Monday, November 12, 2007

Kirschner, Sweller, Clark (2006) - Readings

Note: I'm posting this as an example of the reading I have done while reviewing an academic paper. These are my notes on Kirschner, Sweller and Clark's paper Why Minimal Guidance During Instruction Does Not Work: An Analysis of the Failure of Constructivist,
Discovery, Problem-Based, Experiential, and Inquiry-Based Teaching.

Interestingly, as I review these remarks, not one author has landed on the fundamental problems with the paper. More later.

Primary reference:

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, 75-86.

Mark Montgomery
:Textbook Evaluator

Despite the article’s great strengths, it sometimes devolves into an empassioned plea for more direct instruction. As the authors state at the end of the introduction, the article is not a dispassionate presentation of the evidence. Rather its goal is to discredit constructivism.

The goal of this article is to suggest that based on our current knowledge of human cognitive architecure, minimially guided instruction is likely to be ineffective. The past half-century of empirical research on this issue has provided overwhelming and unambiguous evidence that minimal guidance during instruction is significantly less effective and efficient than guidance specifically designed to support the cognitive processing necessary for learning.


Note how the authors set up a dichotomy: “minimally guided” vs. “guided.” As in most education research, the nuances of reality are just too complicated to address. There are, in practice, various shades of “guided” instruction. On one extreme is the “all drill all the time” school of rote memorization–or 100% guidance. On the other hand is the Rousseavian “let the kid just explore everything all the time, and I’ll just provide a safe space for those explorations” (or 0% guidance). I understand that the authors of this article need to set up categories in order to analyze instructional methods. But be real, folks: the way these things are put into practice by actual teachers with actual students is much less extremist than the academics, pundits, policy makers, and bloggers would have us believe. Most educators I have ever met, dealt with, taught beside, or watched instruct fall somewhere in the glorious, gray middle.

Miles Berry

The authors cite research in which constructivist approaches have produced lower test scores than approaches using direct instruction. Whilst this isn't news, and the 'no significant difference' phenomemon is well documented elsewhere, it does rather come back to a question of what's being measured - if the measure of success is recall of facts and their application to relatively familiar problems, then it's not entirely surprising that pedagogic approaches which emphasise this dimension will produce better results. The experience of teaching to the test is far from uncommon, and high input methods such as detailed revision notes, worked examples and practice papers are certainly felt to be effective by many, if effectiveness is determined by performance on narrow spectrum tests, as indeed it is in UK and US education.


One of the main problems with the authors' argument is, to my mind at least, their failure to acknowledge much by way of a social dimension to learning - their comparison seems largely to be between content delivery led by a teacher and alternative approaches in which learners work things out for themselves.

Of course, traditionally this third way was limited at school level, where the learning network was restricted to those in the class, most of whom would be at roughly the same level, their teacher, the textbook and maybe the school library, but with net access from our classroom and desktops, the number and variety of others with whom learners may engage is orders of magnitude greater. Not only does this extended learning network provide more sources from which students may learn, it also open up a whole host of opportunities for them to teach - as they explain new concepts to each other, and work together to solve problems, I'm sure that the quality of their learning is enhanced, as is the likelihood that such concepts can make it into longer term memory.

Annette Kujawski Taylor, Ph.D.
Professor of Psychology
University of San Diego

I think there is also an issue of each of these methods of instruction in terms of different disciplines. In psychological science "direct inquiry" can lead people hopelessly astray, especially if they begin with incorrect constructions on which to base their future decision making.

As we all know, misconceptions about psychological science abound.

Richard Hake
Cognitive Science and Physics Education Research:
What We’ve Got Here Is Failure to Communicate

It appears that the work of physics education researchers (PER’s) is virtually unknown to some
cognitive scientists CS’s. Several years ago I stumbled upon Allan Collins’ (1999) valuable
article “The Changing Infrastructure of Education Research.” Collins wrote: “Recently
researchers have begun to study teaching and learning in the context of real-world learning
environments,” citing Brown (1992) and Collins (1992). This puzzled me since PER’s have
been studying “teaching and learning in the context of real-world learning environments” for
about three decades. In Section IV, “Empirical Studies,” of McDermott & Redish’s (1999)
“Resource letter on physics education research,” I count over 80 articles, dating from McKinnon
(1971), that feature classroom research. However, it must be admitted that relatively few of
those articles meet the criteria for “design based research” (DBR) suggested by Brown (1992),
Collins (1992), and Kelly (2003).


Another example of what I would regard as a communication failure is provided by the
previously mentioned paper of Kirschner, Sweller, & Clark (2006) with its seemingly non-
sequitur title “Why Minimal Guidance During Instruction Does Not Work: An Analysis of the
Failure of Constructivist, Discovery, Problem-Based, Experiential, and Inquiry-Based
Teaching,” even despite PER evidence reviewed by Hake (2002; 2005b; 2007a,b; in press) for
the effectiveness of all but extreme "discovery teaching.” Kirschner et al. wrote:
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

But here again, as with Klahr and Nigam (2004), “direct instruction” appears to mean to
Kirschner et al. (2006) pedagogy rather similar in some respects to the “interactive engagement” methods shown to be relatively effective by physics education researchers, as can be seen from the Kirschner et al. abstract [my insert at “. . . .[insert]. . . .”]:
Evidence for the superiority of guided instruction. . . [read “interactive engagement”?]. . . is
explained in the context of our knowledge of human cognitive architecture, expert–novice
differences, and cognitive load. Although unguided or minimally guided instructional
approaches are very popular and intuitively appealing, the point is made that these
approaches ignore both the structures that constitute human cognitive architecture and
evidence from empirical studies over the past half-century that consistently indicate that
minimally guided instruction is less effective and less efficient than instructional approaches
that place a strong emphasis on guidance of the student learning process. The advantage of
guidance begins to recede only when learners have sufficiently high prior knowledge to
provide “internal” guidance. Recent developments in instructional research and instructional
design models that support guidance during instruction are briefly described.

Sigmund Tobias, Institute for Urban and Minority Education, Teachers College, Columbia University

I have been impressed by one aspect of the constructivist (let me use that term as shorthand for discovery, problem-based, experiential, and inquiry based teaching that are the subject of the Kirschner et al., paper) position not commented on in the article. It has been claimed, most recently at the Games conference described above, and it seems intuitively reasonable that these approaches induce higher student motivation than approaches relying more heavily on deduction. On the other hand, I am unaware of a study comparing motivation on constructivist and other types of materials.

Traci Sitzmann & Katherine Hildebrand, Training Evaluation Team, Advanced Distributed Learning Co-Lab, Alexandria, VA

Working memory is limited in duration and capacity when processing novel information. Minimally guided instruction burdens trainees’ working memory by requiring them to sort through irrelevant information while locating information that is relevant to their jobs. Working memory cannot be used to commit relevant information to long-term memory while it assesses the relevance of material. In contrast, direct-guided instruction provides trainees with material that is directly applicable to their jobs so working memory can be used to commit training content to their long-term memories. -

Barak Rosenshine, University of Illinois-Urbana

Despite years of research on instruction, and despite research findings such as those which Kirschner et al. (2006) present in their article, constructivism remains the dominant position of curriculum departments in colleges of education. The curriculum departments in colleges of education favor constructivism, and they are able to withstand those empirically oriented educational psychologists who tend to favor direct guided instruction.
The curriculum people always win. Constructivism is a lovely romantic idea of students pursuing their interests and learning on their own, and there is no force of data that can withstand those emotional ideas. Constructivism, after all, is not data-based; rather, it is an emotion about kids and teaching. Chall (2000) called such notions “romantic,” as opposed to “rational,” and apparently the absence of data to support romantic beliefs does not reduce their attractiveness to adherents. I’m surprised that after all these years people like Kirschner et al. and Richard Mayer (2004) don’t understand that the curriculum people always win on this point.

Unnamed student of Paul Kirschner
A cognitive-load approach to collaborative learning: task and learner characteristics

While all levels of education are making use of collaborative1 learning techniques in
both traditional and electronic learning environments, the effectiveness of these types
of education/learning has still not been proven. The results are mixed (at best) and
educators often have to implement extra measures to either ensure that the
participants work together (e.g., requiring a specific number of contributions in
electronic environments or requiring attendance in face-to-face environments) or
ensure that all learners engage in the learning process (e.g., implementing roles,
scripts, and assessment schemes). The basic assumption of this project is that if
individual learners are to work together effectively in groups, the architecture of their
cognitive system and the characteristics of the task must be understood,
accommodated, and aligned. Specifically, this means that the characteristics of the
(group) task must be such that the cognitive system of each of the individuals is not
capable of accommodating its solution AND the group communication and
coordination activities necessary for effectively functioning as a team (i.e., the
cognitive transaction costs) do not impede the collaboration process. This project
uses a cognitive load theory-based approach to develop a method to determine a
groups cognitive load and to investigate how task complexity (i.e., intrinsic cognitive
load) and the learner characteristics expertise and age can inform the design of
effective group-based environments for lifelong learning.

Doug Holton


Problem-based learning (see this overview) in particular is anything BUT "minimally guided." See for example this article by Cindy Hmelo-Silver and Howard Barrows (father of PBL) which details the role of the facilitator in PBL. Kirshner, Sweller, and Clark appear to have created a straw man argument that no one actually believes. Why bother responding to something like that? For one thing, it has been published in a peer-reviewed research journal, and thus can and will be interpreted as scientific "truth." This has happened before. As Hake discusses in his listserv post, an earlier 2004 article by Klahr and Nigam entitled "The Equivalence of Learning Paths in Early Science Instruction" has been used by conservatives to argue that good old fashioned lecture is the best way to teach. So we have to respond to these types of articles and engage in a dialog and not merely dismiss one perspective or the other.

D-Ed Reckoning

Summary from D-Ed Reckoning - Part 1

[ Good Diagram ]
Experts are skillful in an area (domain) because their LTM contains a huge knowledge base of information concerning the area (domain). This knowledge base permits experts to quickly recognize the characteristics of a given situation and serves as a basis for them to determine what to do and when to do it. For example, expert problem solvers derive their skill by drawing on the extensive experience stored in their LTM and then quickly selecting and applying the best procedures for solving problems...


Because WM presents severe limits on the amount of information that can be held in mind simultaneously and on the duration for which it lasts once attention is withdrawn from it, WM is often described as the bottleneck of the human information processing system.

Summary from D-Ed Reckoning - Part 2

Inflexible knowledge is the unavoidable foundation of expertise including that part of expertise that enables students to solve problems by applying existing knowledge to new situations. We call this expertise "problem solving skills." As students work with their knowledge, their store of knowledge in LTM becomes large and increasingly flexible. In other words, the mind's concrete form bias is overcome by the accumulation of a greater store of related knowledge, facts, and examples. The end product of flexible knowledge is expertise which is deep structure of a large domain of knowledge. It is this expertise that we associate with higher-order or critical thinking skills and it can't be taught directly.

Summary from D-Ed Reckoning - Part 3

First the bad news. Practice does not make perfect. If you practice or rehearse to perfection, you will be perfect today. You will likely not be perfect a few days later. It is not good enough to learn something to mastery, you have to overlearn it, past the point of mastery, if you want to retain it in LTM. With sufficient practice the skill or knowledge will become automatic.


No one has yet found a short cut or golden road around the need for sustained practice for learning.

To educators, practice is disparaged as "kill and drill." They don't like it, so they want to avoid it. They are looking for a shortcut around the need for distributed practice to get students to mastery. Fundamentally, they are looking for magic. This is an ongoing theme in modern education. Look at nearly every education reform in recent time and you'll see a common characteristic--the desire to reduce the need for practice.

Summary from D-Ed Reckoning - Part 4

Knowledge helps students in three critical ways:

1. it makes it easier to learn new information,
2. it makes it easier to remember new information, and
3. it improves their ability to think by circumventing the thinking process and by freeing up space in working memory.

Summary from D-Ed Reckoning - Part 5

The theory underlying constructivism is that people learn best in an unguided or minimimally guided environment , generally defined as one in which students, rather than being persented with essential information, must discover or construct essential information for themeselves. Thus, constructivism requires the novice student to search a problem space for problem-relevant information to discover the essential information, i.e., the solution to the problem posed.


Bear in mind, the goal of instruction is rarely just to search for or discover information. The goal of education is to give learners specific guidance about how to cognitively manipulate information in ways that are consistent with a learning goal, and store the result in LTM.

Allegedly, constructivism helps students to derive meaning from learning materials. However, cognitive load theory suggests that the free exploration of a highly complex environment generates a heavy WM load that is detrimental to learning. This suggestion is particularly important in the case of novice learners who lack proper cogntive LTM structures to integrate the new information with their prior knowledge.

while there is little doubt that students remember material they generate themselves better than material that is handed to them. This "generation effect," as it is called, is indeed powerful, and it is due, in part, to forcing the learner to think about the meaning of material (although other techniques can do that as well). Part of the effect does seem to be unique to the actual generation of the answer, over and above thinking about meaning. For this reason constructivists believe that discovery learning should be employed whenever possible. However, given that memory follows thought, one thing is clear: Students will remember incorrect "discoveries" just as well as correct ones.

I find it interesting that the 'summary' in D-Ed Reckoning bears little resemblance to the actual article.

Public Testimony
To the National Math Panel
Dr. Martha Schwartz
November 6, 2006

long-term memory can be considered as the central and dominant structure of human cognition. Long-term memory refers to the enormous pool of knowledge stored over the years and accessible as required. For example, knowing how to read, write, swim, play chess, walk and all the other tasks that we have learned to perform are stored in long-term memory.


However, the current dominant tendency in education is to promote discovery learning.
Kirschner, Sweller and Clark state the following: “Yet many educators, educational researchers, instructional designers, and learning materials developers appear to have embraced minimally guided instruction and tried to implement it” (2006, p. 6). What’s more, much of the literature on ICT endorse this approach as if the capabilities of the tool, the computer, exert some magical power over the learning process.

According to Clark and Feldon (Mayer, 2005, p. 108): “There is a persistent belief among some segments of the education and training communities that the most effective learning experiences are those in which learners navigate unstructured multimedia learning environments or solve novel problems presented without instructional support.”

This should come as no surprise since research on ICT has often been criticized for its lack of rigor and objectivity. Joy and Garcia share this opinion: “Practitioners and consumers of
asynchronous learning networks (ALNs) need to be aware that much of the research in this field is seriously flawed, rendering many of the conclusions inaccurate or open to debate” (2000).

The Integration of ICT in Teaching Practices in Francophone Minority
Settings: Tendencies and Challenges
Clermont Gauthier

Without going into details, it is interesting to review some of the principles enunciated by
researchers and which point to practical directions for teaching.

Multimedia principle. People learn better from words and pictures than from words alone.
Split-attention principle. People learn better when words and pictures are integrated and
not separated.

Modality principle. People learn better from graphics and narration than from graphics
and printed text.

Segmenting principle. People learn better when the message is presented in segments
rather than as a continuous unit.

Coherence principle. People learn better when extraneous material is excluded from the

Personalization principle. People learn better when the words used are in conversational
style rather than formal style and when the words are spoken by a human voice rather
than a machine voice.

Guided-discovery principle. People learn better in discovery-based environments when
guidance is well incorporated.

Worked-out example principle. People learn better when they are presented with
worked-out examples.

Self-explanation principle. People learn better when they are encouraged to generate
explanations during learning.

Animation and interactivity principle. People do not necessarily learn better from
animation than from static diagrams.

Navigation principle. People learn better in hypertext environments when navigation aids
are provided.

Site map principle. People learn better in an online environment when the interface
includes a map showing where the learner is in the lesson.

Fernette and Brock Eide, Eide Neurolearning Blog

When medical students are instructed with case-based instruction - they have superior clinical practice skills, but inferior basic science test performance. So what's the matter with this? Seeback and Kirschner deemphasize this result by mentioning, "But the negatives include lower scores on basic science tests, more study time and a pattern of ordering significantly more unnecessary tests at a much higher cost per patient with less benefit." But who would you rather have for a doctor - one who practiced medicine better or one who knew more answers on a pencil-and-paper test.

Alice Kolb & David Kolb
Experiential Learning Theory Bibliography

Doug Belshaw

As Kathy says, ‘he was explaining that for these musicians, playing this amazing music appeared that easy. Relaxed. Not frantic. There was no desperate need to fill all the space.‘ All well and good, but how is this relevant to educators?

Newbie teachers/trainers often make the same mistake. We fill in every available space, just like those first-time desktop publishers who “abhor a vacuum” and cram words and clip-art into every square millimeter of a flyer.

But real learning takes place between exposures to content! Long-term memory from learning happens after the training. The space between the lessons and practice is where the learning is made permanent. If we don’t leave that space, new content keeps rushing in to overwrite the previous content, before the learner’s brain has a chance to pause, reflect, and synthesize the proteins needed for long-term memory storage.

Kuhlthau, C. & Todd, R. (2007) Guided Inquiry. Retrieved June 14, 2007 from

Kuhlthau and Todd’s research into effective inquiry learning indicates that it must be guided and that the key is in the interventions provided by the teachers and library team. It can no longer be hands off.

“Guided Inquiry is carefully planned, closely supervised targeted intervention …. ……. (by)
teachers to guide students through curriculum based inquiry units that build deep knowledge and deep understanding of a curriculum topic, and gradually lead towards independent learning. ”

Brian Waddell
Teacher Librarian
Karori West & Kelburn Normal Schools

If you can’t beat them, join them: Integrating ICT and guided inquiry learning, a practical solution.

Cognitive load theory at UNSW

Cognitive load theory (CLT) is an instructional theory derived from our knowledge of the evolutionary bases of human cognitive architecture and the instructional consequences that flow from that architecture. A key aspect of the theory is the relation between long-term memory and working memory, and how instructional materials interact with this cognitive system.

Work on CLT was initiated in the early 1980’s at the School of Education, UNSW, and has generated a large range of instructional effects that can be used by teachers, instructors and researchers. We work on projects designed to facilitate learning in curriculum areas such art education, engineering, mathematics, music, reading, science, second language acquisition, and writing, as well as looking at the role of technology and multimedia in learning. Ten Doctoral students working in these areas and using a CLT framework have graduated since 2001.

In recent years, many groups of international researchers located primarily in Europe and the United States have adopted CLT as a theoretical paradigm. We collaborate with a large number of these groups in publications and conference presentations related to CLT. One of the consequences of this collaboration has been the publication of many special issues of leading international journals devoted to CLT, which are listed below.
Dr. H. Leemkuil

When these activities are successfully performed players will gain implicit as well as explicit knowledge during game play. Implicit knowledge is contextualised and difficult to verbalise and therefore difficult to transfer to others or other contexts. Explicit knowledge can be verbalised and can be transferred more easily. However, when some of these activities are not performed or only partially completed, this might influence game play and/or learning from the game.
What's Wrong With Our Schools?

The paper “Applications and

Misapplications of Cognitive Psychology to Mathematics
Education” should be required reading for educators and
math consultants. It is written by three of the most distin-
guished cognitive scientists in the world, John Anderson,
Lynn Reder and Herb Simon, and they argue that educa-
tional movements like constructivism have “questionable
psychological foundations.” Drawing on their own work, and
on numerous studies in cognition, the authors challenge
the central claims of the constructivist school, for example:
that training by abstraction is of little use, that real learning
occurs in authentic situations, that construction needs to
be done in a highly social environment, that knowledge
cannot be instructed by a teacher but rather it can only be
constructed by the learner, that knowledge cannot be rep-
resented symbolically and that knowledge can only be
communicated in complex learning situations.


In The Neuropsychology of Mathematics: Diagnosis and
InterventionSteven Feifer and Philip De Fina argue that
“teaching students proper decision making skills and algo-
rithmic procedures can profoundly influence mathematics


In “The Expert Mind,”
which I discussed in this book, Philip Ross argues that
studies in cognition show that expert abilities can be fostered
in children through practice and rigorous instruction.

In “Mindful of Symbols” cognitive scientist Judy DeLoache
argues that “less may be more when it comes to educational
books for young children.” DeLoache found that the more clut-
tered and distracting the pages of a reader are, the less children
learn from the book. DeLoache also tells a success story:
Using blocks designed to help teach math to young
children, we taught six- and seven-year-olds to do
subtraction problems that require borrowing (a
form of problem that often give young children diffi-
culty). We taught a comparison group to do the same
but using pencil and paper. Both groups learned to
solve the problems equally well—but the group using
the block took three times as long to do so. A girl who
used the blocks offered us some advice after the
study: “Have you ever thought of teaching kids to do
these with pencil and paper? It’s a lot easier.”


Respected education analyst Diane Ravitch succinctly defined them in an essay (May 12, 2005) in the Wall Street Journal:

On one side, beloved by schools of education, are the century-old ideas of progressive education, now called "constructivism." Associated with this philosophy are such approaches as whole language, fuzzy math, and invented spelling, as well as a disdain for phonics and grammar, an insistence that there are no right answers (just different ways to solve problems), and an emphasis on students' self-esteem. ... By diminishing the authority of the teacher, constructivist methods often create discipline problems.

On the other side are those who believe ...

* that learning depends on both highly skilled teachers and student effort;
* that students need self-discipline more than self-esteem;
* that accuracy is important;
* that in many cases there truly are right answers and wrong answers (the Civil War was not caused by Reconstruction); and
* that instructional methods should be chosen because they are effective, not because they fit one's philosophical values.

Convenient chart:

The Ongoing "Education Wars" - Commonwealth Education Organization [Pennsylvania]. Excerpt:
"Most parents have little awareness that an education war is raging over how and what their children are taught, and how education should be delivered in America's schools. The war has formed a fault line running through teacher's lounges, PTOs, school boards, professional organizations, legislatures, colleges and universities, and the various academic disciplines. It is a war between the 'traditionalists' and the 'progressives.' Two divergent underlying philosophies are causing the 'war.' The traditionalists believe that children should be given a strong foundation, rich in content, in a structured environment by teachers trained in their disciplines. Important, relevant facts should be learned and memorized so they become the foundation for higher-level thinking and problem solving. In contrast, progressives believe that children are capable of directing their own learning given the proper guidance, stimulation, and learning environment. Student failure is often blamed on social structures and restrictive traditional classroom practices that they say suppress the natural inclinations of a child toward learning.

What Is an Educrat? by Debra J. Saunders, San Francisco Chronicle, January 4, 1998.
"What is an educrat? ... I use [this term] because it captures a special kind of person in the education world: pinheads who are so process-oriented that they are more excited in the process of learning than the myriad wonders that can be learned. Simply put, educrats believe in process -- as opposed to educators, who believe in results. Educrats focus on how children learn. Educators focus on what they learn. ...
"What is the difference between an educator and an educrat? ..."
The remainder of this powerful article gives many succinct descriptions of those differences! Highly recommended.


Assessing Problem Solving in Simulation Games

CRESST model of learning

Isabelle Girault, David Cross and Cédric d’Ham

Different studies (Arce and Betancourt, 1997; Séré, 2002) emphasize the importance of the
task of experimental procedure design in a learning context. However, the required design is a
difficult task for students (Séré and Beney, 1997). Consequently, students are hardly ever
allowed to design their own experiment. The study by Tiberghien et al. (2001) showed that
“to learn how to plan an investigation in order to address a specific question or problem” was
the least frequent process objective. Although this paper does not focus on direct learning
gains, th2000).

Students’ adaptation to a new situation:
the design of an experimental procedure


25 Learning Principles to Guide Pedagogy and the
Design of Learning Environments

*** ICT U Can!

There is a misconception that Inquiry means letting the kids loose on learning with 'minimal instruction'. Inquiry is not an either / or dichotomy. It does not mean that you either teach them or let them teach themselves. It is a rigorously framed and supported structure that relies on the elements in Vygotsky's zone of proximal development to guide students through their learning. In Kath's model, Inquiry means engaging students in deeper more meaningful learning than direct instruction alone can offer."

("Australian Educator Kath Murdoch...
Kath is the author of the book "Classroom Connections" and we follow her inquiry model in our classrooms in my ICT cluster in Dunedin.

Kath states: Why inquiry?

  • Vehicle for integration of the curriculum
  • fosters connected rather than episodic teaching and learning
  • Caters for range of learning styles
  • Transferable process
  • Taps into students' curiosity

What view of the child does an inquiry approach assume?

Kath states: In an inquiry approach, we must firmly believe in students as:

  • curious
  • searching for meaning
  • intelligent in a range of ways
  • experienced - something to offer
  • thinkers
  • collaborators
  • active
  • risk takers
  • co-learners
  • co-teachers

There is a fundamental assumption made right at the beginning: “Learning, in turn, is defined as a change in long-term memory.” For learning as remembering, certainly guided instruction is best, as the authors claim. But, is learning just about remembering?


Read/Write Web
Nice post that lauds the participatory aspect of e-learning 2.0, pointing to a colorful slide show by Judy O'Connell, highlighting two major aspects: new literacies that are almost all social skills; and the ability to read, write, and interact across a range of platforms. These are applied in Andrew Churches's revision of the verb lists in Bloom's taxonomy.

"Andrew shares this diagram he created with Cmap Tools, the no-cost graphic organizer tool (right-click the image to view it at a larger size):

The mind map above is shared with Andrew's permission. What do you think of the work he's done?"



Artichoke defends the paper...

When I ask teachers “Why inquiry?” many suggest that in adopting inquiry based learning, the pedagogy of student centered exploration will (in some ill defined way) introduce an “authenticity” to the “sequestered/isolated in some age sorted institutional space for 6 hours a day” classroom experience.

Teachers claim that the inquiry classroom will,

  • Rescue us from the dislocation between classroom learning and real life learning.
  • Disconnect us from “learning for the test and then forgetting learning”, and reconnect us with motivated for real life learning.
  • Protect us from “Formica Learning” – the learning that results in a veneer of inert knowledge that coexists alongside deeper naïve beliefs.

They never stop to ask if inquiry is a wag the dog pedagogy, they never ask if we have misidentified what "matters most" and what "matters least" in learning, and they never ask about face validity versus construct validity wrt learning

Do students who experience inquiry based learning environments have an understanding that is deeper, more integrated, more coherent and at a higher level of abstraction than students who learn in “one size fits all” environ


and in the comments...

Ahh Jedd, I agree that this is an article that is certainly worthy of some “learning circle critique” in New Zealand staffrooms,

My interest is not in the article argument itself so much as the way teachers` interpret what the academics represent as constructivism and inquiry learning.

I can believe many of the claims made because I have been nudged by all of the following Kirshner et al thinking when working with schools on constructivist pedagogies in New Zealand.

  • Surprisingly Extensive Scaffolding: Because students learn so little from a constructivist approach most teachers who attempt to implement classroom based constructivist instruction end up providing students with considerable guidance aka scaffolding
  • Subsequent Tests: Learners required to solve problems perform worse on subsequent test problems than learners who study the equivalent worked examples – fits with cognitive load theory
  • Worked examples: Novices studying worked examples – are invariably superior to novices discovering or constructing a solution to a problem
  • Learning nothing: Learners can engage in problem solving activities for extended periods of time and learn almost nothing
  • Expertise reversal effects: issues for how we teach gifted students in inclusive classrooms
  • Process work sheets add value to learning
  • 100 million blow flies can’t be wrong effects: Despite knowing less after instruction – less able learners liked the experience of inquiry

The extensive scaffolding is "a case in point" or "a point in case" – I can never figure out which creates more buy in ….the Magnet and I have developed a pedagogical framework for inquiry that addresses some of the concerns identified by Kirschner et al, and we have been using it in the schools we work with over the past year.

It identifies the role of both teacher and the role of the student, in determining the learning experiences for understanding, through Herron’s four levels of inquiry, (Herron, 1971).

The inquiry design framework we have developed celebrates the ponderous buttock by allowing fence sitting across all levels of “L” word understanding – and supports pedagogical approaches allowing:

  • Teacher driven confirmation/verification approaches (behaviourist learning theory approach), [Problem, procedure and solution given]
  • Teacher student driven guided inquiry (constructivist learning theory approach) [Problem and procedure given] or [Problem given]
  • Student driven and constructed open inquiry (co-creator connectivist learning theory approach).[Students investigate their own topic related questions ]

Both teachers and students can use the same design template to detail the learning experiences and questions planned for the inquiry. Teachers often use Level 0 Inquiry (confirmation/verification), to plan immersion activities for students before they are challenged to develop their own questions for Level 3 student driven open inquiry. Gifted and talented students can do it alone right from the start.

Academics will loathe the eclectism but teachers have enjoyed the pedagogical clarity it gives to planning

Negative Self-Efficacy and Goal Effects Revisited
Albert Bandura
Edwin A. Locke


The authors address the verification of the functional properties of self-efficacy beliefs and document how self-efficacy beliefs operate in concert with goal systems within a sociocognitive theory of self-regulation in contrast to the focus of control theory on discrepancy reduction. Social cognitive theory posits proactive discrepancy production by adoption of goal challenges working in concert with reactive discrepancy reduction in realizing them. Converging evidence from diverse methodological and analytic strategies verifies that perceived self-efficacy and personal goals enhance motivation and performance attainments. The large body of evidence, as evaluated by 9 meta-analyses for the effect sizes of self-efficacy beliefs and by the vast body of research on goal setting, contradicts findings (J. B. Vancouver, C. M. Thompson, & A. A. Williams, 2001; J. B. Vancouver, C. M. Thompson, E. C. Tischner, & D. J. Putka, 2002) that belief in one’s capabilities and personal goals is self-debilitating.

Control theory would have coaches send
their team on the playing field in a self-doubting frame of mind;
otherwise they will play complacently. We would lay heavy bets
against a team coached according to Vancouver’s control theory.
A resilient sense of efficacy provides the necessary staying
power in the arduous pursuit of innovation and excellence. During

social cognitive theory is founded on an
agentic perspective to human self-development, adaptation, and
change (Bandura, 2001). This theory specifies four core features of
human agency, which include intentionality, forethought, self-
reactiveness, and self-reflectiveness. People form intentions that
include plans and strategies for realizing them.

Bill Kerr

Their arguments that inquiry learning approaches:

…ignore both the structures that constitute human cognitive architecture and evidence from empirical studies over the past half century that consistently indicate that minimally-guided instruction is less effective and less efficient than instructional approaches that place a strong emphasis on guidance of the student learning process. (Kirschner et al (2006).

There is commentary by artichoke at this wag the dog blog entry and also at this wiki entry which provoked some critical discussion. Read both arti's original and the comments.

I'm revisiting the Kirschner et al critique of constructivism (Bill Oct10, 2006)

The URL of the paper has changed, just tracked it down, the new URL is here(pdf)

also go here for gel papers:

gel = guided experiential learning.
there are quite a few papers that look interesting, including one on the role of deliberate practice in the role of acquiring expertise, which I agree with Ericsson

the contradiction b/w the Kirschner and Ericsson papers is resolved thus:
constructionism / constructivism does work provided it helps to motivate individuals in effortful study - constructionism can achieve this more readily than other methods, not for all, but does work for those who become motivated - the teacher needs to be expert and understand what is happening

constructionism as developed by Papert et al is a method of subtle (environmental) intervention, yes, there is scaffolding but it is relatively unobtrusive - with scaffolding being removed (the teacher getting out the way and letting students create) where appropriate - other approaches may not enable able students to flourish in this way, they may always keep students chained up

the point is that constructivism used in this way is a form of guided experiental learning, it is just that the method of guidance is much more sophisticated and can create more interesting / enjoyable classroom environments, for both students and teachers

this was certainly explained clearly by Harel and Papert - only Papert's 1980 book Mindstorms is cited in the references, a lot of very good research is ignored in this "authoritative" study

constructionism / discovery learning does not work as some sort of generalised "weeties for the brain" in traditional School environments - however, the claim that children often learn more (through play) before they come to school than they learn at school ought to be not forgotten in thinking these issues through

"Games are... the most ancient and timehonored (sic) vehicle for education. They are the original educational technology, the natural one, having received the seal of approval of natural selection. We don’t see mother lions lecturing cubs at the chalkboard; we don’t see senior lions writing their memoirs for posterity. In light of this, the question, ‘Can games have educational value?’ becomes absurd. It is not games but schools that are the newfangled notion, the untested fad, the violator of tradition. Game-playing is a vital educational function for any creature capable of learning." (Crawford 1982)

the bit in Kirschner et al about long term memory I think is refuted in this paper
The Expert Mind. Read the section on chunking, particularly the last couple of paragraphs:

"Ericsson also cites studies of physicians who clearly put information into long-term memory and take it out again in ways that enable them to make diagnoses. Perhaps Ericsson's most homely example, though, comes from reading. In a 1995 study he and Walter Kintsch of the University of Colorado found that interrupting highly proficient readers hardly slowed their reentry to a text; in the end, they lost only a few seconds. The researchers explained these findings by recourse to a structure they called long-term working memory, an almost oxymoronic coinage because it assigns to long-term memory the one thing that had always been defined as incompatible with it: thinking. But brain-imaging studies done in 2001 at the University of Konstanz in Germany provide support for the theory by showing that expert chess players activate long-term memory much more than novices do.

Fernand Gobet of Brunel University in London champions a rival theory, devised with Simon in the late 1990s. It extends the idea of chunks by invoking highly characteristic and very large patterns consisting of perhaps a dozen chess pieces. Such a template, as they call it, would have a number of slots into which the master could plug such variables as a pawn or a bishop. A template might exist, say, for the concept of "the isolated queen's-pawn position from the Nimzo-Indian Defense," and a master might change a slot by reclassifying it as the same position "minus the dark-squared bishops." To resort again to the poetic analogy, it would be a bit like memorizing a riff on "Mary had a little lamb" by substituting rhyming equivalents at certain slots, such as "Larry" for "Mary," "pool" for "school" and so on. Anyone who knows the original template should be able to fix the altered one in memory in a trice."

Papert on instructionism, pedagogy and constructionism:
"The word instructionism is intended to mean something rather different from pedagogy, or the art of teaching. It is to be read on a more ideological or programmatic level as expressing the belief that the route to better learning must be the improvement of instruction - if School is less than perfect, why then, you know what to do: Teach better. Constructionism is one of a family of educational philosophies that denies this "obvious truth." It does not call in question the value of instruction as such ... The constructionist attitude to teaching is not at all dismissive because it is minimalist - the goal is to teach in such a way as to produce the most learning for the least teaching ... an African proverb: If a man is hungry you can give him a fish, but it is better to give him a line and teach him to catch fish himself"
(The Children's Machine, 139)

Kirschner et al say "Minimal Guidance During Instruction Does Not Work" No qualification there, in that headline grabbing dogma. I say, it can work, but you need good teaching materials and an expert teacher.

Constructivism is not a new idea. It was susinctly put by Plutarch, 2000 years ago. "The mind is not a vessel to be filled but a fire to be kindled." Learning is not a one way transfer of information to the learner, the "chalk and talk" that has characterised schools. A passion for learning is essential. (Tony, May?)

Constructionism, with a N:
Papert's beliefs are rooted very firmly in Piaget's findings about children's learning. Papert worked with Piaget for 5 years, applying his own expertise in maths to help build Piaget's theories.(2) Two points from Piaget stand out:
  • Children build or construct their own intellectual structures.
From this point arises the obligation of the modern teacher to restructure traditional subjects such as maths to fit the child.
(this statement is qualified below in discussion). Hence, Papert has restructured maths by inventing the computing language logo to fit the natural development of the child.
  • Children build on what they know. Piaget's term for children's continual balancing of existing cognitive structures with new experiences is equilibration.
From this point arises the obligation of the modern teacher to investigate the cognitive structures of their students and to interact with those cognitive structures in a subtle, not a heavy handed manner.
Piaget found that incredible amounts of learning occur without formal teaching. In his work, Papert tries to discover and promote the factors that are causing this "hidden" learning and also asks: Why is it that learning often does not occur with formal teaching (and often does occur without formal teaching)?

Harel and Papert (1990) argue that some materials are better with regard to the following criteria:
  • appropriability (some things lend themselves better than others to being made one's own)
  • evocativeness (some materials are more apt than others to precipitate personal thought)
  • integration (some materials are better carriers of multiple meaning and multiple concepts)
(Bill May22)

So, learning materials such as the logo programming language, LEGO TClogo control technology, Instructional Software Design Project (a sophisticated teaching approach) or GameMaker (another evocative low entry high ceiling programming language) in combination with an expert teacher (not just technically expert either) are necessary for the constructionist approach to be effective. (Bill Oct10, 2006)
Fairly comprehensive comments, including discussion, explaining constuctionism at the game learning wiki


Papert's Principle: Some of the most crucial steps in mental growth are based not simply on acquiring new skills, but on acquiring new administrative ways to use what one already knows
- from Minsky's Society of Mind, 10.4

This was used by Papert (in the 1960s) to explain the results of Piaget's conservation experiments. There is a diagram showing the agents involved in Minsky's book. Most previous theories had suggested that children developed different kinds of reasoning as time goes by. Papert suggested how the different ingredients were organised. a mind cannot grow very much merely by accumulating knowledge, it must also develop better ways to use what it knows.

Jeremy Hiebert

Online Self-Organizing Social Systems David Wiley was kind enough to direct me to a paper he wrote with Erin K. Edwards a few years back: Online Self-organizing Social Systems: The Decentralized Future of Online Learning.

I had seen references to it, but couldn't find a copy. This look at online self-organizing social systems ("OSOSS") from a learning perspective is similar to papers I found earlier from David Passmore.

Passmore also looked at PerlMonks (my thoughts here) and investigated learning in Slashdot (my thoughts here), and came to similar conclusions.

From Wiley and Edwards: "While none of the existing OSOSS consider themselves learning communities, learning is happening among their users, and happening in an extremely innovative manner."I'm seeing some evidence to support this in 43 Things, especially in reference to explicit learning communities that form around learning goals.

James Paul Gee
Video Games, Learning, and “Content”,_Learning_and_Content.html

(mostly about how experiences need to be shaped to aid learning)

Learning in video games—learning in terms of the Situated Learning Matrix—is not “anything goes”, or “just turn the learners loose to do their own thing.”41 There is a good deal of guidance in games. Guidance from the game design itself, from the NPCs and the environment, from information given “just in time” and “on demand”, from other players in and out of the game, and from the resources of communities of practice built up around the game.


Postscript to a Debate Between Constructivists and Supporters of Explicit Instruction
Sigmund Tobias

The publication of the article by Kirschtner, Sweller, and Clark (2006) led to a series of rejoinders to that paper (Schmidt, Loyens, Loft & Paas, 2007; Hmelo-Silver Duncan, & Chinn, 2007; Kuhn 2007; Sweller, Kirchner, & Clark, 2007). You may recall that the original article was also discussed in Issues No 3 and 4 of this Newsletter [provide link]. In addition, the original article stimulated the scheduling of a debate at the 2007 conference of the American Educational Research Association regarding the paper’s main assertion that approaches to minimally guided instruction (constructivism, discovery learning, problem based learning, etc…) had failed. Debate participants included two critics of minimal instructional guidance: Paul Kirschner, the first author of the article assisted by comments from the paper’s second author John Sweller, and Barak Rosenshine, a well known critic of minimally guided approaches to learning. Defenders of constructivism included two noted scholars identified with that orientation: David Jonassen, and Rand Spiro.

The purpose of this note is to report a narrowing of the differences regarding the issues posed in the Kirschner et al. (2006) article in the debate as well as in some of the rejoinders to the article (see Hmelo-Silver et al. 2007; Schmidt et al., 2007) and to suggest some research and theoretical development that may ultimately resolve the issues posed.

In the debate the constructivists agreed that some form of direct, or explicit instruction (Sweller et al., 2007) may be useful for students with little prior knowledge of well structured domains. Similarly, Kuhn (2007) acknowledged that there is a place for explicit instruction, and Schmidt et al., (2007), as well as Hmelo-Silver et al. (2007) agreed that some instructional guidance was necessary. During the debate, the constructivists also agreed that some form of guidance, or instructional support (Tobias, 1982) more generally, was essential in all instructional approaches. The constructivists suggested that instruction in ill structured domains, such as problems in medical diagnosis, advanced engineering design, even teaching students the meaning of a complex concept such as “justice” were difficult if not impossible with explicit instructional approaches and could only be managed with instruction following a constructivist orientation. The critics of the constructivist position acknowledged that ill structured domains presented difficult problems and ought to be approached by a variety of means.

A consensus developed in the debate suggesting an interaction among prior knowledge, organization of the subject matter, and instructional method similar to the multivariable aptitude treatment interaction research recommended by Cronbach (2002. The constructivist debaters especially suggested that students’ prior knowledge was a key variable in these interactions, confirming expectations regarding the importance of prior knowledge for instructional adaptations (Tobias, 1989, 2003; Gustaffson & Undheim, 1996). It should be noted that while prior knowledge has been extensively studied in previous aptitude treatment interaction research, there were relatively few studies varying domain structure, a variable that could be profitably followed up by future research.

Instructional Support and Domain Structure
In order to achieve progress in the clarification of these issues it is important for theorists to be more precise about the definition of two terms: guidance or more generally instructional support, and the structure of a domain. Sweller et al. (2007) acknowledged the importance of greater specificity about what guidance meant. Clarification of all forms of instructional support more generally requires specification of a hierarchy so that terms like “minimal” support could be anchored in specific instructional actions rather than vague terms. For example, does asking students to repeat an answer provided by the instructor consist of more or less instructional support than providing knowledge of results, or “worked examples”? Similarly, where do forms of support such as providing explanations, whiteboards, prompts, or hints, to mention some of the supports described in the articles, fit in a hierarchy of instructional support?

It would also be useful to have specific descriptions of how domain structure may be determined. Clearly subjects like the multiplication tables, for example, constitute a well structured domain, and medical diagnosis of rare disease entities is an ill structured domain. Perhaps a second hierarchy ranging from ill structured to well structured domains should be specified, and instantiated by examples such as those given above, to facilitate research on the interaction among prior knowledge, instructional support, and domain structure.

Specification of Cognitive Processes
It should be noted that all instructional approaches need to specify the psychological processes engaged by teaching methods. Even though learning may be strongly influenced by the communities with which learners affiliate, and perhaps even by the environments in which learning occurs (Tobias, 2003), the psychological processes presumably engaged or enhanced by participation in communities need to be specified. Do constructivist activities enhance such processes as attention, retention in working memory, or storage and retrieval from long term memory, to mention only some examples? Similarly, which of these processes, or others, are invoked by explicit instructional methods? Furthermore, which of the approaches stimulates more frequent processing of instructional material either with the processes suggested or by others?

The accuracy of Kirschner et al. (2006), the rejoinders to it (Schmidt et al., 2007; Hmelo-Silver, et al., 2007; Kuhn 2007), and the response to the rejoinders (Sweller, et al., 2007) will ultimately be decided by research, rather than by rhetoric or debates. As noted some time ago (Tobias, 1982), different instructional approaches can lead to varying outcomes only if they either engage different cognitive processes controlling learning, or invoke those processes more or less frequently. Now, that some of the different views have been somewhat clarified by the publications and by the debate, would be a good time to conduct research concerned with the cognitive processes engaged by different teaching methods. Such investigations will ultimately settle the issues discussed in the debate and in the publications dealing with the issue. I join Sweller et al. (2007) in urging that further experimental research is needed, and would add that in order to be most useful, such research should aim to identify and examine both the cognitive processes engaged by the instructional methods used, and the frequency with which the cognitive processes are used by different instructional approaches.



question of finding what you're looking for - do 'beliefs' exist...? Churchland, etc

measuring... like measuring the distance from the edge of the world..
now we have never seen the edge of the world, so we'll crate a baseline reference and say other things are closer & further - so London is, say ... 3000 kn from the edge, and we're at London +500, whatever

and this becomes a system of criticizing subsequent things... like that new tech, a 'globe'
... you can't measure distance to the end of the world, all the calculations are wrong and don't accord with observation

so... you had a 'belief' and you had 'more'... but what are they?


  1. You didn't seem to find the actual published responses to the article.

  2. No I didn't. They don't show up on Google.

    How's that for a 'scientific community'.

  3. Clark was kind enough to put the responses on his own site:

    Clark is also against videogames and animated pedagogical agents, apparently:

  4. Wow. That's an impressive collection of opinions. I learned a bit about how people see the world from reading through it.

    I'm glad you chose this particular paper. A couple of people have quoted it as "proof" that informal learning doesn't work. Heh!

    The authors set up a bogus dichotomy as the core of their arguments. Comparing heavily-guided instruction to free-form exploration is nonsense. (Extremism usually is). And the claim that retention in long-term memory is the appropriate measure of learning shows that the authors lack street smarts.


  5. I'm glad you chose this particular paper. A couple of people have quoted it as "proof" that informal learning doesn't work. Heh!


I welcome your comments - I'm really sorry about the moderation, but Google's filters are basically ineffective.