TLDR: Project Based Learning (PBL) might satisfy the emotional and social aspects of being in a classroom but bypasses the cognitive conditions required for meaningful learning.
I've worked with PBL for over 15 years- first as a teacher at High Tech High and the last ten as someone supporting schools to use PBL. I actually hate (like really, hate) any article that uses PBL in a way that assumes we're discussing the same thing. There is such a wide range of understanding what PBL is, how it's implemented, the culture it's taught in, the conditions of the school and the purpose of why it's used in the first place.
I think it's more helpful to look at a specific project and talk about what you see (and don't see) in the project itself. For me, it's difficult to look at a project like Beyond the Crossfire (https://www.hightechhigh.org/hthcv/project/beyond-the-crossfire/) and not see examples of students learning rigorous content before applying it. Looking at this project specifically, there is a high real-world application (students are film-makers) but also high levels of scaffolding to support the development of the skills required for them to take on these roles.
This is an example of great PBL to me and one that I (and many others who have worked in this space for decades) have argued is the quality-standard we should PBL to...
I'd love to look at a specific project with you and talk more- about what's there and also honestly about what's not.
Thank you for sharing this. I appreciate the way you’re grounding the discussion in a specific project rather than PBL as an abstract ideal.
In the "Beyond the Crossfire" project, I can clearly see strong elements: purposeful application, authentic roles, and implementation of complex skills. I also agree that this is very different from the loosely defined versions of PBL that often circulate under the same label.
One thing I’m genuinely curious about, though (and this is where my interest tends to sit) is assessment. The project description does a good job articulating learning goals across skills, content, and mindsets, but it’s less clear on how those were evaluated:
a. How was students’ conceptual understanding assessed independently of the final product?
b. How did teachers distinguish between strong collaboration or production quality and actual content mastery?
c. What evidence was used to decide whether learning goals were met, beyond successful completion?
I’m asking this not to dismiss the project. But this is usually where discussions of PBL become harder to pin down. I’d enjoy unpacking a project together at that level: what’s explicit, what’s inferred, and what evidence of learning we can confidently point to.
Thank you for the thoughtful reply. I would really be open to talking any time about a specific project or assessment in general. Please feel free to send me an email at loni@imagineif.dk and we can find a time.
To address your questions about assessment specifically, I would ask: how are these things assessed in non-PBL classrooms?
Most of the time in PBL schools, it's done in pretty much the same way. There are PBL teachers who give quizzes, tests, require documentation on Digital Portfolios, assign essays and have rigorous formative and summative assessment benchmarks by using rubrics and other success criteria. I used to have one-on-one sessions with each student after a project was completed where we looked at their Digital Portfolio, final product and I made sure the academic learning goals were reached. I called them "Teach Back" sessions. Sometimes Presentations of Learning are also used or the Exhibition itself becomes an exam of sorts.
The "unicorn" PBL projects have great products that require rigorous academic learning to do them. In those cases, final products can definitely be an assessment tool in itself. A good "rule of thumb" is to see if students can make the product on the first day of the project. If they can- the product is too fluffy and the likelihood of deep learning is low. A vast majority of PBL projects I see have products that are mismatched to the rigor of learning required of the subject, have low criteria for success or have a high degree of choice for kids which can lead to poor academic integration. My experience is that while kids can be naturally curious, they tend to choose the same content or products when given a choice and the quality/depth of that work rarely increases on it's own.
As I wrote before, the definition of PBL matters here and the context in which it's being done also matters. I work with independent schools that have a lot of freedom in their need for learning outcomes and can take PBL in a very loose and child-led manor. Other schools exist in traditional systems where learning outcomes matter a lot to individual children and the schools themselves- so the flavor of PBL will be different.
So would it be fair to say that broadly (except very few contexts) PBL should be used from middle school onwards, and primarily in high school to balance the monotony and pressure of competitive exams?
That’s a reasonable way to think about it, but with an important caveat.
It’s less about age or grade level and more about learner expertise, task complexity, and the kind of learning goal involved. As students gain stable prior knowledge and conceptual fluency, they might benefit more from project-based work.
PBL can play a role in middle and high school especially for application, integration, and transfer but not a substitute for building foundational knowledge, nor as a remedy for exam pressure or classroom monotony. Those are system problems, not instructional ones, don't you think you?
The key question isn’t when PBL starts, but what knowledge students already have and what kind of thinking we’re asking them to do. Hope that makes sense.
I really enjoyed your take on this. I have seen evidence of what you discussed in teachers’ classrooms. It looks like learning but the learning part is not always the center. What can be copied or pasted in? What can I do to make this look pretty? Seem to be in front of the harder parts of learning. One classic example is having students plan a dream trip. You could say they are multiplying decimals, but that isn’t where most of the work is. Most of it is finding the places and prices, often with that assignment students don’t have a budget or understanding of what could be reasonably accomplished in a trip. You could say they are learning geography but often they are just regurgitating what they are finding online. It is important when we plan projects or lessons we see where the work actually is and make sure that matches our learning target.
I wonder what is your evidence to back up these claims. I've spent several years doing research on PBL in science classrooms. It is not a perfect method and it needs to be done by skilled teachers, but most actual studies shows that when it is done right, it supports students' cognitive learning just as good as traditional classroom learning, with the added value of social and emotional engagement, and self-efficacy. Several reviews confirm these claims. You easily dismiss the benefits of PBL as a whole, without any concrete evidence from field studies- I find this populistic and dangerous- though I will be happy to be proven wrong...
Hi Tom, I appreciate this push for evidence. When we look at the 'evidence,' we have to distinguish between Social-Emotional outcomes (self-efficacy, engagement) and Cognitive Architecture (long-term memory encoding).
This post is my conceptual argument majorly based on the Architecture of Complexity (Herbert Simon) and Cognitive Load Theory (Sweller). These are descriptions of the working memory bottleneck of the human mind. Research on CLT and Cog Sci shows that for a novice, a 'noisy' environment like PBL creates a heavy extraneous cognitive load that actively competes with the learning goal.
You mention that when PBL is "done right," it works. But "doing it right" usually involves the teacher providing massive amounts of Direct Instruction and Scaffolding within the project. At that point, is it still the project doing the teaching, or is it the structured instruction hidden inside the project?
My argument isn't that PBL has ZERO value; it’s that it is an Integration Tool, not an Instructional Tool.
Hi Umes, thank you for the thoughtful reply. I agree with all the theories you've mentioned, but it still does not include any actual empirical studies that were performed on PBL in classrooms, and there are plenty of those from the last 30 years, with mostly positive results. Of course, it requires direct instruction in the early stages of adoption from the teacher, like every pedagogical approach. As I said, PBL is not perfect, it is just another pedagogical approach, and a powerful one, that teachers should have in their tool kit. In general, the argument that the most effective teaching approach is the traditional 'pounding' of materials into students' brains is wrong for many reason, despite the current 'populist' wave of bashing against any other approach that had been emprically established in recent years. And above all, the question is always what are your goals and what you are measuring. If you seek to improve students' scores in traditional memorization and basic problem solving, then by all means, you should only teach use teacher-centered traditional excersize-based methods. And again, I appreciate the discussion and open for new evidence-based arguments.
Thank you for writing this op-ed. As a teacher working at a high school that runs on a Project Based Learning model with inter-disciplinary multi-graded studios, I would like to bring some counter points to some of your claims.
-Education is contextual, blanket claiming something to work, or not, most of the time seems far-fetched. Project Based Learning is a tool. How you integrate it, how much of your curriculum is shaped around it, how much agency does the teacher have in co-creating this course, are all important foundational questions to measure if it works for your context or not? Our school's studio curriculum is created around 6 academic pillars:
For example for this trimester, the studios being offered to students: Local Agents of Change (inspired from Marshall Ganz's Public Narrative), Politics on your Plate (Ceramics class for food combinations), Matter Matters (Material Science for Climate Change).
- Yes, there is tendency of Project Based Learning to have a higher weigh towards short term excitement. And that is why we try to use student created reflections and documentation as the most important part of any project, in order to bring some balance of long term mastery. In fact, we define mastery for our students as when they are ready to become teachers to other students, and have a dedicated block of time for peer learning groups for students to choose what they want to learn or teach.
-And in no way I am claiming that any of this works perfectly. We run into our challenges daily, as all schools do. And that's what we do as educators, come together and trying to figure out this inter generational information transfer task at a pivotal point in their brain development.
-To your sub-assembly point, there might be truth to that. We may START with building with shadows and be playing with the breadth a bit, which I also feel is the point of high school, to be interested in how the real world interacts. But once a student is hooked, I can be the worst teacher and the student still wants to solve the problem so they will extract the information out of me. And the certain depth required for technical and soft skills, we make sure students have enough required and elective studios to attend from which will allow them to find their interests from the breadth and grow deeper in them.
Scratch is one of the most beautiful learning platforms made to learn programming. Why does it work at that scale? Low barrier to entry, but the ceiling is limitless. A month ago I demonstrated a speech therapy demo to a early childhood school on Scratch, which took me 15 mins to build.
-And again, none of this is being prescribed as 100% works for all. For some students it can be truly life changing, and for some it can be a very tough time. Does that mean you abandon the tool itself? Or understand it as a tool and hope to use it for its merits and avoid its pitfalls.
Thank you for sharing this detailed perspective from the field. It sounds like your Studios are vibrant environments, and I appreciate your honesty about the daily challenges.
I want to zoom in on your point about 'building with shadows' to get students hooked. From a Herbert Simon perspective, the concern isn't about the hook; it’s about the cost of the detour. If we start with breadth and shadows, we are essentially placing a novice on a 'noisy environment' and hoping their interest will be strong enough to help them navigate the clutter.
For a highly motivated and skilled student, this usually works. But for the student who hasn't mastered the 'stable sub-assemblies' (the math, the syntax, the logic), the shadows never become solid.
Regarding Scratch, I think it actually supports the 'Artificial Learning' argument! Scratch is successful precisely because it decomposes programming. It removes the 'noise' of syntax errors so students can focus on the 'sub-assembly' of logic. It is a beautifully designed, simplified artificial environment. Not a messy, 'real-world' one.
My argument isn't that we should 'abandon PBL as a tool', but that we must stop using PBL to teach the foundations. Projects are for the Integration phase. If we use them as the Instruction phase, we are simply asking novices to act like experts. And cognitive science is clear on this: novices can't think like experts.
I agree with the concern about the illusion of learning, but I disagree that project based learning is the root cause. In my Systems Analysis and Project Management capstone, projects were not used in place of instruction. They were used to operationalize explicitly taught skills. Students were taught scope definition, requirements analysis, dependency mapping, risk identification, governance, and decision rights before engaging in the project.
The capstone was intentionally work relevant to the point that students’ employers flat out encouraged them to prioritize the course because of its immediate applicability. Students also reported continued use of these skills on the job well after the course ended. In this context, the project functioned as a structured application environment with feedback and accountability, not a discovery exercise. When project based learning fails, it is usually a systems design failure involving missing scaffolding, weak assessment, or unclear objectives, rather than an inherent flaw in learning through projects.
Thank you for sharing this. I think this is a brilliant example of what I would call Project-As-Integration rather than Project-As-Discovery.
You’ve actually perfectly illustrated Herbert Simon’s point about 'Near Decomposability.' By explicitly teaching (scope definition, risk identification, and dependency mapping) before the project, the students had opportunities to develop stable sub-skills.
The reason your capstone succeeded (and where most primary school PBL fails) is the Expert-Novice Fallacy. Your students were likely at a stage where those sub-skills were becoming automatic, freeing up their working memory to handle the 'noise' of a real-world project.
In many schools, we do the opposite: we throw novices into the 'noise' before they have the gears. We ask them to 'discover' the laws of physics while building the bridge.
The Systems Design approach you mentioned is exactly the kind of Artificial Learning Simon praised. Something that is highly structured, explicitly taught, and intentionally scaffolded. The 'illusion of learning' only happens when we skip the instruction and scaffolding and hope the project does the teaching for us.
I've worked with PBL for over 15 years- first as a teacher at High Tech High and the last ten as someone supporting schools to use PBL. I actually hate (like really, hate) any article that uses PBL in a way that assumes we're discussing the same thing. There is such a wide range of understanding what PBL is, how it's implemented, the culture it's taught in, the conditions of the school and the purpose of why it's used in the first place.
I think it's more helpful to look at a specific project and talk about what you see (and don't see) in the project itself. For me, it's difficult to look at a project like Beyond the Crossfire (https://www.hightechhigh.org/hthcv/project/beyond-the-crossfire/) and not see examples of students learning rigorous content before applying it. Looking at this project specifically, there is a high real-world application (students are film-makers) but also high levels of scaffolding to support the development of the skills required for them to take on these roles.
This is an example of great PBL to me and one that I (and many others who have worked in this space for decades) have argued is the quality-standard we should PBL to...
I'd love to look at a specific project with you and talk more- about what's there and also honestly about what's not.
Hi Loni,
Thank you for sharing this. I appreciate the way you’re grounding the discussion in a specific project rather than PBL as an abstract ideal.
In the "Beyond the Crossfire" project, I can clearly see strong elements: purposeful application, authentic roles, and implementation of complex skills. I also agree that this is very different from the loosely defined versions of PBL that often circulate under the same label.
One thing I’m genuinely curious about, though (and this is where my interest tends to sit) is assessment. The project description does a good job articulating learning goals across skills, content, and mindsets, but it’s less clear on how those were evaluated:
a. How was students’ conceptual understanding assessed independently of the final product?
b. How did teachers distinguish between strong collaboration or production quality and actual content mastery?
c. What evidence was used to decide whether learning goals were met, beyond successful completion?
I’m asking this not to dismiss the project. But this is usually where discussions of PBL become harder to pin down. I’d enjoy unpacking a project together at that level: what’s explicit, what’s inferred, and what evidence of learning we can confidently point to.
Thanks again for sharing.
Hi Umes,
Thank you for the thoughtful reply. I would really be open to talking any time about a specific project or assessment in general. Please feel free to send me an email at loni@imagineif.dk and we can find a time.
To address your questions about assessment specifically, I would ask: how are these things assessed in non-PBL classrooms?
Most of the time in PBL schools, it's done in pretty much the same way. There are PBL teachers who give quizzes, tests, require documentation on Digital Portfolios, assign essays and have rigorous formative and summative assessment benchmarks by using rubrics and other success criteria. I used to have one-on-one sessions with each student after a project was completed where we looked at their Digital Portfolio, final product and I made sure the academic learning goals were reached. I called them "Teach Back" sessions. Sometimes Presentations of Learning are also used or the Exhibition itself becomes an exam of sorts.
The "unicorn" PBL projects have great products that require rigorous academic learning to do them. In those cases, final products can definitely be an assessment tool in itself. A good "rule of thumb" is to see if students can make the product on the first day of the project. If they can- the product is too fluffy and the likelihood of deep learning is low. A vast majority of PBL projects I see have products that are mismatched to the rigor of learning required of the subject, have low criteria for success or have a high degree of choice for kids which can lead to poor academic integration. My experience is that while kids can be naturally curious, they tend to choose the same content or products when given a choice and the quality/depth of that work rarely increases on it's own.
As I wrote before, the definition of PBL matters here and the context in which it's being done also matters. I work with independent schools that have a lot of freedom in their need for learning outcomes and can take PBL in a very loose and child-led manor. Other schools exist in traditional systems where learning outcomes matter a lot to individual children and the schools themselves- so the flavor of PBL will be different.
Happy to talk more.
-Loni
So would it be fair to say that broadly (except very few contexts) PBL should be used from middle school onwards, and primarily in high school to balance the monotony and pressure of competitive exams?
Hi Arushi,
That’s a reasonable way to think about it, but with an important caveat.
It’s less about age or grade level and more about learner expertise, task complexity, and the kind of learning goal involved. As students gain stable prior knowledge and conceptual fluency, they might benefit more from project-based work.
PBL can play a role in middle and high school especially for application, integration, and transfer but not a substitute for building foundational knowledge, nor as a remedy for exam pressure or classroom monotony. Those are system problems, not instructional ones, don't you think you?
The key question isn’t when PBL starts, but what knowledge students already have and what kind of thinking we’re asking them to do. Hope that makes sense.
I really enjoyed your take on this. I have seen evidence of what you discussed in teachers’ classrooms. It looks like learning but the learning part is not always the center. What can be copied or pasted in? What can I do to make this look pretty? Seem to be in front of the harder parts of learning. One classic example is having students plan a dream trip. You could say they are multiplying decimals, but that isn’t where most of the work is. Most of it is finding the places and prices, often with that assignment students don’t have a budget or understanding of what could be reasonably accomplished in a trip. You could say they are learning geography but often they are just regurgitating what they are finding online. It is important when we plan projects or lessons we see where the work actually is and make sure that matches our learning target.
I have seen students giving amazing presentations at the end of their project, and forget almost everything after a month and even a week.
I wonder what is your evidence to back up these claims. I've spent several years doing research on PBL in science classrooms. It is not a perfect method and it needs to be done by skilled teachers, but most actual studies shows that when it is done right, it supports students' cognitive learning just as good as traditional classroom learning, with the added value of social and emotional engagement, and self-efficacy. Several reviews confirm these claims. You easily dismiss the benefits of PBL as a whole, without any concrete evidence from field studies- I find this populistic and dangerous- though I will be happy to be proven wrong...
Hi Tom, I appreciate this push for evidence. When we look at the 'evidence,' we have to distinguish between Social-Emotional outcomes (self-efficacy, engagement) and Cognitive Architecture (long-term memory encoding).
This post is my conceptual argument majorly based on the Architecture of Complexity (Herbert Simon) and Cognitive Load Theory (Sweller). These are descriptions of the working memory bottleneck of the human mind. Research on CLT and Cog Sci shows that for a novice, a 'noisy' environment like PBL creates a heavy extraneous cognitive load that actively competes with the learning goal.
You mention that when PBL is "done right," it works. But "doing it right" usually involves the teacher providing massive amounts of Direct Instruction and Scaffolding within the project. At that point, is it still the project doing the teaching, or is it the structured instruction hidden inside the project?
My argument isn't that PBL has ZERO value; it’s that it is an Integration Tool, not an Instructional Tool.
Hi Umes, thank you for the thoughtful reply. I agree with all the theories you've mentioned, but it still does not include any actual empirical studies that were performed on PBL in classrooms, and there are plenty of those from the last 30 years, with mostly positive results. Of course, it requires direct instruction in the early stages of adoption from the teacher, like every pedagogical approach. As I said, PBL is not perfect, it is just another pedagogical approach, and a powerful one, that teachers should have in their tool kit. In general, the argument that the most effective teaching approach is the traditional 'pounding' of materials into students' brains is wrong for many reason, despite the current 'populist' wave of bashing against any other approach that had been emprically established in recent years. And above all, the question is always what are your goals and what you are measuring. If you seek to improve students' scores in traditional memorization and basic problem solving, then by all means, you should only teach use teacher-centered traditional excersize-based methods. And again, I appreciate the discussion and open for new evidence-based arguments.
Greetings,
Thank you for writing this op-ed. As a teacher working at a high school that runs on a Project Based Learning model with inter-disciplinary multi-graded studios, I would like to bring some counter points to some of your claims.
-Education is contextual, blanket claiming something to work, or not, most of the time seems far-fetched. Project Based Learning is a tool. How you integrate it, how much of your curriculum is shaped around it, how much agency does the teacher have in co-creating this course, are all important foundational questions to measure if it works for your context or not? Our school's studio curriculum is created around 6 academic pillars:
-Communication and Expression
-Mathematical Thinking
-Social Systems, Civic Engagement & Ethical Thinking
-Technology and Digital Citizenship
-Scientific Inquiry
-Design and Innovation.
For example for this trimester, the studios being offered to students: Local Agents of Change (inspired from Marshall Ganz's Public Narrative), Politics on your Plate (Ceramics class for food combinations), Matter Matters (Material Science for Climate Change).
- Yes, there is tendency of Project Based Learning to have a higher weigh towards short term excitement. And that is why we try to use student created reflections and documentation as the most important part of any project, in order to bring some balance of long term mastery. In fact, we define mastery for our students as when they are ready to become teachers to other students, and have a dedicated block of time for peer learning groups for students to choose what they want to learn or teach.
-And in no way I am claiming that any of this works perfectly. We run into our challenges daily, as all schools do. And that's what we do as educators, come together and trying to figure out this inter generational information transfer task at a pivotal point in their brain development.
-To your sub-assembly point, there might be truth to that. We may START with building with shadows and be playing with the breadth a bit, which I also feel is the point of high school, to be interested in how the real world interacts. But once a student is hooked, I can be the worst teacher and the student still wants to solve the problem so they will extract the information out of me. And the certain depth required for technical and soft skills, we make sure students have enough required and elective studios to attend from which will allow them to find their interests from the breadth and grow deeper in them.
Scratch is one of the most beautiful learning platforms made to learn programming. Why does it work at that scale? Low barrier to entry, but the ceiling is limitless. A month ago I demonstrated a speech therapy demo to a early childhood school on Scratch, which took me 15 mins to build.
-And again, none of this is being prescribed as 100% works for all. For some students it can be truly life changing, and for some it can be a very tough time. Does that mean you abandon the tool itself? Or understand it as a tool and hope to use it for its merits and avoid its pitfalls.
{Typed on phone. Apologies for typos}
Hi Ayush Gandhi,
Thank you for sharing this detailed perspective from the field. It sounds like your Studios are vibrant environments, and I appreciate your honesty about the daily challenges.
I want to zoom in on your point about 'building with shadows' to get students hooked. From a Herbert Simon perspective, the concern isn't about the hook; it’s about the cost of the detour. If we start with breadth and shadows, we are essentially placing a novice on a 'noisy environment' and hoping their interest will be strong enough to help them navigate the clutter.
For a highly motivated and skilled student, this usually works. But for the student who hasn't mastered the 'stable sub-assemblies' (the math, the syntax, the logic), the shadows never become solid.
Regarding Scratch, I think it actually supports the 'Artificial Learning' argument! Scratch is successful precisely because it decomposes programming. It removes the 'noise' of syntax errors so students can focus on the 'sub-assembly' of logic. It is a beautifully designed, simplified artificial environment. Not a messy, 'real-world' one.
My argument isn't that we should 'abandon PBL as a tool', but that we must stop using PBL to teach the foundations. Projects are for the Integration phase. If we use them as the Instruction phase, we are simply asking novices to act like experts. And cognitive science is clear on this: novices can't think like experts.
I agree with the concern about the illusion of learning, but I disagree that project based learning is the root cause. In my Systems Analysis and Project Management capstone, projects were not used in place of instruction. They were used to operationalize explicitly taught skills. Students were taught scope definition, requirements analysis, dependency mapping, risk identification, governance, and decision rights before engaging in the project.
The capstone was intentionally work relevant to the point that students’ employers flat out encouraged them to prioritize the course because of its immediate applicability. Students also reported continued use of these skills on the job well after the course ended. In this context, the project functioned as a structured application environment with feedback and accountability, not a discovery exercise. When project based learning fails, it is usually a systems design failure involving missing scaffolding, weak assessment, or unclear objectives, rather than an inherent flaw in learning through projects.
Hi Dr. Bourne,
Thank you for sharing this. I think this is a brilliant example of what I would call Project-As-Integration rather than Project-As-Discovery.
You’ve actually perfectly illustrated Herbert Simon’s point about 'Near Decomposability.' By explicitly teaching (scope definition, risk identification, and dependency mapping) before the project, the students had opportunities to develop stable sub-skills.
The reason your capstone succeeded (and where most primary school PBL fails) is the Expert-Novice Fallacy. Your students were likely at a stage where those sub-skills were becoming automatic, freeing up their working memory to handle the 'noise' of a real-world project.
In many schools, we do the opposite: we throw novices into the 'noise' before they have the gears. We ask them to 'discover' the laws of physics while building the bridge.
The Systems Design approach you mentioned is exactly the kind of Artificial Learning Simon praised. Something that is highly structured, explicitly taught, and intentionally scaffolded. The 'illusion of learning' only happens when we skip the instruction and scaffolding and hope the project does the teaching for us.