In this article, we review 8 concepts from the literature on learning and performance improvement. These concepts were chosen because they have been empirically shown to influence teaching and learning. These concepts are grouped into those most relevant to educators (cognitive load theory, constructivism, and analogical transfer; Table 1) and those most relevant to learners (goal orientation, metacognition, retrieval, spaced learning, and deliberate practice; Table 2). Despite this grouping, all of the concepts in this article are valuable to both educators and learners. It is also important to note that the study of human learning is a very large field with an immense literature, and our review does not address many important issues. Accordingly, this article is a simple primer for those interested in improving education and, thereby, patient care. We will begin the discussion of each concept with a concrete scenario. These scenarios provide practical examples that relate to the subsequent educational theory discussed below each scenario.
Cognitive Load Theory
Cognitive load scenario: During a busy day in the operating rooms, you supervise a junior resident for his first spinal anesthetic. You provide numerous tips to help the resident with the spinal. You show him how to identify the correct spinal level, how to stabilize the needle when attaching the syringe, and how to slowly inject the isobaric solution. The next day you are working with the same junior resident and encounter another patient requiring spinal anesthesia. Despite numerous opportunities to take advantage of the tips you provided the day before, you are surprised when the resident does not use any of them. What happened?
Working memory is responsible for actively processing and manipulating information in real time. It is central to the management of complex cognitive tasks such as reasoning and comprehension.4 Working memory receives inputs from both the sensory systems (sight, touch, and hearing) and long-term memory. It integrates the new information and experiences with what is already understood. The resulting new understanding is then sent back to long-term memory.5 However, working memory has severe limits on its capacity and is the major bottleneck in the cognitive processing of new information.
Cognitive load theory maintains that it is very easy to overwhelm a learner’s limited working memory capacity because of the cognitive processing that is required to work through a new task or understand a new concept. When this happens it is impossible for meaningful learning to occur.6 The theory posits that working memory is subject to 3 different types of cognitive loads: intrinsic, extraneous, and germane.7,8 Each of these can be manipulated to enhance learning.
Intrinsic load is the cognitive load related to the inherent difficulty of a task. For example, basic arithmetic has a low intrinsic load, whereas differential equations have a much higher intrinsic load. The intrinsic load associated with a particular task is fixed. Often, the educator can help the learner by breaking down learning tasks into smaller subtasks, each with more manageable intrinsic load.9 These components can eventually be integrated for a summative understanding of the overall complex construct.
Extraneous load is information that requires cognitive processing, but which is not central to the learning objective. Extraneous load can arise as a consequence of the technique by which information is presented to a learner, or as a result of other irrelevant cognitive activities that occur while thinking. An educator who presents nonessential details when explaining a complex concept to a learner will increase extraneous load and consume the learner’s limited working memory that is needed for processing the intrinsic and germane load of the problem (see below). Ironically, the text “see below” in the prior sentence is a form of extraneous load because it interrupts the reader’s processing of the text and asks them to look elsewhere in the document to understand the material. To the extent possible, educators should minimize extraneous load when designing instructional methods.9 By minimizing extraneous load, more of the learner’s limited working memory capacity can be devoted to processing elements needed for problem solving, reasoning, decision making, and learning.
Germane load refers to the aspect of working memory that focuses on making sense of information (learning). Germane load is the part of cognition that leads to recognizing relationships and creating frameworks that represent more generic problem structures. A general framework of a problem and its solution is called a schema. For example, the Seldinger technique (advancing a guidewire into a lumen and then threading a device over the guidewire) is a conceptual schema that can be applied to multiple problems, such as percutaneous vascular access, percutaneous endoscopic gastrostomy tube placement, or endotracheal intubation over a bougie. Formation of schemata is critical in improving a learner’s understanding and in developing the ability to apply information to future scenarios. The working memory used to recognize the similarities in deeper structure that exists between related problems is an example of germane load. There is a strain placed on working memory when related examples of a problem are compared and contrasted to reveal similarities and differences between examples. This exercise introduces germane load. This increased demand on working memory is acceptable when the goal is to augment the learner’s understanding. Like extraneous load and intrinsic load, the contribution of germane load to the overall cognitive demands of a learning encounter is modifiable by the educator.7 Instructional design that allows for effective processing of germane load results in more successful formation of schemata. Over time, through effective use of germane load and promotion of schema creation, learning can actually reduce the intrinsic load of a problem by simplifying or encapsulating the concept.7 When conceptual learning is the goal, it is essential to effectively use germane load to develop a deeper understanding of the material by the learner.10
An awareness and understanding of these components of cognitive load theory can help improve an educator’s instructional design.5,11 For example, when the intrinsic load of a learning objective is high, educators who are aware of cognitive load theory can help the learner by dividing the material into smaller components, each having less intrinsic load. In this manner, a savvy educator tasked with teaching about hemodynamics and acute hemorrhage might choose to review stressed blood volume and mean systemic pressure on day 1 and wait until day 2 before discussing the body’s reflexive responses to acute hypovolemia. As another example, educators who are unaware of the detrimental effects of unnecessary extrinsic load may choose to use a verbal approach to explain an inherently graphical concept. Unfortunately, this requires the learner to use additional working memory to convert the words to a mental image. The educator who understands cognitive load theory can remove this unnecessary extraneous load by using a visual example to demonstrate the graphical concept.
Multitasking: An Additional Burden Placed on Working Memory
An unappreciated threat to all forms of cognitive load and therefore to education and learning is multitasking. Undertaking >1 task at the same time, or multitasking, is a method that is sometimes used to improve efficiency. In reality, however, it burdens working memory and slows cognitive function. “Multitasking” is a misnomer. In actuality, humans engage in rapid switching from one task to another. This switching process compromises the efficiency of cognitive processes. By trying to attend to multiple tasks at once, multitasking divides the limited capacity of working memory among the tasks, allotting each task less cognitive bandwidth. Rapid switching between tasks also introduces another type of extraneous cognitive load: switch cost. Switch cost is the additional cognitive burden necessary to change focus from one subject to another.12 Switching between even straightforward tasks has been shown to increase reaction time and error rate.13 Because switching limits the amount of working memory available for processing germane and intrinsic load, multitasking should be avoided in learning environments and in situations that require the timely and accurate completion of a task.
Attempting to complete both physically and mentally demanding tasks concurrently is another form of multitasking. Surprisingly, even modest physical tasks can add to cognitive load. For example, when a subject is required to learn a list of words while concurrently walking along a path (compared with being seated), their ability to later perform simple word recall is significantly reduced.14 Walking along an oval pathway diminished word recall by 17% for middle-aged adults compared with when these same subjects were seated. As the demands of the physical task intensified (i.e., the geometry of the walking path increases in complexity), performance deteriorated even further (32%).
Automaticity: Easing Cognitive Load and Lowering Demands on Working Memory
Multitasking cannot always be avoided. In the practice of anesthesia, it is often necessary to divide one’s attention among multiple demands. Experts in a field are more successful at multitasking than are trainees. The proficiency of experts over novices in multitasking is attributed to automaticity.15 Automaticity, or automatic processing, occurs when an input directly prompts the execution of a learned sequence of actions that are stored in long-term memory, bypassing working memory altogether. Because automatic processing does not use working memory, it proceeds without consuming the subject’s attention.16,17 In contrast, controlled processing refers to the active management of cognitive tasks that require the subject’s attention. Controlled processing uses working memory to accomplish tasks and is therefore subject to working memory capacity limitations. As might be expected, experienced anesthesiologists incorporate much more automatic processing than do novices, thus affording them more available working memory. A novice’s reliance on controlled processing for even basic tasks results in a significant reduction in available working memory that negatively impacts multitasking. As a novice gains experience and tasks become more automated, the increased use of automatic processing subsequently increases the amount of available working memory. This working memory can then be used to attend more cognitively demanding tasks.
Automaticity develops through continued practice, although not all tasks can or should be automated. Ideal tasks for automatic processing are those in which there is a consistent relationship between the stimulus and the response. Tasks in which a single input can result in a number of different possible responses are not suitable for automatic processing because the decision regarding which response to use requires higher cognitive processing. As an example, imagine that an anesthesiologist observes a sudden decrease in an intubated patient’s oxygen saturation. The anesthesiologist immediately places the patient on 100% oxygen and mentally reviews potential causes of hypoxia. On reflection, the anesthesiologist recognizes that the patient’s head was moved for positioning after intubation, correctly identifies a bronchial intubation as the cause of the problem, and adjusts the endotracheal tube’s position accordingly. In this example, automaticity was used during the initial response to desaturation. Automaticity triggered the use of 100% oxygen, because there is a strong and consistent relationship between placing the patient on 100% oxygen and improving the saturation. However, subsequent treatment steps differ depending on the root cause of the hypoxia (mucous plug, disconnected circuit, esophageal intubation, etc.). This varied relationship is best managed through controlled processing because the necessary decision making requires higher cognitive processing and the use of working memory. In short, adjusting the endotracheal tube position is not always the correct next step in managing desaturation.
Educators can help learners recognize processes with consistent relationships and encourage extensive practice in those areas. This will result in automaticity that will free working memory for use on more complex task management. This concept was effectively applied to first-year surgical residents who had been randomly assigned to receive training on suturing. The group that received previous training with suturing showed better intraoperative learning of new educational material than did the group that did not receive previous suturing training.18 This occurred because residents with previous suturing training were able to devote more working memory to understanding what the attending surgeon was teaching during the intraoperative period because the process of suturing had become more automatic.18 The residents who had not received previous suturing training were less able to attend to the intraoperative teaching because they needed to use precious working memory capacity to perform their suturing task. Thus, automaticity provides a latent opportunity for improvement. In other words, automaticity frees working memory capacity that can be used for more complex thinking, planning, deciding, and learning. As demonstrated by the surgical residents who learned more in the intraoperative setting once their suturing skills were somewhat automated, automaticity can result in better overall performance when the learner takes advantage of it in a manner that enables improvements in learning.18
A better understanding of cognitive load theory allows educators to be more effective in teaching. Recognition that working memory acts as a bottleneck for learning is critical. The ability to identify the different types of loads that feed into working memory is useful for educators because it helps them modify their instructional design to minimize unnecessary loads. Further, an appreciation of the detrimental effects of multitasking on cognitive load and the potential benefits of automatic processing allows educators to guide their students toward more efficient and effective learning.
The resident performing the spinal in our initial scenario suffered from cognitive overload. As a novice, his working memory capacity was saturated simply executing the basic steps of placing a spinal. The resident was possibly too cognitively occupied just figuring out which of the numerous vials in the kit contained the local anesthetic for skin anesthesia to internalize the educator’s advice. Thus, when the attending was providing important tips, the learner was not able to cognitively process them because his working memory was fully occupied with other more basic tasks. As a result, the learner did not integrate the tips into his schema for how to place a spinal. Previous review of the kit’s contents and/or previous rehearsal using a spinal simulator may have yielded a better learning outcome. In an analogous manner, if the educator wants to provide feedback to a learner who is under high cognitive load, the educator should have the learner disengage (stop) from performing the task at hand to allow the learner to completely focus on the educator while the feedback is being provided. This will enhance the learner’s ability to cognitively process the feedback in an effective manner.
Constructivism scenario: You are asked to speak to second-year residents about pH management during deep hypothermic circulatory arrest. You work diligently to prepare a lecture reviewing the most recent literature and the theoretical rational behind α-stat and pH-stat management. After the lecture, you leave time for questions. When the residents ask no questions, you feel confident that the presentation was effective. You are surprised and disappointed a few weeks later when you receive your written evaluations. You read comments such as “I found the concepts difficult to understand” and “I felt lost after the first 10 minutes.” Why wasn’t your presentation as effective as you had hoped? How could you prevent this in future lectures?
Learners come to an educational encounter preconditioned by their experiences, intuition, and previous understandings. These are the learner’s preconceptions.19 Preconceptions that are consistent with the material being taught are called anchoring conceptions, and these benefit the learner because they act as a correct foundation on which to build additional understanding. Preconceptions that are inconsistent with a new viewpoint are called misconceptions or alternative conceptions,20 and these typically interfere with learning new concepts or information. Misconceptions often represent a faulty foundation that needs correction before new knowledge may be appropriately integrated.
Constructivism is an educational approach that focuses on how learners process knowledge.21 Constructivism asserts that new knowledge is integrated with a learner’s preconceptions in 1 of 2 ways: assimilation or accommodation.22 Assimilation occurs when the learner incorporates new knowledge seamlessly into a preexisting framework of understanding. With assimilation, new information builds on, rather than disrupts, the learner’s preconceptions. In contrast, when new information challenges a learner’s preconceptions, the learner is required to reframe his or her mental representation through a process called accommodation. Accommodation results in cognitive dissonance, which is initially uncomfortable; however, accommodation is the method by which a misunderstanding can ultimately result in learning.
Unfortunately, having a misconception and then being presented with a correct representation of the concept does not guarantee that appropriate accommodation will occur. For example, a learner may correctly recognize the need to restructure a framework on a particular topic, yet incorrectly perform the restructuring. Learners may also mistakenly view new and conflicting information as erroneous, or an exception to the rule, and fail altogether to recognize that it is their own current framework that is incorrect and in need of adjustment.
It is through the use of dialogue that a constructivist educator discerns the learner’s preconceptions and works to prevent these maladaptations. The constructivist then customizes the educational experience to help the learner integrate the new information that does not currently fit the learner’s preexisting model. An educator who spends time to discover a learner’s previous knowledge base and preconceptions will then have information that can be used to deliver optimal teaching and drive better learning. Accurate determination of a student’s previous conceptions can be quite difficult without such a discussion. Even experienced teachers often presume their students have a level of understanding that is higher than it actually is.23
Constructivist educators frequently use active discovery. Learners build knowledge when they realize that new information conflicts with their current frameworks.24 Thus, constructivists favor exercises that challenge a learner’s current perspective or require the learner to explain their rationale. A recent meta-analysis found that teaching focused on active learning significantly improved performance on standardized tests and decreased failure rates when compared with traditional teaching by lecture.25 When constructivism was used to teach college physics through student collaboration and an interactive-engagement method, the students exposed to the constructivist method outperformed the students from the traditional method of teaching (lecture) with an effect size of 2.26 Thus constructivism is a powerful and effective teaching method.27,28 An example of the constructivist method of teaching is found in Box 1.
Box 1. An Example of a Constructivist Method of Educating Cited Here...
A medical student is observing a mitral valve repair. After the repair has been completed, the surgeon asks that the patient’s pacemaker rate be increased from 60 to 90 bpm. The attending anesthesiologist asks the medical student what will happen to the cardiac output when the heart rate is increased. From his classroom experience, the medical student knows that cardiac output is the product of the heart rate and the stroke volume. The student states that increasing the heart rate by 50% will also increase the cardiac output by 50%. When asked for his rationale, the medical student recites the formula. The attending then increases the rate of the cardiac pacemaker, and the student observes that the measured cardiac output hardly changes. When asked to explain the discrepancy between what he proposed would happen and what actually happened, the student reflects and then correctly states that the stroke volume must not be an independent variable as he had assumed, and thus stroke volume may vary as a function of the heart rate.
In constructivism, the educator, through dialogue, acts as a diagnostician of the learner’s preconceptions. If the learner demonstrates anchoring conceptions (i.e., they correctly understand fundamental concepts and their significance), then education is seamless because knowledge is built from a suitable base. However, more commonly, there are misconceptions and the educator assists the learner to recognize these errors through exercises that promote cognitive dissonance. Cognitive dissonance is the mentally uncomfortable state that occurs when a person realizes new information they are being asked to learn actually conflicts with what they already believe to be true. The dissonance causes learners to either change their minds to accept the new information or to reject it, which then allows them to retain their previous belief. Ideally, the conflict between the learner’s previous understanding and the new understanding offered by the teacher triggers the process of accommodation that results in correction of the misconception.
The most effective educators are skilled in both diagnosing the learner’s misconceptions and, at the same time, crafting a conversation to rectify these misconceptions. The use of questions that require the learner to explain “why” (e.g., “why do you think this patient requires an arterial line?”) or ask the learner to compare and contrast related but different things (e.g., “compare and contrast volume-controlled ventilation and pressure-controlled ventilation”) are ways for the educator to diagnose the learners’ (mis)understandings. Rather than spoon-feeding learners, constructivist educators help learners generate the answers for themselves.29
Our constructivism scenario shows that poor educational results sometimes occur with direct delivery (lectures) of complex concepts, especially when such lectures occur without concurrent assessment of the learner’s baseline understanding. The educator can act as a diagnostician by providing a pretest that would identify gaps in understanding. Alternatively, the educator can stop and pose a question to the audience during the lecture to check whether the learners are able to generate the correct answers. Anonymous electronic polling can assist constructivist educators to identify learner misconceptions. Incorrect responses can then be debriefed and corrected by the educator during the lecture itself.
Analogical transfer scenario: A fourth year medical student is accompanying you in the operating room while you provide anesthesia. She tells you about a liver transplant case that she saw yesterday, mentioning hemodynamic instability and massive transfusion requirements. While she is talking, you notice that your patient has become hypotensive. You quickly scan the work area and realize that the isoflurane has been set too high. Just before correcting the issue, you ask the medical student how she would treat the hypotension. She replies that she would give volume. Why did the medical student reply with this answer? Are there teaching techniques that could help avoid mistakes of this nature?
Analogical transfer is the process by which the solution to one problem is used (transferred) to solve a new and analogous problem in another situation. The source problem, known as the analog, provides a mental template for the understanding of the new problem. For example, while testing various synthetic compounds for antispasmodic properties, pharmacist Otto Schaumann noticed that one of the drugs, meperidine, caused the tail of a mouse to curve in an S-shape. This physical effect had only ever previously been seen with morphine. Consequently, Schaumann conjectured that meperidine might also share the drug’s analgesic effects. Further investigation concluded that meperidine did indeed contain narcotic properties similar to those of morphine.30
Analogical transfer appropriately occurs when the deep structure of 2 problems is the same. The surface structure of the problems can be different even when the deep structure is the same. For example, determining the current flow through a simple electric circuit and determining the rate of water flow in a pipe are 2 problems with different surface structures. However, these 2 problems are analogous because they share a common deep structure with Ohm’s law. In both cases, flow is directly related to a gradient and is inversely related to a resistance. Once there is recognition by a learner that 2 problems share a common deep structure, the odds of analogical transfer increase significantly.31
Low road transfer refers to solving a new problem that shares both surface and deep structure with an analog. An extreme example of low road transfer is learning how to tie the laces of a shoe on the left foot and then applying that to the laces of a shoe on the right foot. Because of the shared surface structure, opportunities for low road transfer are often easily recognized by novices. Conversely, high road transfer refers to solving a new problem that shares little or no surface structure with its analog, but does share deep structure. An example of high road transfer is using the physiological principles that explain a decrease in end-tidal carbon dioxide associated with exsanguinating hemorrhage to explain the decrease in end-tidal carbon dioxide associated with spinal or epidural anesthesia. In both cases, reduced pulmonary blood volume and pulmonary blood flow give rise to increased dead space, although the causal (surface) elements differ somewhat. Novices have a difficult time recognizing analogs when surface structures differ, and educators can provide valuable learning opportunities by assisting novices identify opportunities for high road transfer when they exist.
Despite the great importance of analogical transfer, humans are surprisingly poor at using it in problem solving.32 One study indicated that only 30% of subjects were able to use a previously shown analog to solve a new problem.33 The difficulty arises in correctly identifying and matching the deep structural elements of both the new problem and the analog.
Surface structure can complicate the process in 2 ways. First, when the surface structure of a new problem differs from an appropriate analog, then the analog may fail to be recognized as a solution to the new problem. For example, hypovolemic shock is characterized by hypotension, a small and underfilled heart and a low central venous pressure (CVP). This constellation is easily recognized and volume infusion is the therapy of choice. In contrast, intra-abdominal hypertension (abdominal compartment syndrome) is characterized by hypotension, a small and underfilled heart and a high CVP. The high CVP may lead the learner to believe that the heart is full when in fact it is compressed and underfilled. In this case, the difference in surface structure (low versus high CVP) may result in a learner missing the common deep structure of these 2 forms of shock (an underfilled heart), which may cause the learner to avoid giving the needed volume therapy because of the high CVP.
The second way surface structure can complicate transfer is when a new problem’s surface structure is the same as a possible analog, but the deep structure is different. In this situation, the analog may be incorrectly applied to the similar-appearing new problem. This process of identifying an analog and then using it inappropriately to solve a problem is known as negative transfer. Negative transfer is high when surface elements are similar.34 For example, a hypotensive, hypovolemic, and tachycardic patient should be treated with volume to increase venous return, cardiac output, and arterial blood pressure. In response to this fluid infusion, the heart rate is expected to decrease. If a learner somehow thought that directly reducing the tachycardia would help treat the hypotension and did not appreciate that the tachycardia was reflexive and resulted from the primary problem of hypotension and hypovolemia, then the learner may incorrectly use a β-blocker to slow the heart rate. The learner would have recognized that in certain cases the resolution of hypotension is associated with resolution of the tachycardia (surface structure), but failed to understand the deeper structure of the problem, resulting in negative transfer. Of note, learners who can successfully identify and incorporate analogs when they do exist (positive transfer) are also better at avoiding negative transfer.34
Educators can improve the learners’ successful use of analogical transfer by helping them develop schemata for problems that share similar deep structure. A schema acts as the abstract representation of a deep structure that is common to multiple problems and is useful in helping a learner to recognize opportunities for transfer. For example, the term “full stomach” can be viewed as a construct or a schema that means anything that predisposes the patient to aspiration. There are many examples that fit this abstract concept of “risk of aspiration” such as recent ingestion of a meal, severe reflux disease, achalasia, pregnancy, bowel obstruction, emergency surgery, etc. The benefit of a schema is that it functions as an analog already stripped of its surface elements. By developing schemata, learners decrease the likelihood of negative transfer and establish a repository of analogs that promote the probability of successful positive transfer.35
Educators who teach with more generic representations of a problem help learners create accurate schemata. Teaching with concrete examples has been shown to be less effective at promoting analogical transfer than teaching with abstract examples.36 However, concrete examples tend to be more engaging for learners. Thus, presenting a concrete problem followed by a generic example may be an optimal instructional design to encourage transfer. If educators choose to use concrete examples exclusively, then offering multiple examples of the same type of problem with varied surface elements will help learners identify common deep structures and thereby create schemata. Concerns have been raised that the use of abstract teaching might compromise learning because it detracts from the authenticity of the educational experience. However, a recent study of third-year medical students indicated that differences in instructional authenticity (i.e., use of paper-based cases versus live standardized patients) did not affect subsequent clinical performance.37 Interestingly, the authors felt that perhaps any advantage gleaned from increased authenticity might have been mitigated by drawbacks introduced from the increased cognitive load of a more complex instructional format.
Novice learners typically focus on surface structure and miss the deep structure. This makes analogical transfer difficult for the novice. When a novice makes an error in transfer, it is important for the educator to explain why the error is incorrect as opposed to simply telling the learner what to do without offering an explanation. Part of the educator’s role is to assist the learner recognize common deep structure. This explanation will promote deeper learning that will help with subsequent transfer (effect sizes of 0.9–1.6).38
Verbalization of a problem’s structure by the learner can also help promote successful analogical transfer because it assists in schema formation. When learners are required to explain how they arrived at a solution (a form of constructivism), they increase the likelihood of future positive transfer.32 As the complexity of a problem increases, learners often have more difficulty articulating their thought processes, even if they are able to correctly solve the problem.39 If a learner cannot express how the solution to a problem is determined, analogical transfer between that problem and similar problems will not happen. Explicit awareness of the deep structure of an analog is requisite for successful analogical transfer to occur.
In our initial scenario, the medical student exhibited negative transfer. She used surface structure (hypotension) to incorrectly transfer the solution used on a hypotensive hypovolemic patient (volume infusion) to treat a hypotensive normovolemic patient who had an isoflurane overdose. The educator now has an opportunity to use elements of constructivism by asking the student to compare and contrast different causes for hypotension in these different cases (hypovolemia and vasodilation). It is likely that the student will reveal his/her misunderstandings during the discussion, and the educator can then help correct them.
Goal Orientation: Learning Versus Performance Orientation
Goal orientation scenario: While on call one night you give a resident the choice to take over a laparoscopic cholecystectomy on a healthy patient or to start an emergent repair of a ruptured aortic aneurysm. The resident opts to take care of the patient having the cholecystectomy. You recognize that this resident routinely avoids challenging yet highly educational cases in favor of cases he can easily manage. Why is this? Are there strategies that can be used to help individuals choose more challenging cases that will enhance their clinical skills?
Achievement goal orientation, or simply goal orientation, refers to the implicit goals that individuals have for themselves when they are placed in an achievement situations (i.e., they have the opportunity to succeed or fail). These goals illustrate a person’s dominant beliefs regarding self-improvement. They also predict how an individual will respond to challenges and failure. There are 2 predominant goal orientations: “performance orientation” and “learning orientation.” Individuals with a performance orientation have a primary goal of validating their abilities, mainly by demonstrating their abilities to others. Because their primary goal is to validate their abilities, they also try to avoid revealing that they do not understand something or are not able to perform a task correctly. In contrast, individuals with a learning orientation have the primary goal of increasing their competency or mastery of a topic. Their actual mastery matters much more to them than how their mastery is perceived by others. A frequent misconception arises because of the terminology used in labeling goal orientations. “Performance orientation” and “learning orientation” do not mean that individuals focus on performing and learning, respectively. Instead, they refer to how an individual construes challenges, setbacks, or failures. In particular, when performance-oriented individuals encounter a setback, they will conclude that they are not able to do the task because they view a setback as diagnostic of their inherent (in)ability. They see abilities as fixed. In contrast, when learning-oriented individuals encounter a setback, they will conclude that they need to increase their effort or change their strategy to improve their competency because they construe a setback as diagnostic of a need to improve. Learning-oriented individuals view their abilities as malleable.
These 2 different goal orientations have been shown to directly influence how resident physicians view feedback regarding their performance.40 Residents who have a strong learning orientation perceive feedback as beneficial. They do not see negative costs associated with receiving feedback. Consistent with their goal of mastery, a learning-oriented resident will use the feedback to help improve their performance. In contrast, residents who have a strong performance orientation perceive feedback as costly. They do not recognize the benefits associated with feedback. Performance-oriented residents view feedback as a mechanism to point out the areas of weakness that interferes with their goal of validating their skills and abilities. The relationship of performance orientation to the goal of validating one’s ability and attempting to impress others was recapitulated with medical students.41 The overall measure of performance orientation increased significantly in the third-year medical students during their clinical rotations. This may have occurred as they increased the goal of validating themselves, especially in front of faculty members and residents. Importantly, these orientations are independent constructs, and thus, a single individual can have a low, medium, or high learning orientation and either a low, medium, or high performance orientation.42 For the purposes of performance improvement, a high learning orientation seems to be the key and needed attribute.
How an individual approaches a challenge or a setback can differ depending on their dominant goal orientation. Performance-oriented individuals will tend to avoid challenging situations for fear of appearing inept. Similarly, if the task requires a great deal of effort, or does not come easily, a performance-oriented individual is likely to simply conclude that they are “not good” at the task. Furthermore, if a task is attempted and failure results, these individuals will be very concerned with the resulting negative judgments made by others about their abilities. They will work to sidestep being placed in a similar situation in the future. Conversely, learning-oriented individuals view challenges as opportunities for growth. They see failures as useful for improving and they strive to learn and grow from their mistakes.43 Learning-oriented individuals see effort and strategy as normal factors to be used for improving performance.
In an educational environment, particularly one beset with frequent challenges, it is ideal that the learner possesses a strong learning orientation. Although an individual may default toward either a performance orientation or a learning orientation, situational cues and a learner’s educational environment can modify these tendencies.42 This means that performance-oriented learners can be influenced toward a learning orientation through discussion and education. Such interventions have been shown to produce a significant change in an individual’s goal orientation.44 On a societal level, different cultures can affect the learning orientation of entire populations of people.45,46 Thus, an educator can positively influence a learner’s reaction to failure by creating an environment that emphasizes a learning orientation. This is especially beneficial for those learners who are naturally more performance oriented.47 College students who received a low score on a test and who had an acute intervention that increased learning orientation were more interested in taking a remedial class to improve their scores than were the students who had an intervention that increased their performance orientation (d = 1.1).48 In another study, college students who performed poorly on a test and were randomly assigned to an acute intervention that increased learning orientation were found to be more open to remediation and were less defensive (d = 1.6).49 Educators who praise effort and strategy, encourage creativity and discovery, deemphasize the importance of natural ability, and reduce praise for static traits such as intelligence are promoting a learning orientation.
In our initial scenario, the resident appeared to hold a strong performance orientation because he consistently chose cases that he could successfully do so that he would validate his abilities to others. This will ensure that he “looks good,” but it will not help the resident improve. The educator can help the resident choose the more challenging and educational cases (the ruptured abdominal aneurysm) by indicating to the resident that taking on challenges is the most effective path to improving performance. The educator must also ensure that when failures occur, no indications are made that would imply that these failures are because of some inherent limitation in the abilities of the resident. Instead, when challenges occur, the educator can provide concrete suggestions that will help improve the skills of the resident. Addressing setbacks with strategic actions and focused effort is consistent with a learning orientation. Praising a resident for effective actions they took or for useful efforts they made will also support a learning orientation. Because residency is characterized by frequent challenges, it behooves both the educator and the learner to adopt a learning orientation. In contrast, praising or complaining to a resident in a manner that implies that traits are static encourages a performance orientation. Telling a resident that they are “smart” (even though it is praise) focuses on innate ability over effort, self-improvement, and persistence, and will encourage a performance orientation.
Metacognition scenario: You are assigned to provide anesthesia for a patient requiring tracheal resection and reconstruction. The night before the surgery, you read a chapter on the intricacies of the surgical procedure as well as a journal article on appropriate anesthetic management. After your review, you go to bed feeling well prepared for the case. However, the next day you are not able to easily remember some of the key concepts you thought you had mastered the night before. How did this happen? What can you do to prevent a similar episode in the future?
Self-directed learning is a dynamic process in which learners choose objectives, monitor progress, and adjust study patterns accordingly.50 A learner’s ability to accurately monitor what he or she truly understands plays a critical role in effective learning. Metacognition is the act of thinking about one’s own thinking and includes recognizing when one does or does not actually understand something. For example, if you were discussing a drug’s mechanism of action with a colleague and you realized that you did not know how the drug worked on a deeper level, then you would have used metacognition to recognize the limits of your understanding. Metacognition also pertains to motor skills.51 This means that some people can have poor motor skills and be unaware of it because of their poor metacognition. A related topic is metacomprehension, which is the term used to describe the awareness or understanding an individual has regarding his or her own level of comprehension about a topic. Accurate metacomprehension correlates with improved study patterns and better performance during testing.52 More efficient learning occurs from learners with high metacomprehension, because they allocate the appropriate amount of time to subjects that are less well understood and spend little time reviewing items that have already been mastered. Unfortunately, human metacognition is generally poor.53,54
Learners should develop study habits that focus on improving metacognition. Deliberately creating a plan of how to approach a study topic can result in increased metacognition. For example, if a learner desires to know more about regional anesthesia, he or she should map the study guide dictating a systematic review of the subcategories of the topic (e.g., first reviewing upper extremity anatomy, then lower extremity anatomy, followed by the study of the characteristics of local anesthetics, and finishing with strategies for ultrasound optimization). Once planning is completed and study has commenced, it is then important to periodically evaluate understanding to ensure that mastery of the previously reviewed concepts has been retained. Evaluation can be accomplished by self-testing through practice questions or flash cards, or even simply through deliberate mental review. The key is that by focusing on the process of learning and by regular self-evaluation, students can increase their awareness of what is known and what needs additional review (i.e., their metacognition) and thus improve learning.
Study Techniques That Improve Metacognition
Certain specific study techniques can naturally improve metacognition. For example, reading a passage more than once increases a learner’s understanding of the material and also improves the learner’s metacognition (i.e., it enhances the learner’s awareness of which parts of the passage are well understood and which require more review).55 Taking practice tests is another method that can aid in improving metacognition.56
Summarizing study material after a study session is another learning technique that has been examined extensively. Generating written summaries of the material reviewed has been shown to improve comprehension.57 The act of summarizing better enables learners to determine which concepts are fundamental. They can then focus attention on these central elements. In addition, creating a summary acts as a form of self-testing, a practice which is known to improve metacognition. Some investigators argue that the enhanced comprehension that comes with summarization is primarily a result of improved metacognition.57
Of note, when summarization is used, metacognition is improved to a greater degree when material is summarized after a time delay.52,58 Summarizing immediately after reviewing material allows a learner to easily recount concepts that are stored only in short-term memory. Much of this material will never transition to long-term memory. Thus, immediate and effortless recall after study can fool learners into believing that they have mastered the material when, in fact, that material may not shift into long-term memory. By summarizing after a delay, a learner will more accurately assess what concepts have actually been learned. Thus, delayed summarization alerts learners to material that has not been learned in a durable fashion and that deserves additional attention. By incorporating delayed summarization into their study, learners can improve overall comprehension and develop greater metacognition.57 This can directly translate into better learning as reflected by better test performance.52
In our tracheal resection scenario, it is likely that the clinician was misled into believing that he or she fully understood the case because of a solid grasp of the steps and anesthetic considerations immediately after having read the material. Unfortunately, only a fraction of the recently read material made it into long-term memory, and thus, much of the material was lost and not available the following day when it was needed. This situation could be improved by simply taking a short (15–30 minutes) break after finishing the material and then trying to recount the salient features of the material. Any failure to recall key steps would then be detected, and these gaps could be repaired by reviewing the unlearned material.
Retrieval scenario: You have always preferred teaching using the Socratic method where educators repeatedly ask learners questions to facilitate instruction. One day a resident tells you he strongly dislikes this method of education and believes it is less effective. The resident feels that when he is put on the spot he is unable to focus on learning. He states he prefers when the attending physicians simply tell him what they are thinking rather than “quizzing” him. Are the resident’s concerns valid? Should you change your teaching style? Are there benefits to teaching by the Socratic method that might outweigh the discomfort felt by some learners when asked questions directly?
Traditionally, students have learned new material by encoding, a process of taking in information by repeatedly reviewing the information until it is learned. Examples of encoding in formal education are reading textbooks and listening to lectures. Research has shown that this approach is not optimal for long-term retention.59 Instead, study methods that require a student to retrieve (or pull out) information from memory will result in much better long-term retention of the information compared with simply rereading the material. One analysis demonstrated that, under certain conditions, study that incorporated retrieval resulted in correct answers on final testing 80% of the time, whereas a control group who used restudy without retrieval only answered 33% to 36% of final testing questions correctly.60
In 1978, Slamecka and Graf61 investigated whether students who simply read a passage repeatedly would learn as much as students who attempted to recall details from memory after reading a passage. Slamecka and Graf found that the cohort who learned by retrieving answers from memory retained knowledge much better over the long term. Numerous subsequent studies have confirmed that learning using active retrieval (i.e., practice tests where the answers are not provided until after the test is completed) results in superior long-term memory when compared with learning that focuses on repeated encoding (i.e., rereading textbooks, relistening to lectures).62,63 One study demonstrated that the learner’s long-term retention (1 week) was improved by 4 SDs (Cohen d = 4) when using retrieval strategies compared with using the normal strategy of simply rereading the material.60 This reinforces the concept that repeated retrieval is one of the most effective strategies to ensure that new material is learned in a durable fashion. More recently retrieval-based learning was directly compared with concept mapping, which is a strategy used to enhance understanding of conceptual material like that found in biological systems. Retrieval-based learning outperformed the well-known concept mapping technique with an effect size of 1.5.64 Thus, retrieval-based learning augments conceptual learning in addition to its well-known effects on simple information retention. Retrieval exercises such as self-testing are also an effective way of improving metacognition.59 Self-testing reveals gaps in knowledge and allows for more efficient and focused learning.55
Conventional wisdom says that after information is learned to the point where it can be recalled once, there is no benefit from additional study.60 Research indicates otherwise and points to the usefulness of repeated recall. Continued study by repeated retrieval, even after the target study material can be successfully recalled once, results in significant positive effects on long-term retention. In contrast, additional study by encoding seems to provide little long-term benefit.60
Learners are generally unaware of the benefits of retrieval-based learning,60 and thus educators should encourage learners to incorporate these methods into their study processes. Textbooks, journal articles, and didactic lectures are well suited for the initial exposure to new material. However, if the material is deemed important to know for long-term purposes, then further study of that material should incorporate active recall instead of simple review. Flashcards, when used appropriately, function well to promote retrieval. Similarly, actively recalling fundamental algorithms (i.e., advanced cardiac life support, malignant hyperthermia, difficult airway) will benefit a learner to a greater extent than repeated readings of the algorithm. Both educator and learner would profit from embracing greater use of the Socratic method in education,65 because it is based on the principle of retrieval (and constructivism). By choosing retrieval-based exercises over simple encoding, learners can increase metacognition and enhance long-term information retention.
In our scenario, the learner prefers to be told information instead of being made to answer questions. However, the evidence favors having students retrieve the information after it has been presented. Thus, it is reasonable to tell the resident what he needs to know but then the educator needs to follow up to ensure that the learner can self-produce (retrieve) the information. The very act of retrieval is an act of learning and will help the resident to know the material in a much more durable fashion. Self-testing will also accomplish this goal and may be a useful approach for the motivated learner.
Spaced learning scenario: Your board examination is coming up in 6 months and you are planning your study schedule. In college, you performed quite well by cramming for final examinations at the end of each semester. In medical school, you continued to get good grades; however, your method of preparing for examinations changed such that you spread your studying over a longer period of time. Which strategy should you use to prepare for your board examination? Is one superior to the other?
Spaced learning is the practice of learning information by studying the material with distinct time intervals between each study period. In contrast, massed learning is learning information without significant interruption. Cramming for an examination is an example of massed learning. The results of >100 years of study66,67 strongly support spaced learning over massed learning to achieve long-term retention of studied material. This benefit is known as the spacing effect, and it applies both to studying for knowledge acquisition and to practicing for motor skill development.68 Spaced education has been used to increase practicing clinicians’ medical knowledge of inappropriately ordered medical tests (d = 1.1).69 This translated into 40% fewer inappropriate tests being ordered on actual patients.69
Hundreds of studies analyzing spaced learning have emerged, and the precise details of the spacing itself can appear quite different. Interstudy intervals can range from a few seconds to a few days, spacing schedules can be fixed or variable,70 studied retention times can vary significantly, and the general domain of the material can differ dramatically from study to study. However, meta-analyses identify at least 3 consistent themes with regards to spaced learning.67 First, regardless of the interstudy interval, spaced learning is always superior to massed learning. For example, a total study time of 2 hours is more effective when used as 4 30-minute sessions (with space between each session) compared with a single 2-hour session regardless of the time between sessions. Second, longer interstudy intervals are generally associated with better long-term retention (provided the interstudy interval is not increased to the point that the learner completely forgets the previous learning episode). Finally, spaced learning works equally well with verbal tasks and motor tasks, with the most substantial gains observed in tasks of lower complexity. Paradoxically, learners usually prefer massed learning and believe that it is more effective than spaced learning. This occurs despite their own test performance demonstrating the opposite results.54 This means that learners can be fooled by the illusion that massed learning is more effective than spacing when in fact the opposite is true. This is an example of failure in the domain of metacognition. Interestingly, the spacing effect is not as readily apparent with computational or mathematical tasks.71
In the practice of medicine, the concept of spaced learning can be applied broadly. New residents striving to learn standard dosages of rarely used medications should focus on brief daily reviews of these values as opposed to studying them in longer, less frequent study sessions. Similarly, seasoned physicians hoping to improve their transesophageal echocardiography image acquisition abilities might opt for more regular practice with departmental simulators instead of attending an intensive 3-day crash course. Even at the curriculum level, educators may decide that 2 separate 1-month rotations in cardiac anesthesia are preferable to a single 2-month rotation. Given the evidence in favor of spaced learning, it is surprising that its application is not more ubiquitous.
In our scenario concerning how best to study for an upcoming examination, the evidence strongly supports spacing the study sessions over time instead of consolidating the study into a short high-intensity episode. For an equivalent total study duration, spaced learning will result in better long-term retention of the material. Both spaced learning and retrieval are examples of knowledge acquisition and retention strategies. Although they are distinct strategies, both of these approaches act as evidence that traditional methods of learning by lecture or simple review can be improved on.
Deliberate practice scenario: You hear about a colleague who recently canceled an elective case scheduled for general anesthesia because she recognized a small spontaneous pneumothorax on the patient’s preoperative chest radiograph. You wonder if you would have caught such a subtle finding and recognize the need to improve your proficiency at reading chest radiographs. What is the best method to approach such a goal?
To become expert in a given domain, it is essential to have experience. However, experience alone is insufficient to achieve mastery. By examining the practice methods of expert performers, Ericsson et al.72 found that elite performers engaged in much more deliberate practice than lower level performers. Deliberate practice is a specific type of practice with the singular goal of improving performance. It is highly structured, focuses on overcoming weaknesses, and requires significant effort. Performance is monitored, and feedback is provided with the goal of enhancing performance. Deliberate practice focuses on well-defined tasks that are manageable in size and scope. The topic for practice should appropriately challenge the learner (i.e., the learner may not initially be successful at the task). The learner should also be afforded abundant opportunities for repetition to allow for correction of errors.72
It is also important to understand which activities do not represent deliberate practice. In a retrospective examination of the study habits of musical performers in an elite institution, the total amount of time spent in general musical activities did not vary between elite and average students.73 Examples of general musical activities include time spent performing for an audience and playing for fun or practicing a piece from beginning to end without breaking it down and focusing more heavily on challenging sections. These general activities do not reflect deliberate practice and do not appear to promote mastery. When the various musical activities were further broken down and analyzed by type, it became clear that top students focused on a significantly larger portion of their time engaged in deliberate practice than did middle or bottom performers.73 In an analysis of elite chess players, this theme emerged again.74 Players who dedicated more time to studying chess positions and comparing their moves with those of a grandmaster subsequently attained higher rankings than did players who focused on playing games in tournaments or for fun. Across multiple other disciplines (sports,75 memory,76 spelling,77 and typing78), the evidence supports the conclusion that elite performance can be linked to the amount of time spent in deliberate practice with effect sizes ranging from 0.4 to 1.2.79
Deliberate practice has been successfully applied to improving performance in a variety of medical skills. For example, deliberate practice has been used to significantly improve the performance of advanced cardiac life support,80 central line placement,81,82 lumbar puncture,83 and thoracentesis84 with effect sizes ranging from 1.9 to 3.9.
Additional factors also play a role in the development of elite performance. In addition to training and practice, Howe et al.’s85 review of the literature identified early experiences in a field, individual preferences, opportunities, and an individual’s habits, as determinants of excellence. Notably, there appears to be only modest evidence supporting the notion that innate talents or abilities are required to develop expertise in most fields.85 Even if natural aptitude does confer a performance benefit, large additional gains can still be obtained through the consistent engagement in deliberate practice.86 The concept that years of experience alone result in domain expertise has also been called into question.87 In the field of medicine, physicians who have been in practice longer have been shown to be at risk of providing lower quality care.88 This is likely because of a lack of continued focused training. A recent study explored how physicians learn in the workplace and found that physicians in practice did not engage in meaningful deliberate practice.89 Conversely, high performers can maintain their abilities even into older ages if they persist in regular deliberate practice.90,91
There are abundant opportunities for learners in the medical field to incorporate deliberate practice into their education. For example, medical learners desiring to improve their ability to read electrocardiograms might gather and review a large number of tracings, interpret them on their own, and then check their conclusions with an expert such as a cardiologist who specializes in electrophysiology. Ideally, a learner would receive feedback from the cardiologist when an error was made. As learners become more adept, increasingly difficult electrocardiograms should be introduced to keep them practicing at the edge of their skills. This concept is broadly applicable and could easily be applied to interpreting arterial blood gas results, reading head computed tomograms, or diagnosing valvular lesions by transesophageal echocardiography.
Deliberate practice can also serve a particularly important role in developing abilities related to scenarios that are both difficult and uncommon. Studies have indicated that the reliable superiority of true experts is principally evident in more challenging cases.92,93 Mastery of these more complex problems should be the goal of every learner aspiring for expert performance. The growth of medical simulation offers opportunities for deliberate practice in rare or challenging situations. Even in everyday practice, a motivated learner can create ways to rehearse vital but infrequently used skills. For example, practicing fiberoptic intubation on a patient who could be intubated via direct laryngoscopy is a method of increasing proficiency in a skill that is only necessary in infrequent but challenging situations.
The literature highlights that domain mastery can be achieved by those willing to engage in appropriate practice patterns, but it takes an enormous amounts of work to attain such expertise. Ericsson87 and Ericsson et al.94 estimate that it may require thousands of hours of deliberate practice to achieve elite performance. In the pursuit of excellence, there are no shortcuts. As Sloboda95 aptly stated, “There is absolutely no evidence of ‘fast-track’ for high achievers.”
Our initial scenario demonstrates that sophisticated or advanced levels of medical care are achievable when individuals engage in deliberate practice deigned to increase performance in targeted areas. There is every reason to believe that any clinician who is willing to engage in deliberate practice could ultimately reach this level of performance.
CAVEATS AND LIMITATIONS
The concepts reviewed in this article (cognitive load theory, constructivism, analogical transfer, goal orientation, metacognition, retrieval, spaced learning, and deliberate practice) rest on empirical evidence to support them (Table 3). The primary limitation of these concepts arises when teachers and learners attempt to implement them to improve education. Cognitive load theory is a nice example of a concept that has significant empirical evidence to support it as consequential for teaching and learning96 but that has not been widely adopted into educational practice. Integrating cognitive load theory into educational practice would require developing educational content and then designing the educational process to optimally use cognitive load theory for the most efficient learning. There is little information about how to do this, never mind doing it in an optimal fashion. Most medical educators focus on content (the subject matter that they want to teach) and very few then design the teaching of that material using cognitive load theory to maximize teaching. Fortunately, Issa et al.11 have published an example showing how they effectively used cognitive load theory to redesign a series of lectures for medical students on circulatory shock, which resulted in significantly better learning of the material at 4 weeks (Cohen d = 1.17). Such examples are currently rare. The limitations are thus 2-fold: many educators still do not know these concepts and far fewer know how to implement these concepts as strategies for better education. This partly stems from the fact that many of these concepts have been studied in highly controlled experimental settings but have not then been extensively tested in common educational environments. Fortunately, all of these concepts are now well supported and are represented in widely available peer-reviewed journals. The next steps to improving medical education are broader appreciation of these and other essential educational concepts by educators, followed by effective application of these principles to common educational environments.
As discoveries continue to expand our understanding of medicine, effective learning and teaching remain essential to both the development and the maintenance of a physician’s knowledge base. Although there is no replacement for the amount of time dedicated to study, a better understanding of established learning and teaching methods can dramatically improve the effectiveness of the time spent in these activities. During the early years of training, it is fundamental that medical schools and residency programs use educational practices based on evidence as they plan curricula for educating physicians. After the completion of formal training, specific learning methods designed to enhance education and performance will assist practicing physicians as they strive to both maintain and grow their knowledge over the course of their career.
Name: Joseph Weidman, MD.
Contribution: This author helped write the manuscript.
Attestation: Joseph Weidman approved the final manuscript.
Name: Keith Baker, MD, PhD.
Contribution: This author helped write the manuscript.
Attestation: Keith Baker approved the final manuscript.
This manuscript was handled by: Gregory J. Crosby, MD.
1. Miller RD Anesthesia. 1981 New York Churchill Livingstone
2. Miller RD Miller’s anesthesia. 20158th ed. Philadelphia Elsevier/Saunders
3. . A modification in the training requirements in anesthesiology: requirements for the third clinical anesthesia year. American Board of Anesthesiology. Anesthesiology. 1985;62:175–7
4. Baddeley A. The episodic buffer: a new component of working memory? Trends Cogn Sci. 2000;4:417–23
5. Mayer RE. Applying the science of learning to medical education. Med Educ. 2010;44:543–9
6. Sweller J. Cognitive load during problem solving: effects on learning. Cogn Sci. 1988;12:257–85
7. Sweller J, van Merrienboer JJG, Paas FGWC. Cognitive architecture and instructional design. Educ Psychol Rev. 1998;10:251–96
8. van Merriënboer JJ, Sweller J. Cognitive load theory in health professional education: design principles and strategies. Med Educ. 2010;44:85–93
9. Chandler P, Sweller J. Cognitive load theory and the format of instruction. Cogn Instr. 1991;8:293–332
10. Cheon J, Grant MM. The effects of metaphorical interface on germane cognitive load in Web-based instruction. Education Tech Research Dev. 2012;60:399–420
11. Issa N, Mayer RE, Schuller M, Wang E, Shapiro MB, DaRosa DA. Teaching for understanding in medical classrooms using multimedia design principles. Med Educ. 2013;47:388–96
12. Yeung N, Monsell S. Switching between tasks of unequal familiarity: the role of stimulus-attribute and response-set selection. J Exp Psychol Hum Percept Perform. 2003;29:455–69
13. Rogers RD, Monsell S. Costs of a predictible switch between simple cognitive tasks. J Exp Psychol Gen. 1995;124:207–31
14. Lindenberger U, Marsiske M, Baltes PB. Memorizing while walking: increase in dual-task costs from young adulthood to old age. Psychol Aging. 2000;15:417–36
15. Stefanidis D, Scerbo MW, Korndorffer JR Jr, Scott DJ. Redefining simulator proficiency using automaticity theory. Am J Surg. 2007;193:502–6
16. Schneider W, Shiffrin RM. Controlled and automatic human information processing: I. Detection, search, and attention. Psychol Rev. 1977;84:1–66
17. Shiffrin RM, Schneider W. Controlled and automatic human information processing: II. Perceptual learning, automatic attending and a general theory. Psychol Rev. 1977;84:127–90
18. Kurahashi AM, Harvey A, MacRae H, Moulton CA, Dubrowski A. Technical skill training improves the ability to learn. Surgery. 2011;149:1–6
19. Bransford JD, Brown AL, Cocking RR How People Learn: Brain, mind, Experience, and School. 1999 Washington, DC National Academy Press
20. Wadsworth BJ Piaget’s Theory of Cognitive and Affective Development: Foundations of Constructivism. 19965th ed. White Plains Longman Publishing
21. Michael J. Where’s the evidence that active learning works? Adv Physiol Educ. 2006;30:159–67
22. von Glasersfeld E. Cognition, construction of knowledge, and teaching. Synthese. 1989;80:121–40
23. Eckert TL, Dunn EK, Codding RS, Begeny JC, Kleinmann AE. Assessment of mathematics and reading performance: an examination of the correspondence between direct assessment of student performance and teacher report. Psychol Sch. 2006;43:247–65
24. Haak DC, HilleRisLambers J, Pitre E, Freeman S. Increased structure and active learning reduce the achievement gap in introductory biology. Science. 2011;332:1213–6
25. Freeman S, Eddy SL, McDonough M, Smith MK, Okoroafor N, Jordt H, Wenderoth MP. Active learning increases student performance in science, engineering, and mathematics. Proc Natl Acad Sci USA. 2014;111:8410–5
26. Hake RR. Interactive-engagement versus traditional methods: a six-thousand-student survey of mechanics test data for introductory physics courses. Am J Phys. 1998;66:64–74
27. Smith MK, Wood WB, Adams WK, Wieman C, Knight JK, Guild N, Su TT. Why peer discussion improves student performance on in-class concept questions. Science. 2009;323:122–4
28. Osborne J. Arguing to learn in science: the role of collaborative, critical discourse. Science. 2010;328:463–6
29. Metcalfe J, Kornell N. Principles of cognitive science in education: the effects of generation, errors, and feedback. Psychon Bull Rev. 2007;14:225–9
30. Reynolds AK, Randall LO Morphine and Related Drugs.. 1975 Toronto University of Toronto Press
31. Holyoak KJ, Koh K. Surface and structural similarity in analogical transfer. Mem Cognit. 1987;15:332–40
32. Norman G. Teaching basic science to optimize transfer. Med Teach. 2009;31:807–11
33. Gick ML, Holyoak KJ. Schema induction and analogical transfer. Cogn Psychol. 1983;15:1–38
34. Novick LR. Analogical transfer, problem similarity, and expertise. J Exp Psychol Learn Mem Cogn. 1988;14:510–20
35. Gentner D, Loewenstein J, Thompson L. Learning and transfer: a general role for analogical encoding. J Educ Psychol. 2003;95:393–408
36. Kaminski JA, Sloutsky VM, Heckler AF. Learning theory. The advantage of abstract examples in learning math. Science. 2008;320:454–5
37. La Rochelle JS, Durning SJ, Pangaro LN, Artino AR, van der Vleuten CP, Schuwirth L. Authenticity of instruction and student performance: a prospective randomised trial. Med Educ. 2011;45:807–17
38. Moreno R. Decreasing cognitive load for novice students: effects of explanatory versus corrective feedback in discovery-based multimedia. Instr Sci. 2004;32:99–113
39. Casale MB, Roeder JL, Ashby FG. Analogical transfer in perceptual categorization. Mem Cognit. 2012;40:434–49
40. Teunissen PW, Stapel DA, van der Vleuten C, Scherpbier A, Boor K, Scheele F. Who wants feedback? An investigation of the variables influencing residents’ feedback-seeking behavior in relation to night shifts. Acad Med. 2009;84:910–7
41. Artino AR Jr, Dong T, DeZee KJ, Gilliland WR, Waechter DM, Cruess D, Durning SJ. Achievement goal structures and self-regulated learning: relationships and changes in medical school. Acad Med. 2012;87:1375–81
42. Button SB, Mathieu JE, Zajac DM. Goal orientation in organizational research: a conceptual and empirical foundation. Organ Behav Hum Decis Process. 1996;67:26–48
43. Elliott ES, Dweck CS. Goals: an approach to motivation and achievement. J Pers Soc Psychol. 1988;54:5–12
44. Dweck CS. Motivational processes affecting learning. Am Psychol. 1986;41:1040–8
45. Heine SJ, Lehman DR, Ide E, Leung C, Kitayama S, Takata T, Matsumoto H. Divergent consequences of success and failure in japan and north america: an investigation of self-improving motivations and malleable selves. J Pers Soc Psychol. 2001;81:599–615
46. Lee C, Tinsley C, Bobko P. Cross-cultural variance in goal orientations and their effects. Appl Psychol Int Rev. 2003;52:272–97
47. Kohli AK, Shervani TA, Challagalla GN. Learning and performance orientation of salespeople: the role of supervisors. J Market Res. 1998;35:263–74
48. Hong Y-y, Chiu C-y, Dweck CS, Lin DMS, Wan W. Implicit theories, attributions, and coping: a meaning system approach. J Pers Soc Psychol. 1999;77:588–99
49. Nussbaum AD, Dweck CS. Defensiveness versus remediation: self-theories and modes of self-esteem maintenance. Pers Soc Psychol Bull. 2008;34:599–612
50. Dunlosky J, Thiede KW. What makes people study more? An evaluation of factors that affect self-paced study. Acta Psychol (Amst). 1998;98:37–56
51. Martini R, Shore BM. Pointing to parallels in ability-related differences in the use of metacognition in academic and psychomotor tasks. Learn Indiv Differ. 2008;18:237–47
52. Thiede KW, Anderson MCM, Therriault D. Accuracy of metacognitive monitoring affects learning of texts. J Educ Psychol. 2003;95:66–73
53. Maki RHHacker DJ, Dunlosky J, Graesser AC. Test predictions over text material. In: Metacognition in Educational Theory and Practice. 1998 Mahwah Lawrence Erlbaum Associates Publishers:117–44
54. Kornell N, Bjork RA. Learning concepts and categories: is spacing the “enemy of induction”? Psychol Sci. 2008;19:585–92
55. Dunlosky J, Rawson KA. Why does rereading improve metacomprehension accuracy? Evaluating the levels-of-disruption hypothesis for the rereading effect. Discourse Process. 2005;40:37–55
56. Shaughnessy JJ, Zechmeister EB. Memory-monitoring accuracy as influenced by the distribution of retrieval practice. BPS. 1992;30:125–8
57. Thiede KW, Anderson MCM. Summarizing can improve metacomprehension accuracy. Contemp Educ Psychol. 2003;28:129–60
58. Anderson MC, Thiede KW. Why do delayed summaries improve metacomprehension accuracy? Acta Psychol (Amst). 2008;128:110–8
59. Karpicke JD, Roediger HL III. Repeated retrieval during learning is the key to long-term retention. JMemL. 2007;57:151–62
60. Karpicke JD, Roediger HL III. The critical importance of retrieval for learning. Science. 2008;319:966–8
61. Slamecka NJ, Graf P. The generation effect: delineation of a phenomenon. J Exp Psychol Hum Learn. 1978;4:592–604
62. Roediger HL III, Karpicke JD. The power of testing memory: basic research and implications for educational practice. Perspect Psychol Sci. 2006;1:181–210
63. Hirshman E, Bjork RA. The generation effect: support for a two-factor theory. J Exp Psychol Learn Mem Cogn. 1988;14:484–94
64. Karpicke JD, Blunt JR. Retrieval practice produces more learning than elaborative studying with concept mapping. Science. 2011;331:772–5
65. Kost A, Chen FM. Socrates was not a pimp: changing the paradigm of questioning in medical education. Acad Med. 2014;90:20–4
66. Wixted JT. On common ground: Jost’s (1897) law of forgetting and Ribot’s (1881) law of retrograde amnesia. Psychol Rev. 2004;111:864–79
67. Cepeda NJ, Pashler H, Vul E, Wixted JT, Rohrer D. Distributed practice in verbal recall tasks: a review and quantitative synthesis. Psychol Bull. 2006;132:354–80
68. Lee TD, Genovese ED. Distribution of practice in motor skill acquisition: learning and performance effects reconsidered. Res Q Exerc Sport. 1988;59:277–87
69. Kerfoot BP, Lawler EV, Sokolovskaya G, Gagnon D, Conlin PR. Durable improvements in prostate cancer screening from online spaced education a randomized controlled trial. Am J Prev Med. 2010;39:472–8
70. Karpicke JD, Bauernschmidt A. Spaced retrieval: absolute spacing enhances learning regardless of relative spacing. J Exp Psychol Learn Mem Cogn. 2011;37:1250–7
71. Moss VD The Efficacy of Massed Versus Distributed Practice as a Function of Desired Learning Outcomes and Grade Level of the Student. 1996 Ann Arbor, MI ProQuest Information & Learning:5204
72. Ericsson KA, Krampe RT, Tesch-Römer C. The role of deliberate practice in the acquisition of expert performance. Psychol Rev. 1993;100:363–406
73. Sloboda JA, Davidson JW, Howe MJA, Moore DG. The role of practice in the development of performing musicians. Br J Psychol. 1996;87:287–309
74. Charness N, Tuffiash M, Krampe R, Reingold E, Vasyukova E. The role of deliberate practice in chess expertise. Appl Cogn Psychol. 2005;19:151–65
75. Ericsson KA, Ward P. Capturing the naturally occurring superior performance of experts in the laboratory: toward a science of expert and exceptional performance. Curr Dir Psychol Sci. 2007;16:346–50
76. Ericcson KA, Chase WG, Faloon S. Acquisition of a memory skill. Science. 1980;208:1181–2
77. Duckworth AL, Kirby TA, Tsukayama E, Berstein H, Ericsson KA. Deliberate practice spells success: why grittier competitors triumph at the National Spelling Bee. Soc Psychol Personal Sci. 2011;2:174–81
78. Keith N, Ericsson KA. A deliberate practice account of typing proficiency in everyday typists. J Exp Psychol Appl. 2007;13:135–45
79. Macnamara BN, Hambrick DZ, Oswald FL. Deliberate practice and performance in music, games, sports, education, and professions: a meta-analysis. Psychol Sci. 2014;25:1608–18
80. Wayne DB, Butter J, Siddall VJ, Fudala MJ, Linquist LA, Feinglass J, Wade LD, McGaghie WC. Simulation-based training of internal medicine residents in advanced cardiac life support protocols: a randomized trial. Teach Learn Med. 2005;17:210–6
81. Barsuk JH, Cohen ER, McGaghie WC, Wayne DB. Long-term retention of central venous catheter insertion skills after simulation-based mastery learning. Acad Med. 2010;85:S9–12
82. Khouli H, Jahnes K, Shapiro J, Rose K, Mathew J, Gohil A, Han Q, Sotelo A, Jones J, Aqeel A, Eden E, Fried E. Performance of medical residents in sterile techniques during central vein catheterization: randomized trial of efficacy of simulation-based training. Chest. 2011;139:80–7
83. Barsuk JH, Cohen ER, Caprio T, McGaghie WC, Simuni T, Wayne DB. Simulation-based education with mastery learning improves residents’ lumbar puncture skills. Neurology. 2012;79:132–7
84. Wayne DB, Barsuk JH, O’Leary KJ, Fudala MJ, McGaghie WC. Mastery learning of thoracentesis skills by internal medicine residents using simulation technology and deliberate practice. J Hosp Med. 2008;3:48–54
85. Howe MJ, Davidson JW, Sloboda JA. Innate talents: reality or myth? Behav Brain Sci. 1998;21:399–407
86. Ericsson KA. Why expert performance is special and cannot be extrapolated from studies of performance in the general population: a response to criticisms. Intelligence. 2014;45:81–103
87. Ericsson KA. Deliberate practice and the acquisition and maintenance of expert performance in medicine and related domains. Acad Med. 2004;79:S70–81
88. Choudhry NK, Fletcher RH, Soumerai SB. Systematic review: the relationship between clinical experience and quality of health care. Ann Intern Med. 2005;142:260–73
89. van de Wiel MWJ, Van den Bossche P, Janssen S, Jossberger H. Exploring deliberate practice in medicine: how do physicians learn in the workplace? Adv Health Sci Educ. 2011;16:81–95
90. Masunaga H, Horn J. Expertise and age-related changes in components of intelligence. Psychol Aging. 2001;16:293–311
91. Ericsson KA. How experts attain and maintain superior performance: implications for the enhancement of skilled performance in older individuals. J Aging Phys Act. 2000;8:366–72
92. Simpson SA, Gilhooly KJ. Diagnostic thinking processes: evidence from a constructive interaction study of electrocardiogram (ECG) interpretation. Appl Cogn Psychol. 1997;11:543–54
93. Wolf FM, Miller JG, Bozynski ME, Schlesinger A, Rosen DS, Weiner GM, Fliegel JE. Effects of experience and case difficulty on the interpretation of pediatric radiographs. Acad Med. 1994;69:S31–3
94. Ericsson KA, Nandagopal K, Roring RW. Toward a science of exceptional achievement: attaining superior performance through deliberate practice. Ann N Y Acad Sci. 2009;1172:199–217
95. Sloboda JAEricsson KA. The acquisition of musical performance expertise: deconstructing the ‘talent’ account of individual differences in musical expressivity. In: The Road to Excellence: The Acquisition of Expert Performance in the Arts and Sciences, Sports, and Games.. 1996 Hillsdale Lawrence Erlbaum Associates, Inc:107–26
96. Mayer RE. Applying the science of learning to medical education. Med Educ. 2010;44:543–9
97. Mamede S, van Gog T, Moura AS, de Faria RM, Peixoto JM, Rikers RM, Schmidt HG. Reflection as a strategy to foster medical students’ acquisition of diagnostic competence. Med Educ. 2012;46:464–72
98. Deslauriers L, Schelew E, Wieman C. Improved learning in a large-enrollment physics class. Science. 2011;332:862–4
© 2015 International Anesthesia Research Society
99. Cohen J Statistical Power Analysis for the Behavioral Sciences. 19882nd ed. Hillsdale L. Erlbaum Associates