Rosenshine’s Principle 4: Provide models: Providing students with models and worked examples can help them learn to solve problems faster.

“Research Findings

Students need cognitive support to help them learn to solve problems. The teacher’s modelling and thinking aloud while demonstrating how to solve a problem are examples of effective cognitive support…  ” (Rosenshine)

Proponents of inquiry-based learning and teaching problem-solving argue that we learn something more deeply when we discover it for ourselves and that students cannot become good problem solvers when they are just shown how to do things or just given facts.

Inquiry and problem-based learning models tend to operate on the randomness as genesis principle, which humans have to use when faced with a completely new type of problem, no background knowledge in long-term memory to deal with it, and no expertise (knowledgeable others) to borrow from. Under these conditions, we are forced to use a process of random trial and error to solve a problem. This can eventually work, and most scientific and technological breakthroughs have probably relied on this mode of problem-solving. However, it is a slow and inefficient way to learn and makes no sense in a school environment where experts are available, as well as information on the best ways to solve particular problems. So the question begs, why ask students to struggle their way to learning the best way to do something and risk error, misunderstanding and frustration when we can just teach it directly? 

Rosenshine points to research showing that the best way to teach students to solve problems is for the teacher (an expert) to take students, step by step, through their thinking as they model processes for working something out in a step-by-step way. Common sense, right? Well, schools and teacher educators have spent a lot of time and money convincing themselves that this way of teaching was not the way to go. An expert (teacher) demonstrating to novices (students), step by step, how they do something went out of vogue and was replaced by vague ideas of inquiry-based, student-led learning, relegating teachers to the position as facilitator of learning (a guide on the side) rather than an instructor (the sage on the stage). This disastrous era that began around the 1970s still endures in some schools, most of them very prestigious and expensive schools!

Education lost its way because of a fundamental misunderstanding of what makes good problem solvers. Teaching Problem-solving skills and asking students to inquire their way to knowing is based on the shaky premise that problem-solving is a discrete set of teachable skills and, when learned by students, can be applied across any area where a problem needs to be solved. Having students bumble their way to a solution to a problem is an ineffective way to get information into their long-term memories, and because problems are domain-specific, learning to fix a problem under the hood of a car will be of no assistance to solving an algebra problem. There are simply no transferable skills, routines or dispositions that cross over.

It turns out that experts in a field don’t have mystical, transferrable problem-solving skills that are applicable to any area. Instead, they have lots of hours of practice in a given area of knowledge. A brilliant physicist may be helpless with a flat tyre. Their physics brilliance would not transfer to fitting spare tyres. The flat tyre is a problem to be solved, but not in a domain familiar to the Physicist.

It turns out that experts have lots of knowledge stored in their long-term memory related to a particular domain. When faced with a novel problem or a problem with some unknown elements, they draw on these deep knowledge networks in their long-term memory and apply them to this new situation that has some similarities to a problem they’ve solved before. From the onlooker’s perspective, this might seem like creative thinking and problem-solving when, in fact, it’s the work of a highly experienced person who’s previously seen and solved lots of problems similar to this (just not this exact one) and made one small next step, (intellectual hop if you will), to solve the new(ish) problem.  

Studies on Chess masters showed that the best players have more board configurations stored in long-term memory than their less expert opponents. They use this vast store of board configurations to decide an excellent next move. They don’t think many moves ahead or possess fantastic creativity. They look at a board configuration and recognise it from experience, whereas their opponent may not. They know the best option. When two players compete, the better player will have more board configurations stored in long-term memory and will win more often because of this.  

Teachers have more subject-specific (domain-specific) content knowledge than their students. They have, stored in long-term memory, knowledge and procedures for domain-specific tasks, from writing a simple sentence to solving a complex calculus problem. Effective teaching involves presenting students with this declarative and procedural knowledge in well-sequenced manageable pieces by demonstrating them and then sharing the task with students, with the teacher doing a part of a task, students doing part, and then eventually, students doing the entire task independently under the watchful eye of the teacher. The teacher shows the students how an expert does something; the students borrow these expert routines and eventually, through guided practice, reorganise them in their own long-term memories. You may know this process as worked examples and the ‘I do’, ‘We do’, ‘You do’ approaches popularised by explicit and direct instruction models.  

This is more time-efficient than posing problems to students and asking them to solve them through trial and error. I’m not saying this has no educational merit; there’s much to learn from mistakes and missteps –ask any inventor or scientist. The problem with an instructional model dominated by this approach is that it wastes instructional time and risks students embedding misconceptions and less efficient ways to do tasks and procedures. Trial and error can also demotivate students who have insufficient domain-specific knowledge stored in long-term memory to know where to start.

Students need a robust subject-specific knowledge base to inquire and problem-solve from; otherwise, they flounder, and no amount of struggle will ever be productive.

Playberry Laser T1-2 is a teacher-supportive multisensory literacy resource for primary teachers to support their teaching in line with research. We’ve taken the planning and resource design load to free teachers to focus on building content knowledge and sharpening their delivery in line with Rosenshine’s Principles of Instruction.

Reference: 

Rosenshine, B. (2012). Principles of instruction: Research-based strategies that all teachers should know. [online] American Educator, American Educator, pp.12–39. Available at: https://www.aft.org/sites/default/files/periodicals/Rosenshine.pdf.