Every profession has training myths, but law enforcement carries some especially dangerous ones. Near the top of the list is the belief that if a drill feels smooth, productive, and confidence-building, then it must be effective. That assumption is comforting to students, instructors, and administrators alike. It is also one of the most persistent lies in police training. Because when the pressure is real, the cues are incomplete, and the environment turns chaotic, what felt best during practice is often what fails first. In other words, it is a lie because immediate performance is not the same thing as durable learning. And in police work, that distinction is not academic. It is operational, constitutional, and, in some cases, fatal. 

Robert Bjork’s work on memory and metamemory remains so important because it exposes not just how people learn, but how badly they often misjudge their own learning. In his 1994 publication, Memory and Metamemory Considerations in the Training of Human Beings, Bjork was not reporting one isolated experiment. He was synthesizing a broad body of research on learning, retention, retrieval, transfer, and self-assessment to answer a far more important question: what kinds of training conditions actually produce skill that survives after the training event is over? His answer was deeply inconvenient for traditional training culture. The very conditions that make performance look best during practice are often not the conditions that produce the strongest retention, recall, or transfer later (Bjork, 1994). 

That insight lands with particular force in policing. Our problem is not whether an officer can run a clean drill when the task is fresh, predictable, coached, and heavily supported. Our problem is whether that officer can still retrieve the right perception, judgment, and action when time has passed, the cues are incomplete, stress is rising, and the environment no longer resembles the training bay. Bjork’s central point was that training must do more than produce a stored memory. It must produce a memory that remains accessible across time and shifting contexts. That is where weak training gets exposed (Bjork, 1994). 

Bjork begins with a truth that many instructors still resist. The goal of training is not short-term performance inside training. The goal is long-term post-training performance in the environments that actually matter. He emphasizes that even relatively superficial changes in context can impair retrieval and performance. That should unsettle any agency still relying on static, familiar, and tightly controlled exercises as evidence of readiness. It is one thing for an officer to perform well in a known environment under ideal sequencing. It is something else entirely for that officer to recall and apply the same skill in a different setting, under pressure, with competing demands and imperfect information (Bjork, 1994). 

Bjork also reminds us that memory is not a recording device. Human beings do not store knowledge literally. We encode it by relating it to what we already know, and we retrieve it through cue-dependent, reconstructive processes. That matters because retrieval is not merely proof of learning. Retrieval is itself part of learning. When a person successfully recalls something under demanding conditions, that act strengthens later access. In practical terms, this means that watching, re-reading, and repeating under ideal support are not the same as retrieving under pressure. One difficult, effortful recall may do more for long-term performance than several smooth repetitions that never require the learner to solve anything (Bjork, 1975; Bjork, 1994). 

That is the heart of what later became widely known as “desirable difficulties.” Bjork showed that many training conditions that make learning feel worse actually make learning endure better. Variable practice, contextual interference, distributed practice, reduced feedback, and retrieval-based learning all tend to suppress performance during acquisition while improving retention and transfer afterward. That is the paradox. Training can look worse while teaching better. Conversely, it can look excellent while leaving behind very little that survives (Battig, 1979; Bjork, 1994; Schmidt & Bjork, 1992; Shea & Morgan, 1979). 

Why Difficult Training Produces Better Retention

When practice is blocked, predictable, and repeatedly cued, learners improve quickly. When feedback is constant and immediate, drills look cleaner. When repetitions are massed together, trainees often feel like they are locking in mastery. But Bjork’s synthesis makes clear that these conditions can act like crutches. They prop up performance during practice while weakening the learner’s ability to perform later without the supports. The learner mistakes fluency for competence. The instructor mistakes smoothness for mastery. The institution mistakes documentation for readiness (Bjork, 1994; Schmidt, 1991). 

Bjork’s metamemory argument is what makes this especially dangerous. Learners are often poor judges of what learning feels like. Rapid progress during practice feels like real learning. Struggle, delay, and error feel like failure. Yet those feelings often point in exactly the wrong direction. Trainees can come to prefer the training methods that produce the weakest long-term learning simply because those methods feel better while they are happening. Bjork cites the example of British postal workers learning keyboard skills. The employees trained under less efficient massed-practice conditions were more satisfied than those trained under the more efficient spaced-practice schedule. That single finding ought to shake every trainer who has ever used student satisfaction, confidence, or perceived smoothness as a proxy for quality (Baddeley & Longman, 1978; Bjork, 1994). 

In policing, that mistake is amplified by culture. Officers, like most learners, are naturally drawn to methods that make them feel competent. Agencies are naturally drawn to methods that are easy to administer, easy to score, easy to document, and easy to defend internally. Static drills, qualification strings, blocked repetitions, and constant coaching check all of those boxes. They produce visible order. They create fast progress. They make everyone feel like something meaningful is happening. But Bjork’s work warns that what feels best during acquisition may be exactly what teaches worst for real-world performance (Bjork, 1994). 

Police Training's False Sense of Competence

That is where the police motor learning literature becomes so important. Di Nota and Huhta argue that police training should be understood as complex motor learning, and they explicitly frame situational awareness and decision-making as essential motor skills in policing, not merely abstract cognitive add-ons. They describe how officers move from initial skill acquisition to higher proficiency through chunking, concatenation, and repeated reinforcement of meaningful action patterns. They also make a critical warning: without validated strategies, officers may encode ineffective patterns and stimulus-response tendencies instead of effective critical thinking skills (Di Nota & Huhta, 2019). 

That warning aligns almost perfectly with Bjork. Training that looks successful during acquisition can still be building the wrong thing. Anderson and his colleagues make the next part of the case. Their review on acute stress physiology and motor performance makes clear that high-level motor skills in policing are affected by autonomic, endocrine, and neuromuscular stress responses. Adaptive arousal can improve alertness and focus, but excessive stress degrades performance. Their conclusion is blunt: skill decay under high acute stress is inevitable, although evidence-informed training can mitigate it (Anderson et al., 2019). 

That matters because many agencies still train as if a skill demonstrated in calm conditions will automatically survive high-threat conditions. It will not. The nervous system does not work that way. The officer who performs beautifully when every variable is known and every step is rehearsed may still struggle when the situation becomes dynamic, ambiguous, and threatening. Stress changes perception. Stress changes attention. Stress changes timing. Stress changes movement. The profession ignores that at its own peril (Anderson et al., 2019). 

Stress Reveals What Training Actually Built

From a NeuralTac perspective, this is the center of gravity. NeuralTac begins with a premise that too many conventional training programs avoid: the brain under stress will not reveal what was merely rehearsed. It will reveal what was actually built. That means training design has to be organized around encoding, retrieval, discrimination, contextual variability, and adaptive execution. It means we cannot let student comfort become the design standard. It means we cannot let short-term polish masquerade as durable skill. 

That does not mean fundamentals do not matter. They do. Basic mechanics still have to be chunked, stabilized, and refined. Di Nota and Huhta describe how continuous streams of movement are broken down into manageable “chunks” and then recombined into longer sequences that can be performed with less mental effort (Di Nota & Huhta, 2019). 

But that is only the beginning. Once the learner has a workable internal model of the skill, the training environment must change. The learner has to retrieve more. Solve more. Adapt more. Perform under greater variation. Couple movement with language, perception, judgment, and changing environmental demands. Otherwise the officer may develop polished habits that exist only inside the artificial conditions that created them. 

This is also where naturalistic decision-making research helps sharpen the point. The literature makes an important distinction between the aspects of performance that should become automatic and the aspects that should not. Some task components must be automated so they consume less mental bandwidth during high-workload conditions. But decision-making itself cannot be reduced to rigid automaticity because real environments vary too much. In other words, we want officers to have reliable mechanics, communication habits, indexing, movement efficiency, and other component skills. We do not want them to replace judgment with canned response (Klein, 1998). 

That is not a small distinction. It is one of the dividing lines between effective preparation and institutional self-deception. 

From Learning Science to Legal Liability

The legal implications are impossible to ignore. Under 42 U.S.C. § 1983, officers and agencies face civil exposure when constitutional rights are violated under color of law. Under Monell, municipalities are not liable on a pure respondeat superior theory, but they may be liable when a constitutional injury is tied to an official policy, custom, or practice (Monell v. Department of Social Services, 1978). And under City of Canton v. Harris, failure to train can create municipal liability when it rises to deliberate indifference to the rights of persons with whom police come into contact (City of Canton v. Harris, 1989; 42 U.S.C. § 1983). 

That is a high standard, but it is not meaningless. It does not ask whether an agency provided some training. It asks whether the agency ignored an obvious need for training that actually prepares officers for recurring constitutional demands. That is why Popow v. City of Margate still matters. The case is not merely a historical citation. It is a warning. Training is not judged in the abstract. It is judged against the actual conditions in which officers are expected to perform. If an agency arms officers, sends them into uncertain public encounters, and relies on training models that maximize short-term performance while neglecting realistic transfer, stress adaptation, movement, backdrop, decision pressure, and environmental variability, that agency is not merely behind the science. It is building the very factual record plaintiffs’ lawyers know how to use (City of Canton v. Harris, 1989; Monell v. Department of Social Services, 1978; Popow v. City of Margate, 1979). 

This is where Bjork’s work becomes more than a learning-theory footnote. He gives us the language to explain why so much traditional police training looks good right up until the moment it matters. He explains why students can sincerely believe they are learning when they are mostly becoming familiar. He explains why instructors can be seduced by visible progress. He explains why administrative preference, student preference, and actual effectiveness may point in three different directions. And he explains why agencies that confuse performance with learning are not just making a pedagogical mistake. They may be manufacturing operational failure and legal risk at the same time (Bjork, 1994). 

Training for Reality, Not Appearance

One of the hardest responsibilities of a serious trainer is protecting productive struggle from institutional pressure. The student wants to feel successful. The instructor wants to see quick progress. The administrator wants clean documentation. The organization wants a training package that is easy to schedule, easy to measure, and easy to defend. But the profession cannot afford to let those pressures dictate training design. Good training often looks less flattering in the short term. It may feel slower. It may produce more errors. It may frustrate students who mistake effort for failure. That does not make it bad. It may mean it is finally doing the work that real learning requires (Bjork, 1994; Schmidt & Bjork, 1992). 

There is also an ethical dimension here. False confidence is dangerous in police work because it distorts behavior at exactly the moment honest self-appraisal is most needed. The officer who understands the limits of his readiness may move with appropriate caution. The officer who has been trained under flattering conditions may walk into a crisis with a confidence his nervous system has not earned. That is not just a training defect. It is a public safety problem. 

The lesson here is uncomfortable, but it is not complicated. What students prefer is not always what helps them most. What instructors enjoy delivering is not always what transfers best. What administrators can easily measure is not always what matters. Bjork’s findings force us to confront that reality. The police performance literature confirms it. NeuralTac gives us a framework for operationalizing it. And liability doctrine reminds us that agencies that ignore it do so at their own risk (Anderson et al., 2019; Bjork, 1994; Di Nota & Huhta, 2019). 

The profession does not need more training that looks polished and collapses under stress. It needs training that survives contact with reality. That kind of training is usually harder to build, harder to administer, and harder for students to love in the moment. But it is also the kind that protects officers, protects the public, and stands on far firmer ground when the courts come asking what the agency knew, what it trained, and whether the failure was obvious all along. 

References 

Anderson, G. S., Di Nota, P. M., Metz, G. A. S., & Andersen, J. P. (2019). The impact of acute stress physiology on skilled motor performance: Implications for policing. Frontiers in Psychology, 10, Article 2501. 

Baddeley, A. D., & Longman, D. J. A. (1978). The influence of length and frequency of training session on the rate of learning to type. Ergonomics, 21(8), 627-635. 

Bahrick, H. P. (1979). Maintenance of knowledge: Questions about memory we forgot to ask. Journal of Experimental Psychology: General, 108(3), 296-308. 

Battig, W. F. (1979). The flexibility of human memory. In L. S. Cermak & F. I. M. Craik (Eds.), Levels of processing in human memory (pp. 23-44). Erlbaum. 

Bjork, R. A. (1975). Retrieval as a memory modifier. In R. L. Solso (Ed.), Information processing and cognition: The Loyola symposium (pp. 123-144). Erlbaum. 

Bjork, R. A. (1994). Memory and metamemory considerations in the training of human beings. In J. Metcalfe & A. Shimamura (Eds.), Metacognition: Knowing about knowing (pp. 185-205). MIT Press. 

Bjork, R. A., & Bjork, E. L. (1992). A new theory of disuse and an old theory of stimulus fluctuation. In A. Healy, S. Kosslyn, & R. Shiffrin (Eds.), From learning processes to cognitive processes: Essays in honor of William K. Estes (Vol. 2, pp. 35-67). Erlbaum. 

Bransford, J. D., Franks, J. J., Morris, C. D., & Stein, B. S. (1979). Some general constraints on learning and memory research. In L. S. Cermak & F. I. M. Craik (Eds.), Levels of processing in human memory (pp. 331-354). Erlbaum. 

City of Canton v. Harris, 489 U.S. 378 (1989). 

Connick v. Thompson, 563 U.S. 51 (2011). 

Di Nota, P. M., & Huhta, J.-M. (2019). Complex motor learning and police training: Applied, cognitive, and clinical perspectives. Frontiers in Psychology, 10, Article 1797.

Landauer, T. K., & Bjork, R. A. (1978). Optimum rehearsal patterns and name learning. In M. M. Gruneberg, P. E. Morris, & R. N. Sykes (Eds.), Practical aspects of memory (pp. 625-632). Academic Press.