receive attention. In the last section, the advantages of supervised learning will be reexamined in light of the self-guided learning and metacognition literatures using a review of pertinent learning theories. In particular, we concentrate on learning theories that examine the ability of educators to determine a learner's needs, to challenge the learner, and to support the learner. Finally, the discussion section will elaborate on the need to reorient our questions when considering the concept of self-guided learning. Rather than asking questions concerning the effectiveness of particular technologies, we propose that our research questions focus on our understanding of trainees' natural propensities while learning in the unsupervised context and on exploring conditions that will maximize the educational benefit of self-guided learning.
Data Regarding the General Effectiveness of Self-Regulated Learning Processes
Educational psychologists who study self-guided learning tend to emphasize the positive aspects of what they refer to as self-regulated learning.14 Although there are several schools of thought in the self-regulated learning literature, there seems to be a general agreement that trainees are capable of effectively regulating their cognition, motivation, and/or behavior to achieve their learning goals.15,16 That is, researchers in this domain presume that trainees possess the cognitive tools needed to learn competently when unsupervised, but tend to use these tools suboptimally. Because it is taken as largely uncontroversial that enhancing the use of these cognitive tools will improve learning, research in this field tends to focus on the design of specific tools to help learners maximize the use of their preexisting skills. For example, many researchers identify self-monitoring, defined as an individual's awareness of his or her psychological content, as an important process in self-guided learning.16,17 Several studies of the self-monitoring process have focused on making self-monitoring activities more explicit—testing, for example, the efficacy of asking participants to keep records of their learning behaviors. Whether learning skills such as dart throwing18 or knowledge domains such as graduate-level statistics,19 participants who self-recorded details about their performance had enhanced self-efficacy and enhanced performance scores relative to “nonrecording” control groups. Thus, self-regulated learning theorists conclude that self-recording is an explicit behavior participants can use to enhance the underlying psychological process of self-monitoring, which in turn improves unsupervised, self-regulated learning. Clearly, then, learners show some capacity to become more aware of their learning and, through this increased awareness, improve the outcomes of their learning efforts.
However, researchers who explore self-guided learning in this way tend not to raise questions related to the underlying cognitive mechanisms that support the process. That is, they appear less interested in how self-monitoring works than in the fact that it does. Thus, while the work demonstrates that there are some effective, naturally occurring self-regulatory learning processes (such as self-monitoring), and that increasing the explicit use of these processes has some value, it is not clear from this literature whether it is possible to improve the processes themselves. By contrast, experimental psychologists have taken a more in-depth look at self-monitoring and other metacognitive processes when participants learn in unsupervised contexts.
Mechanisms of Self-Guided Learning: Lessons From Research on Metacognition
Flavell21 defined metacognition as “one's knowledge concerning one's own cognitive processes.” In the field of metacognition, researchers aim to understand our natural propensities when we learn at a more specific and detailed level than those working in the field of self-regulated learning. The questions relevant to self-guided learning that metacognition researchers ask include (1) How are metacognitions used during self-guided learning to influence what trainees study and for how long? (2) Do trainees shape their learning strategies to use metacognitions effectively? and (3) How might metacognitive errors undermine effective self-guided learning? In this section, the current evidence available to help answer each of those questions will be reviewed briefly, and an example that reflects a representative theoretical perspective will be explored.
One area of metacognition research has focused on understanding the processes by which learners select the specific material to be learned, that is, what learners choose to study and for how long. Consistent with the self-regulated learning literature, there is evidence to suggest that learners have some capacity to understand their learning needs. They will identify items that they already know and eliminate these from the learning set,21,22 and they will accurately sort the remaining items based on the perceived levels of difficulty involved in learning them.23 Further, there is some evidence that they can use this perception of difficulty strategically in their selection of items to learn. When under no constraints or pressures, learners tend to concentrate their initial efforts on items that they perceive as more difficult to learn in what is described as a “difficult to easy” learning schedule.23 But, when placed under time pressure, they will sequence items from easy to difficult, initially selecting what they perceive to be easy items or items that are closest (i.e., most proximal) to being learned,21,24,25 and when an explicit reward structure is present, they will tend to sequence items based on value rather than difficulty,22 again suggesting some level of strategic control. Further, there is some evidence to suggest that learners will stop studying when the rate of learning drops too low.21,25 For example, in one of few studies to address this question in the health professions, medical trainees seemed to accurately identify their own learning asymptotes when practicing one-handed knot-tying.26 That is, participants who were allowed to stop practicing at their own discretion performed as well on a later test as participants who were asked to continue practicing even after they determined that they wanted to stop. The researchers did not address why the participants decided to stop, but it appears that additional practice beyond the point at which they felt they were no longer learning did not, in fact, lead to further learning.
Despite these relatively positive findings about the value of metacognitive processes in making decisions about what to learn and how long to study, there is evidence to suggest that these processes are far from perfect when used to select strategies to maximize (or at least enhance) learning. For example, frequent retrieval practice, or self-testing, has been shown to be a potent strategy for enhancing learning27–29 even at the expense of additional study time.28,30–31 However, evidence suggests that learners will fail to incorporate self-testing strategies into their personal theories of how they learn and, under unconstrained conditions, will generally fail to self-test effectively.27 Rather, they tend to avoid situations in which they might commit an error despite the positive consequences that the error would provide (i.e., recognizing the need for more learning33). It has been suggested that such challenges are likely avoided for the purposes of maintaining one's positive self-concept14 despite the resulting loss of important learning information. Even when they do choose to self-test, learners do so suboptimally. For example, they tend to delay self-testing until long after the feedback would have been maximally efficacious for guiding learning.28 It is worth noting that when the option to self-test is more salient, participants will choose the option and will benefit from doing so.34 This suggests that educators can create simple choices that prompt participants to use strategies they would not spontaneously engage. Therefore, such interventions represent a method for enhancing the efficacy of self-guided learning.
However, even when learners do try to evaluate their performance, they often radically overestimate the level of learning they have achieved. Bjork35 attributes these illusions of competence to a trainee's misreading or misunderstanding of her progress during training. Several metacognitive errors contribute to these illusions of competence. For example, when self-testing, learners will often do so incorrectly and expose themselves to the hindsight bias. That is, they will look at the answer before properly challenging themselves with an effort to recall the item. In the presence of the answer, they will mistake recognition for the ability to recall, and the ease of recognition leads them to overestimate future recall performance. As another example, learners have a propensity to interpret the ease with which they are currently processing information as predictive of success in future processing (mistaking performance for learning).36 Each illusion of competence has at its core the participant's misunderstanding of the complexities of his own memory. Participants, it would appear, do not have an accurate mental model of how learning works,29 are often misled by short-term performance,37,38 and exhibit biased subjective assessments of ability.39 At times, participants appear to self-monitor effectively in the moment, but they often do not possess accurate theories about why particular strategies are effective. Those theories will affect the self-control policies that they adopt, and, consequently, suboptimal strategies may be employed.
How Can Expert Supervision Help Maximize Learning Outcomes?
In light of the evidence in favor of self-guided learning from the self-regulated learning literatures and the cautionary tales that arise from the metacognition literature, it may be helpful to turn our attention to some educational theories that expound the value that expert educators provide for their trainees. Among the many positive contributions made by expert educators, three are particularly relevant to the issues raised in the sections above: their ability to identify the learner's skill level and set the level of challenge appropriately, their ability to challenge the learner in ways in which he would not challenge himself in order to ensure that he does not fall prey to illusions of competence, and their ability to support the learner in those difficult conditions. We will elaborate on each briefly.
Clinical educators are often content experts who are able to effectively analyze a task or problem that they must teach to others and then create a plan for how to teach the material to trainees at various levels of knowledge and ability. In addition to this form of task analysis, however, the expert educator must also be able to “diagnose” each individual learner's level of understanding and ability.11,40,41 Combining these skills of task- and trainee analysis, clinical educators can tailor instruction and training interventions to each trainee's individual needs, ensuring that the trainee is being appropriately challenged. Practice without challenge can actually be detrimental to long-term learning.27,33,35 Indeed, errors may be more probable over the long term if they are not induced during training.33 So, a key role of the educator is to induce such challenges and errors appropriately in the learner. In the kinesiology literature, the Challenge Point Framework10 concentrates on how educators can effectively challenge trainees with the appropriate level of task difficulty. Guadagnoli and Lee10 describe “functional difficulty” as the interaction between the task's inherent difficulty, the skill level of the trainee, and the environmental conditions in which the task is performed. Importantly, this implies that functional difficulty is a dynamic construct and must be regularly modified so that it will maximize the potential for learning. As a trainee's learning curve plateaus, educators can create a new learning challenge by increasing the functional task difficulty, thereby maintaining the optimal challenge point for the trainee. A similar procedure is recommended in the expertise literature where an educator or coach designs “deliberate practice” activities to improve the performance of a particular trainee.12
Challenging learners effectively, however, is not only a matter of ensuring that they learn continuously at the right level of challenge. Excellent teachers must also challenge the learner in ways that he might not challenge himself because of the metacognitive illusions of competence. To this end, educators must often use training measures that likely appear counterintuitive and even irrational from the perspective of the metacognitively self-monitoring learner. Bjork,35 for example, coined the term “desirable difficulties” to describe manipulations that introduce challenges and difficulties for the trainee that may depress immediate training performance but which typically improve long-term performance (i.e., learning). Across the verbal and motor learning literatures, this phenomenon has earned the label “the performance-learning paradox.”38 Examples of this paradox, manipulations that improve long-term learning at the cost of immediate performance, include the use of random versus blocked practice,42 the use of learning time for scheduled self-tests versus continued study,43 and the provision of summary versus concurrent feedback.44 In each example, the latter intervention leads to better short-term performance during practice, but the first option leads to better long-term learning, likely because the trainee must use encoding processes in practice that are representative of those that the posttraining (e.g., clinical) environment will demand.39 Based on the psychology and kinesiology literatures, then, the best approach to enhance health professions' curricula is to structure the conditions of training optimally. This need for improved structure of training exists not just in the context of traditional training programs but also in modern settings that incorporate educational technologies such as computer-assisted instruction.39 In this era of remodeling curricula, it seems necessary that educators and researchers probe the benefits of desirable difficulties which on the surface appear irrational but, in practice, serve as critical factors that can maximize learning outcomes.
Finally, it is important to remember that the good teacher not only challenges and pushes learners but also supports learners in their efforts to rise to those learning challenges. This supportive role of the educator is highlighted nicely in Vygotsky's8 social constructivist model of education. Vygostky's Zone of Proximal Development is defined as “the phase in development in which the [trainee] has only partially mastered a task but can participate in its execution with the assistance and supervision of an adult or more capable peer.”45 This definition serves to remind us that the optimal challenge point may lie in the range of functional difficulty where a trainee is able to complete a task with the assistance of others, rather than in the range of difficulty where he or she can achieve the task alone.8 This more socially constructed model of the learning process, then, defines expert teaching not only as the process of setting the parameters of the task to keep challenging the learner but also as identifying and responding to those moments when the learner requires support in order to address that challenge.46
Collectively, these three roles of the educator, described in light of the literature on self-regulated learning and metacognitive errors, set an agenda for enhancing the value of the unsupervised learning contexts that are proliferating in our health professional education milieus. In the discussion, we will elaborate on how we might incorporate structural supports into the educational technologies being used by our trainees that replicate those educator roles.
The recent trend toward providing self-guided learning opportunities during the initial stages of health professions education has been motivated by technological, economic, ethical, and theoretical pressures. Self-guided learning requires the trainee to determine his learning needs, formulate learning goals, identify learning resources, select and employ adequate strategies, and evaluate learning outcomes. With so many factors to consider, we should not be surprised by data which suggest that trainees have difficulty coordinating self-guided learning. Indeed, those data would appear to give credence to the value of expert educators in traditional apprenticeship training. An interesting discrepancy arises here when a trainee's opportunities to self-guide his learning in the clinical training context are contrasted with his opportunities in the unsupervised learning contexts we have highlighted in this review. Increasingly, evidence is accumulating which suggests that residents desire less direct supervision than clinical supervisors would like to provide.47 Confusion may arise when we juxtapose residents' sense of “oversupervision” in the clinical context with the thesis of this review: that trainees are increasingly being exposed to unsupervised learning contexts. To clarify, we believe that the issue of supervision has become heavily context-specific; patient safety concerns are driving the increase in supervision in the clinical context,2 whereas once those patient safety concerns are relaxed (i.e., in unsupervised learning contexts with technologies), the drive for greater supervision is diminished.
While certain contexts demand educator supervision to help ameliorate the potential negative outcomes associated with trainee learning and patient safety, the reality is that trainees will continue to learn clinical skills and knowledge on their own. Further, self-guided learning has received theoretical support because it can lead to better learning when trainees understand how to optimize training and tailor it to their unique needs. Trainee motivation may also increase when they are active and autonomous in their learning decisions. Given these divergent pressures and preferences surrounding how to reconcile the concept of self-guided learning with the desire for clinical supervision, we suggest a need for the health professions education community to reorient our thinking on this topic.
During formal educational activities in which trainees self-guide their learning, the environment should be structured with safeguards present that direct trainees down the path to competence rather than the path to incompetence. Hence, we suggest a modification of the concept of self-guided learning as learning that occurs while the trainee is unsupervised, but not necessarily undirected. Using an approach that emphasizes the interaction between the trainee and his environment,3,8,10 educators can involve themselves in a trainee's self-guided learning while maintaining the trainee's feeling of autonomy. Rather than being present physically, the educator can be present through thoughtful manipulations of the trainee's self-guided learning environment. Rather than assuming the trainee will make effective learning decisions, manipulations may be designed to take advantage of his natural propensities when learning on his own and direct his decisions toward effective learning activities. An educator's use of learning theory to influence instructional design for unsupervised contexts has been called directed self-guided learning 13 and finds support in various educational concepts like the zone of proximal development8 and scaffolding.41 Importantly, vigilance is needed to avoid too much director involvement, which may prevent the educational activity from flexibly meeting trainees' needs and may also limit trainee motivation. Therefore, the directed self-guided learning model requires reflective thinking by both the educator and the trainee to ensure the activity is optimal. To test the quality of this new conceptualization of self-guided learning, there is a need for research that adds to our understanding of how the different phenomena outlined in this review manifest in the context of health professions education. Supplementary work will need to evaluate how we can redress certain issues given the affordances and constraints inherent to our training contexts.
Even if the formal self-guided learning training context is optimized, the metacognitive errors literature39 provides evidence that trainees may not develop habits of learning because certain learning experiences do not “sink in” at a metacognitive level. An iconic example is the medical trainee who does not request clinical support from her supervisor in a timely manner. She may avoid seeking help because of metacognitive errors such as not being able to recognize when she needs support.39,48 Her decision to seek clinical support is also complicated because she is influenced by the nature of the clinical question and by factors related to herself and her supervisor. Kennedy et al48 have shown that the trainee perceives risks to her professional credibility in contexts where, for example, seeking support may be associated with a question that she believes falls within her expected scope of practice, may affect how her performance is evaluated, and/or may aggravate a supervisor she feels is not approachable. This example illustrates that the questions of how best to manage the push and pull of tensions between trainees and educators in a directed self-guided learning system can be expanded to daily clinical practice and informal instances of self-guided learning. Therefore, the links between self-monitoring, self-guided study, learning outcomes, and changes in trainees' subjective beliefs and motivation deserve full exploration when trainees self-guide their learning during formal curricular experiences and informal daily situations. Researchers and educators alike should begin creating and testing interventions that aim to enhance objective and subjective learning such that trainees perform well, manipulate their training environment optimally in the moment, and carry forward those lessons to future informal self-guided learning episodes.
If we consider self-guided learning in this light, different research questions arise. The first example of a potential program of research is the study of which cues have the greatest influences over metacognitive judgments of learning when trainees learn clinical skills and knowledge. Specific questions that can be addressed include which cues are attended to and why, what is the validity of cues in predicting future performance, when and how do trainees use certain cues, and how do the products of metacognitive judgments, accurate or inaccurate, influence subsequent behavior and the management of educational resources. Study of the cues and heuristics used by health professions trainees will permit a better understanding of the mechanisms underpinning self-monitoring and self-guided learning.
A second research program could focus on the alignment of trainees' subjective and objective learning curves as they acquire clinical knowledge or a particular skill set. Commonly used objective measures of performance, such as hand motion analysis49,50 and expert-based assessments,51 can be compared with subjective indices such as trainee judgments of how well one has learned the material, judgments of one's rate of learning,21 and judgments of perceived increases in effort.52 Such a study would be novel because, in previous work, subjective assessments of learning are often summative, and trainees are not asked to assess their learning in the moment. The resulting data could be used to help educators understand common illusions of competence and would also provide hints on how to diagnose and address such biases.
Third, a potential research program could examine how trainees allocate their study time spontaneously in self-guided learning conditions. Various variables can be considered: Under time pressure, do trainees always pursue the easiest items in an effort to “get by”? Does reward-driven behavior always arise when different incentives are distributed among to-be-learned items? What implications would the expectedly different study behaviors have for learning outcomes and/or patient safety?
A fourth research program could have as its focus which desirable difficulties are effective in our training contexts. Considering the increased use of educational technologies, a good approach will be to test the utility of desirable difficulties with these nontraditional modes of learning. This line of research would expand the theoretical framework of how to challenge trainees and would also generate evidence that adds to our understanding of how and why desirable difficulties and other challenging conditions influence learning.
Finally, a sociological research program could explore how the dynamic relationship between a trainee, her environment, and the educator affects her sense of feeling supported in her self-guided learning. Though variables such as motivation and self-efficacy have not been addressed fully in the present review, a trainee's personal perceptions will certainly impact the quality and quantity of learning in self-guided learning contexts. Given that the educational technology boom is now well entrenched in curricula in the health professions, simulation laboratories and other nontraditional testing grounds would serve as good initial contexts to enhance our understanding of the processes of self-monitoring, self-control, and self-guided learning. However, education researchers must carefully consider the extent to which self-guided learning in the simulation setting can be likened and translated to unsupervised clinical practice.53
The new conceptualization of self-guided learning offered in this review requires educators to recognize that self-monitoring is context- and content-specific.54,55 This specificity means that the particular clinical domain the trainee is learning (procedural skills versus physical examination skills versus diagnostic reasoning, etc.) will impact and determine the best training approach. Consequently, training regimes will have to be constructed independently and iteratively with the specific context in mind, be it simulation, Web-based modules, or clinical training experiences. Recently, attention has been paid to the models of supervision that can help trainees practice medicine autonomously and effectively.53 Comparatively less attention has been placed on maximizing the informal opportunities that trainees have to practice and modify those same lifelong learning skills, though there are exceptions.56 Self-monitoring and self-guided learning should not be framed as individualistic activities; external resources are necessary for the trainee to experience the greatest educational benefit. Considering self-guided learning from this perspective may help health professions education researchers to reorient our research efforts toward a goal of determining how to take advantage of trainees' natural propensities, both good and bad, so that we maximize self-guided learning opportunities.
The authors would like to thank Dr. Heather Carnahan and Dr. Lorelei Lingard for their thoughtful comments on earlier versions of this review paper.
RB is supported by a postdoctoral fellowship from the Natural Sciences and Engineering Research Council (NSERC) and a postdoctoral fellowship from the Centre for Health Education Scholarship, University of British Columbia. AD is supported by a Discovery Grant from the Natural Sciences and Engineering Research Council (NSERC).
2Kennedy TJ, Lingard L, Baker GR, Kitchen L, Regehr G. Clinical oversight: Conceptualizing the relationship between supervision and safety. J Gen Intern Med. 2007;22:1080–1085.
3Zimmerman BJ. A social cognitive view of self-regulated academic learning. J Educ Psychol. 1989;81:329–339.
4Knowles MS. Self-Directed Learning: A Guide for Learners and Teachers. New York, NY: New York Association Press; 1975.
5Cook DA. The failure of e-learning research to inform educational practice, and what we can do about it. Med Teach. 2009;31:158–162.
6Kneebone R. Evaluating clinical simulations for learning procedural skills: A theory-based approach. Acad Med. 2005;80:549–553.
7Winne PH, Stockley DB. Computing technologies as sites for developing self-regulated learning. In: Schunk DH, Zimmerman BJ, eds. Self-Regulated Learning: From Teaching to Self-Reflective Practice. New York, NY: The Guilford Press; 1998:106–137.
8Vygotsky LS. Mind in Society: The Development of Higher Psychological Processes. Oxford, England: Harvard U Press; 1978.
9Bruner JS. The Culture of Education. Cambridge, Mass: Harvard University Press; 1996.
10Guadagnoli M, Lee T. Challenge point: A framework for conceptualizing the effects of various practice conditions in motor learning. J Mot Behav. 2004;36:212–224.
11Irby DM. How attending physicians make instructional decisions when conducting teaching rounds. Acad Med. 1992;67:630–638.
12Ericsson KA. An expert-performance perspective of research on medical expertise: The study of clinical performance. Med Educ. 2007;41:1124–1130.
13Brydges 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.
14Bandura A. Social Foundations of Thoughts and Action: A Social Cognitive Theory. Englewood Cliffs, NJ: Prentice-Hall; 1986.
15Hofer BK, Yu SL, Pintrich PR. Teaching college students to be self-regulated learners. In: Schunk DH, Zimmerman BJ, eds. Self-Regulated Learning: From Teaching to Self-Reflective Practice. New York, NY: Guilford Publications; 1998:57–85.
16Schunk DH. Social cognitive theory and self-regulated learning. In: Zimmerman BJ, Schunk DH, eds. Self-Regulated Learning and Academic Achievement: Theoretical Perspectives. 2nd ed. Mahwah, NJ: Lawrence Erlbaum Associates Publishers; 2001:125–151.
17Zimmerman BJ. Self-efficacy: An essential motive to learn. Contemp Educ Psychol. 2000;25:82–91.
18Zimmerman BJ, Kitsantas A. Self-regulated learning of a motoric skill: The role of goal setting and self-monitoring. J Appl Sport Psychol. 1996;8:60–75.
19Lan WY. The effects of self-monitoring on students' course performance, use of learning strategies, attitude, self-judgment ability, and knowledge representation. J Exp Educ. 1996;64:101–115.
20Flavell JH. Metacognitive aspects of problem solving. In: Resnick LB, ed. The Nature of Intelligence. New Jersey, NJ: Lawrence Erlbaum; 1976.
21Metcalfe J, Kornell N. A region of proximal learning model of study time allocation. J Mem Lang. 2005;52:463–477.
22Ariel R, Dunlosky J, Bailey H. Agenda-based regulation of study-time allocation: When agendas override item-based monitoring. J Exp Psychol Gen. 2009;138:432–447.
23Metcalfe J. Is study time allocated selectively to a region of proximal learning? J Exp Psychol Gen. 2002;131:349–363.
24Metcalfe J, Kornell N. The dynamics of learning and allocation of study time to a region of proximal learning. J Exp Psychol Gen. 2003;132:530–542.
25Kornell N, Metcalfe J. Study efficacy and the region of proximal learning framework. J Exp Psychol Learn Mem Cogn. 2006;32:609–622.
26Jowett N, LeBlanc V, Xeroulis G, MacRae H, Dubrowski A. Surgical skill acquisition with self-directed practice using computer-based video training. Am J Surg. 2007;193:237–242.
27Karpicke JD. Metacognitive control and strategy selection: Deciding to practice retrieval during learning. J Exp Psychol Gen. 2009;138:469–486.
28Karpicke JD, Roediger HL 3rd. The critical importance of retrieval for learning. Science. 2008;319:966–968.
29Kornell N, Bjork RA. The promise and perils of self-regulated study. Psychon Bull Rev. 2007;14:219–224.
30Baddeley AD, Longman DJA. The influence of length and frequency on training sessions on the rate of learning to type. Ergonomics. 1978;21:627–635.
31Simon DA, Bjork RA. Models of performance in learning multisegment movement tasks: Consequences for acquisition, retention, and judgments of learning. J Exp Psychol Appl. 2002;8:222–232.
32Kornell N, Bjork RA. Optimising self-regulated study: The benefits—and costs—of dropping flashcards. Memory. 2008;16:125–136.
33Eva KW. Diagnostic error in medical education: Where wrongs can make rights. Adv Health Sci Educ Theory Pract. 2009;14(suppl 1):71–81.
34Kornell N, Son LK. Learners' choices and beliefs about self-testing. Memory. 2009;17:493–501.
35Bjork RA. Memory and metamemory considerations in the training of human beings. In: Metcalfe J, Shimamura A, eds. Metacognition: Knowing About Knowing. Cambridge, Mass: MIT Press; 1994:185–205.
36Koriat A, Bjork RA. Illusions of competence in monitoring one's knowledge during study. J Exp Psychol Learn Mem Cogn. 2005;31:187–194.
37Nelson TO, Leonesio RJ. Allocation of self-paced study time and the labor-in-vain effect. J Exp Psychol Learn Mem Cogn. 1988;14:676–686.
38Schmidt RA, Bjork RA. New conceptualizations of practice—Common principles in 3 paradigms suggest new concepts for training. Psychol Sci. 1992;3:207–217.
39Bjork RA. Assessing our own competence: Heuristics and illusions. In: Gopher D, Koriat A, eds. Attention and Performance XVII—Cognitive Regulation of Performance: Interaction of Theory and Application. Cambridge, Mass: MIT Press; 1999:435–459.
40Irby DM. What clinical teachers in medicine need to know. Acad Med. 1994;69:333–342.
41Wood D, Bruner JS, Ross G. The role of tutoring in problem solving. J Child Psychol Psychiatry. 1976;17:89–100.
42Brydges R, Carnahan H, Backstein D, Dubrowski A. Application of motor learning principles to complex surgical tasks: Searching for the optimal practice schedule. J Mot Behav. 2007;39:40–48.
43Roediger HL, Karpicke JD. Test-enhanced learning: Taking memory tests improves long-term retention. Psychol Sci. 2006;17:249–255.
44Xeroulis GJ, Park J, Moulton CA, Reznick RK, Leblanc V, Dubrowski A. Teaching suturing and knot-tying skills to medical students: A randomized controlled study comparing computer-based video instruction and (concurrent and summary) expert feedback. Surgery. 2007;141:442–449.
45Wertsch JV, Rogoff B. Editor's notes. In: Rogoff B, Wertsch JV, eds. Children's Learning in the “Zone of Proximal Development.” San Francisco, Calif: Jossey-Bass Inc.; 1984:1–6.
46Tharp RG, Gallimore R. A theory of teaching as assisted performance. In: Rousing Minds to Life: Teaching, Learning, and Schooling in Social Context. New York, NY: Cambridge University Press; 1988:27–43.
47Farnan JM, Johnson JK, Meltzer DO, Humphrey HJ, Arora VM. On-call supervision and resident autonomy: From micromanager to absentee attending. Am J Med. 2009;122:784–788.
48Kennedy TJ, Regehr G, Baker GR, Lingard L. Preserving professional credibility: Grounded theory study of medical trainees' requests for clinical support. BMJ. 2009;338:b128.
49Brydges R, Classen R, Larmer J, Xeroulis G, Dubrowski A. Computer-assisted assessment of one-handed knot tying skills performed within various contexts: A construct validity study. Am J Surg. 2006;192:109–113.
50Datta V, Mackay S, Mandalia M, Darzi A. The use of electromagnetic motion tracking analysis to objectively measure open surgical skill in the laboratory-based model. J Am Coll Surg. 2001;193:479–485.
51Martin JA, Regehr G, Reznick R, et al. Objective structured assessment of technical skill (OSATS) for surgical residents. Br J Surg. 1997;84:273–278.
52Koriat A, Ma'ayan H, Nussinson R. The intricate relationships between monitoring and control in metacognition: Lessons for the cause-and-effect relation between subjective experience and behavior. J Exp Psychol Gen. 2006;135:36–69.
53Kennedy TJ, Regehr G, Baker GR, Lingard LA. Progressive independence in clinical training: A tradition worth defending? Acad Med. 2005;80(10 suppl):S106–S111.
54Eva KW, Regehr G. Self-assessment in the health professions: A reformulation and research agenda. Acad Med. 2005;80(10 suppl):S46–S54.
55Son LK, Metcalfe J. Metacognitive and control strategies in study-time allocation. J Exp Psychol Learn Mem Cogn. 2000;26:204–221.
56Evensen DH, Salisbury-Glennon JD, Glenn J. A qualitative study of six medical students in a problem-based curriculum: Toward a situated model of self-regulation. J Educ Psychol. 2001;93:659–676.