Effective Learning Does Not Emulate the Professional Workplace

The most effective learning techniques require substantial cognitive effort from students and typically do not emulate what experts do in the professional workplace. Direct instruction is necessary to maximize student learning, whereas unguided instruction and group projects are typically very inefficient.

This post is part of the book The Math Academy Way (Working Draft, Jan 2024). Suggested citation: Skycak, J., advised by Roberts, J. (2024). Effective Learning Does Not Emulate the Professional Workplace. In The Math Academy Way (Working Draft, Jan 2024). https://justinmath.com/effective-learning-does-not-emulate-the-professional-workplace/

Want to get notified about new posts? Join the mailing list and follow on X/Twitter.

It’s a common myth that effective methods of practice emulate what experts do in the professional workplace.

In reality, a well-known phenomenon in cognitive psychology is that instructional techniques that promote the most learning in experts, promote the least learning in beginners, and vice versa. This is called the expertise reversal effect (first introduced by Sweller et al., 2003). As Kirschner & Hendrick summarize (2024, pp.67):

“As the novice is not a miniature expert, it’s important to realize that what may work very well for an expert (e.g. discovery learning, problem-based learning [in the sense of working in groups to solve an open-ended problem], inquiry learning) usually doesn’t work well or is even harmful and counterproductive for the novice (and vice versa).”

Additionally, in the professional workplace, employees engage in activities that maximize group output, which is totally different – and in some ways, opposite – from maximizing individual learning.

Direct Instruction is Needed

Definition and Importance

It is true that many highly skilled professionals spend a lot of time solving open-ended problems, and in the process, discovering new knowledge as opposed to obtaining it through direct instruction. However, this does not mean that beginners should do the same. The expertise reversal effect suggests the opposite – that beginners (i.e. students) learn most effectively through direct instruction.

Direct instruction is intuitively obvious. If a coach is trying to get a student to become a great chess player or pianist, they don’t tell the student “go play around and come back with something insightful.” Rather, the coach explicitly demonstrates a skill and then provides corrective feedback to the student as they practice the skill. As Kirschner & Hendrick describe (2024, pp.68):

“While an expert can be given a problem to be solved after having been taught a certain technique or principle, a novice should be given a more structured approach to using that principle for solving the same problem, for example in the form of a worked example.”

Indeed, this is backed up by decades of research. As prominent psychologists Richard Clark, Paul Kirschner, and John Sweller summarize (2012):

*“Decades of research clearly demonstrate that for novices (comprising virtually all students), direct, explicit instruction is more effective and more efficient than partial guidance. So, when teaching new content and skills to novices, teachers are more effective when they provide explicit guidance accompanied by practice and feedback, not when they require students to discover many aspects of what they must learn.
…
We also have a good deal more experimental evidence [since the 1960s] as to what constitutes effective instruction: controlled experiments almost uniformly indicate that when dealing with novel information, learners should be explicitly shown all relevant information, including what to do and how to do it. We wonder why many teacher educators who are committed to scholarship and research ignore the evidence and continue to encourage minimal guidance when they train new teachers.

After a half century of advocacy associated with instruction using minimal guidance, it appears that there is no body of sound research that supports using the technique with anyone other than the most expert students. Evidence from controlled, experimental (a.k.a. “gold standard”) studies almost uniformly supports full and explicit instructional guidance rather than partial or minimal guidance for novice to intermediate learners. These findings and their associated theories suggest teachers should provide their students with clear, explicit instruction rather than merely assisting students in attempting to discover knowledge themselves.”*

Unguided Instruction has a History of Pseudoscience

Clark, Kirschner, & Sweller (2012) explain that unguided instruction persists by cloaking itself in a different disguise each time it is debunked:

*“Richard Mayer (a cognitive scientist at the University of California, Santa Barbara) examined evidence from studies conducted from 1950 to the late 1980s comparing pure discovery learning (defined as unguided, problem-based instruction) with guided forms of instruction. He suggested that in each decade since the mid-1950s, after empirical studies provided solid evidence that the then-popular unguided approach did not work, a similar approach soon popped up under a different name with the cycle repeating itself.

Each new set of advocates for unguided approaches seemed unaware of, or uninterested in, previous evidence that unguided approaches had not been validated. This pattern produced discovery learning, which gave way to experiential learning, which gave way to problem-based and inquiry learning, which has recently given way to constructivist instructional techniques.”*

As they elaborate elsewhere (Kirschner, Sweller, & Clark, 2006), these unguided approaches are often based on modeling the activities of professionals:

“Examples of applications of these differently named but essentially pedagogically equivalent approaches include science instruction in which students are placed in inquiry learning contexts and asked to discover the fundamental and well-known principles of science by modeling the investigatory activities of professional researchers (Van Joolingen, de Jong, Lazonder, Savelsbergh, & Manlove, 2005).”

They also explain that the current formulation, constructivist instruction, uses scientific camouflage but is not actually scientific itself:

*“Turning again to Mayer’s review of the literature, many educators confuse ‘constructivism,’ which is a theory of how one learns and sees the world, with a prescription for how to teach.

In the field of cognitive science, constructivism is a widely accepted theory of learning; it claims that learners must construct mental representations of the world by engaging in active cognitive processing. Many educators (especially teacher education professors in colleges of education) have latched on to this notion of students having to ‘construct’ their own knowledge, and have assumed that the best way to promote such construction is to have students try to discover new knowledge or solve new problems without explicit guidance from the teacher.

Unfortunately, this assumption is both widespread and incorrect. Mayer calls it the ‘constructivist teaching fallacy.’… Learning requires the construction of knowledge. Withholding information from students does not facilitate the construction of knowledge.”*

In his critical review, Mayer (2004) had plenty more to say:

“The research in this brief review shows that the formula constructivism = hands-on activity is a formula for educational disaster.
…
Like some zombie that keeps returning from its grave, pure discovery continues to have its advocates.
…
Pure discovery did not work in the 1960s, it did not work in the 1970s, and it did not work in the 1980s, so after these three strikes, there is little reason to believe that pure discovery will somehow work today.
…
[T]he issue addressed in this article is not whether constructivism is a good idea for education. but rather whether the educational implications attributed to constructivism are good ideas. In the case of discovery methods, the implications attributed to constructivism are not good ideas.
…
The debate about discovery has been replayed many times in education, but each time, the research evidence has favored a guided approach to learning.”

These interpretations are echoed throughout the literature. As other prominent psychologists John Anderson, Lynne Reder, and Herbert Simon state (1998):

*“A consensus exists within cognitive psychology that people do not record experience passively but interpret new information with the help of prior knowledge and experience… However, denying that information is recorded passively does not imply that students must discover their knowledge by themselves, without explicit instruction, as claimed by radical constructivists.”

“Radical constructivism emphasizes discovery learning, learning in complex situations, and learning in social contexts, while strongly distrusting systematic evaluation of educational outcomes… [C]ertain of its devotees exhibit an antiscience bias that, should it prevail, would devote any hope for progress in education.
…
Little positive evidence exists for discovery learning and it is often inferior. Discovery learning, even successful in enabling the acquisition of the desired construct, may require a great deal of valuable time that could have been spent practicing the construct (which is an active process, too) if it had been learned from instruction. Because most learning only takes place after the construct has been discovered, when the search is lengthy or unsuccessful, motivation commonly lags. As D. P. Ausubel wrote in 1968, summarizing the findings from the research on discovery learning:

‘Actual examination of the research literature allegedly supportive of learning by discovery reveals that valid evidence of this nature is virtually nonexistent. It appears that the various enthusiasts of the discovery method have been supporting each other research-wise by taking in each other’s laundry, so to speak, that is, by citing each other’s opinions and assertions as evidence and by generalizing wildly from equivocal and even negative findings.’”*

Unguided Instruction is Logically and Scientifically Inconsistent

Anderson, Reder, & Simon (1998) also explain that opponents of direct instruction are, ultimately, opponents of extensive practice – a position that is clearly problematic:

*“Some argue that direct instruction leads to ‘routinization’ of knowledge and drives out understanding:

‘The more explicit I am about the behavior I wish my students to display, the more likely it is that they will display the behavior without recourse to the understanding which the behavior is meant to indicate; that is, the more likely they will take the form for the substance.’

An extension of this argument is that excessive practice will also drive out understanding. This criticism of practice (called ‘drill and kill,’ as if this phrase constituted empirical evaluation) is prominent in constructivist writings. Nothing flies more in the face of the last 20 years of research than the assertion that practice is bad.

All evidence, from the laboratory and from extensive case studies of professionals, indicates that real competence only comes with extensive practice. By denying the critical role of practice, one is denying children the very thing they need to achieve competence… the grain of truth in the drill-and-kill criticisms [is that]: Students need to be engaged when they are studying.”*

Likewise, there are critical issues with the idea of learning primarily from complex situations:

*“First, a learner who is having difficulty with many of the components can easily be overwhelmed by the processing demands of the complex task. Second, to the extent that many components are well mastered, the student will waste a great deal of time repeating those mastered components to get an opportunity to practice the few components that need additional practice.

A large body of research in psychology shows that part training is often more effective when the part component is independent, or nearly so, of the larger task… Practicing one’s skills periodically in full context is important to motivation and to learning to practice, but not a reason to make this the principal mechanism of learning.”*

Along these lines, Clark, Kirschner, & Sweller (2012) further explain that, in addition to being supported by a mountain of experimental evidence, the superiority of direct instruction follows intuitively from modern understandings of working and long-term memory:

“These two facts — that working memory is very limited when dealing with novel information, but that it is not limited when dealing with organized information stored in long-term memory — explain why partially or minimally guided instruction typically is ineffective for novices, but can be effective for experts. When given a problem to solve, novices’ only resource is their very constrained working memory. But experts have both their working memory and all the relevant knowledge and skill stored in long-term memory.”

As Sweller, Clark, and Kirschner (2010) elaborate elsewhere:

*“Recent ‘reform’ curricula both ignore the absence of supporting data and completely misunderstand the role of problem solving in cognition. If, the argument goes, we are not really teaching people mathematics but rather are teaching them some form of general problem solving, then mathematical content can be reduced in importance. According to this argument, we can teach students how to solve problems in general, and that will make them good mathematicians able to discover novel solutions irrespective of the content.

We believe this argument ignores all the empirical evidence about mathematics learning. Although some mathematicians, in the absence of adequate instruction, may have learned to solve mathematics problems by discovering solutions without explicit guidance, this approach was never the most effective or efficient way to learn mathematics.
…
[L]ong-term memory, a critical component of human cognitive architecture, is not used to store random, isolated facts but rather to store huge complexes of closely integrated information that results in problem-solving skill. That skill is knowledge domain-specific, not domain-general. An experienced problem solver in any domain has constructed and stored huge numbers of schemas in long-term memory that allow problems in that domain to be categorized according to their solution moves.

In short, the research suggests that we can teach aspiring mathematicians to be effective problem solvers only by providing them with a large store of domain-specific schemas. Mathematical problem-solving skill is acquired through a large number of specific mathematical problem-solving strategies relevant to particular problems. There are no separate, general problem-solving strategies that can be learned.
…
Whereas a lack of empirical evidence supporting the teaching of general problem-solving strategies in mathematics is telling, there is ample empirical evidence of the validity of the worked-example effect. A large number of randomized controlled experiments demonstrate this effect (e.g., Schwonke et al., 2009; Sweller & Cooper, 1985). For novice mathematics learners, the evidence is overwhelming that studying worked examples rather than solving the equivalent problems facilitates learning.

Studying worked examples is a form of direct, explicit instruction that is vital in all curriculum areas, especially areas that many students find difficult and that are critical to modern societies. Mathematics is such a discipline. Minimal instructional guidance in mathematics leads to minimal learning (Kirschner, Sweller, & Clark, 2006).”*

Unguided Instruction Leads to Major Issues in Practice

Clark, Kirschner, & Sweller (2012) also describe what actually happens in classrooms that do not use direct instruction:

*“In real classrooms, several problems occur when different kinds of minimally guided instruction are used.

First, often only the brightest and most well-prepared students make the discovery.

Second, many students, as noted above, simply become frustrated. Some may disengage, others may copy whatever the brightest students are doing — either way, they are not actually discovering anything.

Third, some students believe they have discovered the correct information or solution, but they are mistaken and so they learn a misconception that can interfere with later learning and problem solving. Even after being shown the right answer, a student is likely to recall his or her discovery — not the correction.

Fourth, even in the unlikely event that a problem or project is devised that all students succeed in completing, minimally guided instruction is much less efficient than explicit guidance. What can be taught directly in a 25-minute demonstration and discussion, followed by 15 minutes of independent practice with corrective feedback by a teacher, may take several class periods to learn via minimally guided projects and/or problem solving.”*

These issues are also backed up by numerous studies:

“Hardiman, Pollatsek, and Weil (1986) and Brown and Campione (1994) noted that when students learn science in classrooms with pure-discovery methods and minimal feedback, they often become lost and frustrated, and their confusion can lead to misconceptions. Others (e.g., Carlson, Lundy, & Schneider, 1992; Schauble, 1990) found that because false starts are common in such learning situations, unguided discovery is most often inefficient.”

To emphasize, these issues are so problematic that they can actually result in negative educational progress:

“Not only is unguided instruction normally less effective; there is also evidence that it may have negative results when students acquire misconceptions or incomplete or disorganized knowledge.”

But despite these issues, the students who learn least in unguided settings still tend to prefer it because it feels less effortful:

”…[W]hen learners are asked to select between a more-guided or less-guided version of the same course, less-skilled learners who choose the less-guided approach tend to like it even though they learn less from it. It appears that guided instruction helps less-skilled learners by providing task-specific learning strategies. However, these strategies require learners to engage in explicit, attention-driven effort and so tend not to be liked, even though they are helpful to learning.”

Of course, experienced, effective teachers are well acquainted with these issues and (rightfully so) brush off any recommendations to use unguided learning:

”…[M]any experienced educators are reluctant to implement — because they require learners to engage in cognitive activities that are highly unlikely to result in effective learning. As a consequence, the most effective teachers may either ignore the recommendations or, at best, pay lip service to them (e.g., Aulls, 2002).”

This sentiment is sharply echoed by Mayer (2004):

”…[T]he contribution of psychology is to help move educational reform efforts from the fuzzy and unproductive world of educational ideology — which sometimes hides under the banner of various versions of constructivism — to the sharp and productive world of theory-based research on how people learn.”

To top it all off, as Kirschner, Sweller, & Clark (2006) summarize, even on the rare occasion that a student does manage to learn in an unguided setting, their learning tends to be shallower than it would have been in a strongly guided setting:

*“Moreno (2004) concluded that there is a growing body of research showing that students learn more deeply from strongly guided learning than from discovery. Similar conclusions were reported by Chall (2000), McKeough, Lupart, and Marini (1995), Schauble (1990), and Singley and Anderson (1989).

Klahr and Nigam (2004), in a very important study, not only tested whether science learners learned more via a discovery versus direct instruction route but also, once learning had occurred, whether the quality of learning differed. Specifically, they tested whether those who had learned through discovery were better able to transfer their learning to new contexts. The findings were unambiguous. Direct instruction involving considerable guidance, including examples, resulted in vastly more learning than discovery. Those relatively few students who learned via discovery showed no signs of superior quality of learning.”*

As Kirschner & Hendrick summarize (2024, pp.76):

”…[I]f you want your students to learn to solve problems, they first need both the declarative and procedural knowledge within the subject area of the problem in question. This is also true if you want to teach them to communicate, discuss, write, or whatever twenty-first century skill people are talking about. You can’t communicate about something, write about something, discuss or argue about something, etc., without first knowing about that something and then also knowing the rules (i.e. the procedures) for doing it.”

Many Hands Make Light Work… and Light Learning

Professionals often work in groups because it gives them an economic advantage. Real-world projects are often extremely complex and require a massive amount of highly skilled labor across a wide variety of disciplines. The amount of work necessary to bring the project to fruition might exceed what one person can put forth over their entire lifetime, and the number of skill domains covered by the work might be more than any one person can hope to master in a single lifetime. This problem is solved by constructing a team where each member is highly skilled in one or more of the relevant domains, and there are enough members to complete the workload in a feasible amount of time.

The goal of division of labor in the professional workplace is to maximize the output of a team. On the surface, it might seem like a tempting strategy to apply in the classroom: won’t maximizing the output of a classroom effectively maximize the learning of individual students? But the answer is a resounding no. Division of labor is division of learning, which means that it actually minimizes the learning of individual students.

To maximize the learning of individual students, it is necessary to actively engage every individual student on every single piece of material to be learned. Division of labor is the complete opposite of that, since each student actively learns only the material that corresponds to their individual responsibility in the division of labor. The rest of the project, they observe only passively, if at all. At best, each student only learns a tiny fraction of the material. At worst, one student ends up doing all the work while the rest of the group learns nothing.

As Anderson, Reder, & Simon (1998) summarize:

*“Some of the learning contexts recommended in radical constructivist writings involve tasks that can be solved by a single problem solver, but the movement more and more is to convert these to group learning situations… While a person must learn to deal with the social aspects of jobs, all skills required for these jobs do not need to be trained in a social context… Training independent parts of a task separately is preferable, because fewer cognitive resources will be required for performance, thereby reserving adequate capacity for learning.
…
A review by the National Research Council (NRC) Committee on Techniques for the Enhancement of Human Performance noted that… relatively few studies ‘have successfully demonstrated advantages for cooperative versus individual learning,’ and that a number of detrimental effects arising from cooperative learning have been identified — the ‘free rider,’ the ‘sucker’ [reducing effort to avoid being taken advantage of by free riders], the ‘status differential’ [low-ability team members lose social status and reduce effort] and ‘ganging up’ [directing group effort towards circumventing the intended efforts of the task] effects [Salomon & Globerson, 1989].

The NRC review of cooperative learning notes a substantial number of reports of no-differences, but, unfortunately, a huge number of practitioner-oriented articles about cooperative learning gloss over difficulties with the approach and treat it as an academic panacea. It is applied too liberally without the requisite structuring or scripting to make it effective… A reported practice among some students is to divide the labor across classes so that one member of a group does all of the work for a project in one class, while another carries the burden for a different class. Clearly these are not the intended outcomes of cooperative learning but will occur if thoughtful implementation and scripting of the learning situation are not evident.”*

Granted, fun, collaborative group activities can sometimes be useful for increasing student motivation and softening the discomfort associated with intense, individualized deliberate practice. However, they do not directly move the needle on student performance – rather, they “grease the wheels” and reduce psychological friction during the process of deliberate practice. Performance improvements come directly from deliberate practice.

References

Anderson, J. R., Reder, L. M., Simon, H. A., Ericsson, K. A., & Glaser, R. (1998). Radical constructivism and cognitive psychology. Brookings papers on education policy, (1), 227-278.

Clark, R., Kirschner, P. A., & Sweller, J. (2012). Putting students on the path to learning: The case for fully guided instruction. American Educator, 36 (1), 5-11.

Kirschner, P., & Hendrick, C. (2024). How learning happens: Seminal works in educational psychology and what they mean in practice. Routledge.

Kirschner, P. A., Sweller, J., & Clark, R. E. (2006). Why minimal guidance during instruction does not work: An analysis of the failure of constructivist, discovery, problem-based, experiential, and inquiry-based teaching. Educational psychologist, 41 (2), 75-86.

Mayer, R. E. (2004). Should there be a three-strikes rule against pure discovery learning?. American psychologist, 59 (1), 14.

Sweller, J., Ayres, P. L., Kalyuga, S., & Chandler, P. (2003). The expertise reversal effect. Educational Psychologist, 38 (1), 23-31.

Sweller, J., Clark, R., & Kirschner, P. (2010). Teaching general problem-solving skills is not a substitute for, or a viable addition to, teaching mathematics. Notices of the American Mathematical Society, 57 (10), 1303-1304.

Want to get notified about new posts? Join the mailing list and follow on X/Twitter.