Acitivity 1

Preparation

(I was in the reaction group for Schmidt et al., 2007)

1. The paper’s main claim(s) minimal guidance is less effective, but PBL is not always minimal guidance, it is a method in which the amount of guidance is flexible. PBL is in fact compatible with human cognitive architecture, and maybe even more so than direct instruction.

2. The evidence that supports the claim(s)

discussion activates prior knowledge:

• Schmidt et al. (1989), novices recalled 40% more after discussing a problem, experts 11% more (hints at expertise-reversal effect)
• Schmidt (1993), health science students recalled almost 2x as much after discussing a problem

support/scaffolding is needed and available:

• Schmidt (1994), the expertise of the tutor is more important if prior knowledge is low, or instruction unstructured (consistent with CLT)

PBL leads to better application:

• Dochy et al. (2003) meta-analysis

1. How it responds to the arguments by Kirschner, Sweller and Clark

• it shows that KSC misunderstand how PBL works: PBL has the goal “scaffolding for student independence”. KSC confuse this goal (for experts) with the method used (for novices).
• it shows that PBL takes into account CLT, because PBL elements exist to manage cognitive load:
  • seven-step training + restricted resources -> extreanous load
  • simple-to-complex order + tutor assistence + distributing load among group -> intrinsic load
  • discussions + tutor prompts -> germane load
• it addresses the argumentation & evidence used
  • complexity mentioned by KSC is CLT-complexity, not PBL-complexity, so the comparison is not entirely fair
  • evidence mentioned by KSC is for individual learning, while PBL is about group-based learning, so the comparison is not entirely fair
  • curriculum comparisons are not randomized + may show ceiling effects
  • pbl optimizes for different transfer (not solving specific new problems per se, but how to solve problems in general; you test direct instruction by giving new problems to solve, you test pbl by testing how well you learn next time), so again, comparison might not be entirelt fair

1. What does it challenge or leave unresolved?

• it challenges the categorization, not the main claim, because they agree with the claim, but just think it does not apply to PBL
• the argument about implementation is not strong, because PBL can be badly implemented, and badly implemented PBL is minimal guidance, and KSC critisize the PBL they see in practise
• authentic problems can have high intrinsic load, even for the “simple” problems (especially because “simple” problems are PBL-simple, not necessarily CLT-simple!!)
• the claim “works better than direct instruction” is not entirely supported, and there is no experimental evidence that compares PBL and worked examples (because the study discussed compares problem discussion to discussion of unrelated problem, not to worked example)
• the learning still happens individually, from the provided resources, which is, if you think about it, actually just direct instruction. so they essentially argue for direct instruction wrapped in authentic problems. i wonder what the effectiveness of PBL is on its own, if you leave out the direct instruction part.

Presenting

1. Share your paper’s main claim, its strongest support, and how it responds to Kirschner, Sweller and Clark

Our paper’s main claim is that problem-based learning is not (always) minimally guided, but that it is a form of instruction that is flexible in the amount of guidance, and that it is in fact compatible with human cognitive architecture, and maybe even more so than direct instruction is. It agrees with KSC that minimal guidance is an ineffective instructional method, but disagrees with the categorisation of PBL as minimally guided. It shows that PBL takes cognitive load into account with various elements. It uses 7 clearly structured steps and a restricted list of resources (so the learner doesn’t have to search), which both reduce extraneous load. It orders problems from simple to complex, the cognitive load is distributed over the group, and there is a tutor to assist in case a group gets stuck. These all limit intrinsic load to a managable level (because it is true that the authentic problems used in PBL problems have a baseline higher instrinsic load than similar direct instruction approaches would have). And the group discussion and tutor prompting can both contribute to a higher germane load, and thus more learning. The empirical evidence shows that the group discussions in PBL lead to activation of prior knowledge, which for novices leads to a 40% higher recall rate, and for experts 11% higher, which hints at the expertise-reversal effect (but both are still higher!).

Feedback/critique

(These questions the other group should answer, but because I am at home, I’ll do the critique myself, and then exchange with a classmate later.)

1. Does the paper support, challenge, or put limits on KSC’s claim?

It limits the scope of the claim (to exclude PBL as minimal guidance), but it does not challenge the argument.

1. What specific point in the paper supports that position? (e.g. a study result, an example, or something else)

The specific points would be the PBL elements and their effect on cognitive load, like the tutor giving just-in-time information or distributing cognitive load among a group.

1. Does this argument convince you?

I agree that (seeing the arguments provided by Schmidt et al.) PBL is not minimal guidance. So it was wrong of KSC to group PBL into that category. I however, do not entirely agree with that PBL could be better aligned with cognitive architecture, because I essentially see it as window dressing (wow a term i learned during OVL!) on direct instruction. I think you’d get the same results if you start with any other preinstructional strategy that activates prior knowledge.

Group discusion

(I’ve skipped the plenary discussion, because I didn’t see the presentations of other groups..)

Conclusion

“Constructivist learning environments that offer minimal guidance and ignore human cognitive architecture are based on ideology and are ineffective”

1. What’s your personal position?

I think the statement on its own would be true, but I don’t agree with the way the statement is formulated or how it portrays constructivism.

I think the above statement is maybe formulated a little strongly. A few thoughts:

• If something ignores human cognitive architecture, it is almost by definition ineffective, right? So the statement is almost the same as “snow is white”. (I forgot the fancy term we use during exams for this type of statement… but it’s bad reasoning nonetheless)

• The claim that constructivists learning environments like this exist at all is not supported at all..? Like, I feel like they paint constructivism badly and in the extreme. If it’s really that bad, you woulnd’t need to formulate it so strongly.

• Who says its based on ideology? It can also be based on outdated information? Either way, it’s a claim about motivation, not about effectiveness. Something can be both ideological and effective, in which case it wouldn’t be all that bad.

1. How do the different articles nuance your interpretation?

I have only read one of the articles in its entirety, which essentially agreed with the core claim of KSC. From the abstracts of the other articles, I gather that: Hmelo-Silver et al. make the same argument, but they want to exclude inquiry learning rather than PBL. And Kuhn redirects the entire debate onto what the goals of education should be, which is IMO frankly irrelevant. So regarding the core claim, not much. However, having read this one article, I do wonder whether the other ideas about learning named by KSC (discovery learning etc.) that are lumped into minimal guidance, are not miscategorized also. I feel like KSC oversimplify and mispresent these instructional methods.

So generally, I read the article by KSC more critically now.

1. Did your position change because of the discussion?

I am still pretty much on the side of KSC, which is also primarily because I as a learner prefer direct instruction. Which may have something to do with that illusion of learning, now that I think about it…

Academic reading

“Disputes about the impact of instructional guidance during teaching have been ongoing for at least the past half-century…”

1. Who says so? Is it true? The authors of the paper claim this. You can only judge whether this is true or not when you see the evidence they would provide for this claim, and the rest of the sentence is cutt-off, so we don’t know if they cite a source or provide some kind of explanation or something. So it may or may not be true. I’m inclined to just accept it as true and continue reading. I can always revisit it later.

2. What’s a ‘dispute’? An argument, discussion, fight? A lack of consensus, I suppose.

“On one side of this argument are those advocating the hypothesis that people learn best in an unguided or minimally guided environment…”

1. Is it really one side? Framed black-white? Again, depends on the evidence provided.

2. Are ‘unguided’ and ‘minimally guided’ the same? Depends on your definition of both. And you could argue that they may be very similar in practise.

“…generally defined as one in which learners, rather than being presented with essential information, must discover or construct essential information for themselves…”

1. Who defined it this way? The authors.

2. This definition is debatable: many constructivists argue for guided discovery or scaffolded inquiry, not pure self-discovery. Dated argument/definition, or…? Yeah, so if constructivist argue for GUIDED discovery, it’s not minimal guidance, or unguided, is it? The definition of minimally guided or unguided instruction doesn’t apply to them at all. At most the categorisation of constructism as minimally guided or unguided is wrong, not the definition itself. Or maybe I don’t quite get this comment?

“Learning, in turn, is defined as a change in long-term memory.”

Activity 2

Group discussion

1. Where are students making sense of ideas?

• Maths: initially during the explaining, and probably again while solving the second problem.
• History: during the comparison.
• Science: supposed to happen while physically positioning and moving the objects, and maybe triggering cognitive dissonance in the progress. in this case it doesn’t, because groups get stuck or use materials without knowing what they’re doing.

1. What guidance do they get?

• Maths: a worked example
• History: two cases and evidence cards
• Science: only physical materials and a prompt

Preparation

1. What is missing from students’ current understanding?

• the earth is tilted
• the angle determines the amount of energy
• days are longer in summer than in winter
• hemispheres have opposite seasons

…probably these

the main thing they need to know to see that distance is unrelated is:

• earth is closest to the sun in january, and then its winter for us, so that would be opposite with distance
• if its summer for us, its winter in australia, which also cannot be explained by distance alone

1. What should they be able to explain by the end of the lesson?

seasons are caused by the tilt of the earth, which changes the angle of light and the length of the day throughout the year.

Designing

1. Check (mis)conceptions of seasons by asking something like “why is it warmer in summer?”, students all write it down and then a few explain.
2. Let them investigate, but have some scaffolding as to how, like:
   • shine flashlight on paper at 2 angles, compare how big the spot is
   • have the globe with the tilt, rotate it around lamp, and mark europe and australia (make sure tilt direction stays fixed!!)
3. Teacher can prompt for something like “if distance causes summer, what season should it be in australia in july?” and “where is the most light on the globe?”
4. Students revise their initial explanation, and maybe also explain it to a peer or something.

Checking design

1. Use your assigned perspective to review your lesson. Discuss what your perspective reveals. What change would improve the lesson?

• Motivation: Does the lesson support students’ competence, autonomy, or sense of connection?

1. Competence: in the original task, students got stuck => failure experience => lower self-efficacy. with cleared structure, they would have higher self-efficacy & competence, so more motivation.
2. Autonomy: structured steps do reduce autonomy, so from this viewpoint motivation would be less.
3. Relatedness: working in groups satisfies relatedness. Maybe choosing the groups themselves would be benificial too?

• Metacognition: How will students notice whether they understand?

in the original lesson, students had no indication whether they understood or not, and maybe using materials felt like false productivity / illusion of learning. there is no pre- or post-test.

writing down an explanation/prediction at the start and changing at the end if its is wrong can show that they now understand.

• Self-regulation: How will students plan and manage their work?

structured steps are given, so planning is less of a problem, I think, because the teacher regulates. you can solve that by giving them the steps and making them determine the order themselves. but the goal of the lesson isn’t really self-regulation, is it? also, maybe some kind of support in telling them how to work together would be in order too.

• Cognitive architecture: Where might students get stuck or overloaded? What support could help?

Orbit, rotation, tilt, angle, energy and day length are all interacting elements, so there might be a high intrinsic load for beginners.

understanding separate elements in separate steps first, can maybe reduce intrinsic load a little.

and marking things physically on the globe can also reduce/offload the cognitive load partly.

So the problem is basically these perspectives “pulling” towards two sides:

• from CLT perspective, we need more structure
• from motivational perspective (and also a little self-regulation), we need more freedom/autonomy

the “solution” is probably the ordering/sequencing: first more structure and later on more autonomy (which is basically what schmidt et al. called “scaffolding for independence” btw.)

Revising design

• What will students work out or revise?

Work out:

• low-angle light spreads over bigger surface area
• tilt => hemispheres have different angle => tilt causes seasons

Revise: initial distance explanation

• What guidance will the teacher provide?

  • structured steps, and the materials
  • the prompts
  • short demo of the orbit probably, to reduce cognitive load

• What changes did you make based on looking at your design from another perspective?

  • Cognitive architecture: split the hands-on part into structured steps + offload information physically onto the globe + a demo before doing it themselves (modeling!!)

  • Megacognition: a ‘pretest’ and ‘posttest’ in the form of writing down a prediction/explanation and revising it at the end.

  • Self-regulation: i am really in doubt about this one. maybe offer the steps and make students order then? or maybe give some kind of instruction on team-work, with roles..?

  • Motivation: situational interest (trigger curiosity), higher self-efficacy /competence due to increased structure

Based on your designs, does constructivism mean minimal guidance, or can teacher guidance support students’ active thinking?

I think teacher support is crucial if you want to maintain competence/self-efficacy, and I also think that elaboration (‘active thinking’) is a separate process that can be triggered independently of whether information is given or discovered. IMO, the main goal of self-discovery during learning is to increase motivation through autonomy, but I think this is unfit for beginners. And I also think constructivism doesn’t equal learning through self-discovery/minimal guidance only. Direct instruction can still cause self-construction of knowledge, so IMO constructivism and self-discovery/minimal guidance are confounded here.

Final activity

“To really learn in this course, make sure you…”

read the literature ahead of the tutorial session, and really prepare it well. you will get 1000% more from the tutorial session that way

^the two times i prepared less thouroughly or read literature after the fact are the weeks i have the worst understanding off :(

and attendance! its so important! which is ironic because i am writing this in a compensatory assignment i have to make because i was absent..