Thinking and education, will they go together?

Onderwijs en denken, een onheilig verbond

Ben Wilbrink

Is thinking a verb? tweet thread. We urgently need insights in how thought processes unwind, what their (knowledge) conditions are, what it is to validly ‘test for thinking’.

The intention is to develop a new blog format, on controversial topics. The idea is to choose a Dutch innovation in math education — ‘mathematical thinking activities’ [in Dutch: wiskundige denkactiviteiten, WDA], or a more general topic: thinking is not accessible through introspection, therefore talking of ‘lsquo;learning to think (critically’ and ‘21st century skills’ is obscurantism (Jon Elster, 2015).

Thinking is not accessible for introspection, therefore talk of learning to think is idle.

Thinking is not accessible for introspection. Somehow it happens (autonomously), the results become available in working memory tweet May 2, 2015

Introspection. How can it be possible?

tweet december 20, 2015 Introspection on autonomous brain processes is fiction. What we call ‘thinking’ are merely their results.

Stellan Ohlsson (2011). Deep Learning: How the Mind Overrides Experience. Cambridge University Press. info

Not much is said by Ohlsson explicitly on introspection. However, the whole book implicitly is about the impossibility of valid introspection on autonomous mental processes.

Remembering a name is ‘thinking’: failing to remember a name, what can one do? How can one ‘learn to remember’? Teach it? Ohlsson explains. Constructivism/progressivism claims to know how to teach (children) how to ‘learn to think’ (like, for example, a mathematician, like a 21st century thinker). Is ‘learning to think’ an oxymoron? I’ll try to flesh out the controversy between cognitive science and educational theories the coming weeks.

The power of cognition resides, not in limited working memory, but in richly filled declarative memory: domain knowledge does the trick.

Mario Lopez: Can you teach somebody to learn to remember?

Paul Westerbeek: Remembering a name isn’t thinking. Operations on contents of memary are.

Talk of thinking, as for example by logicians, is not in fact talk of cognitive processes, but of their results. For an illustration, see my annotation on Guttenplan (1991) below.

Jan Albert Brinkman (1990). The analysis of fault diagnosis tasks. Do verbal reports speak for themselves? Dissertation Technical University of Eindhoven. (Promotoren J. Moraal en Jen Algera. Aha) pdf

Research on two somewhat contrasting methodologies:

  1. Richard E. Nisbett & T. D. Wilson (1977). Telling more than we can know: Verbal reports on mental processes. Psychological Review, 84, 231-259.
  2. K. Anders Ericsson & Herbert A. Simon (1980). Verbal reports as data. Psychological Review, 87, 215-251. abstract; scan of article
    K. Anders Ericsson & Herbert A. Simon (1984). Protocol analysis: Verbal reports as data. Cambridge, Mass.: MIT Press. scan Ch. 1; info book

Spatial thinking skills?

tweet that poses the problem. See the discussion on Twitter (I will summarize that on this page. Soon)

Cognitive science

Samuel Guttenplan (1986/1991). The languages of logic. An introduction to formal logic. Blackwell. isbn 0631146253 info

This book is not psychology at all, even where its terminology on thought suggests it is. It might be useful to contrast the logician and the psychologist, so study the followong passage. Psychologically, the ‘thoughts’ are assumed to be generated by previous thoughts (which does explain nothing at all), a process that implicitly is assumed to be not complicated, therefore a deus ex machina. In psychological theory this is not acceptable. The Guttenplan exposition might be excellent logic, it isn’t psychology (he does not claim it to be psychology, of course [I hope]). What Guttenplan here calls ‘thinking’ is, for all practical purposes, nothing more than a written series of sentences; no thinking at all, but presumably the results of some thinking. It is important here to be explicit and clear: Guttenplan does not explain what needs an explanation if one’s intention, Guttenplan’s intention, would to spell out some rules for one’s thinking, instead of only for the (acceptability of) results of that thinking.

Philip Johnson-Laird (1993). The computer and the mind. An introduction to cognitive science. info

Part IV is on thought. In ch. 12, on deduction, he uses validity as a criterium of thought; why should one do that? There seems to be nothing in this chapter on any explaining proces in the head (in stead of in a computer program). Ch. 13, on induction, is a good story on concepts; here also thought is being analyzed in terms of what I would call external characteristics: explanatory processes stay hidden in a black box. The chapter on creativity (h. 14) is disappointing: no explanation deserving that label at all. Johnson-Laird even calls doing arithmetic determinstic! Wow, what a misconception.

It is a popular book, the Johnson-Laird theory and research on mental models gets barely mentioned in this Part IV.

Chi, M.T.H. (1996). Constructing self-explanations and scaffolded explanations in tutoring. Applied Cognitive Psychology, 10, 33-49. pdf

Niels Taatgen and John Anderson (2010). The Past, Present, and Future of Cognitive Architectures. abstract and free download; pdf

C. Lebiere (1998). The dynamics of cognition: An ACT-R model of cognitive arithmetic. Ph.D. dissertation, Carnegie Mellon University. pdf

Not cognitive science?


Anthony Radice (January 7, 2015). The myth of thinking skills. blog

Anthony is on Twitter to follow @AnthonyRadice1.

Yes, (institutional) protagonists of 21st C skills are trying to resurrect the kind of education that Latin was supposed to offer. And math! Let’s not forget the darker side: selectivity by means of Latin & math, cooling out kids of the ‘lower classes’. The same is happening again, under the cloak of meritocracy: emphasizing problem solving & creativity puts a premium on IQ. Learning psychology would emphasize knowledge acquisition, putting a premium on effort. Cognitive science shows knowledge (from long term memory, not from internet) to be conditio sine qua non for problem solving. Yes, ‘Aiming for pupils to store core knowledge in long term memory.’ is the right thing to do, not wasting student time.

Donald Clark (June 16, 2013). Latin – 10 reasons why it should NOT be taught in schools. blog

Donald Clark (April 04, 2012). Thorndike (1874 – 1949) – experimental rigour, transfer and why Latin is a waste of time blog

Helen C. Reed, Claire Stevenson, Marije Broens-Paffen, Paul A. Kirschner, Jelle Jolles (2015). Third graders’'verbal reports of multiplication strategy use: How valid are they? Learning and Individual Differences, 37, 107-117. pdf

Onderdeel van het proefschrift van Helen Reed Mathematical thinking, learning and performance. The article uses the term ‘thinking’ in a loose sense, the same might be true as regards ‘strategy use’. Thinking not being available for introspection, what do the investigators use as indicators of supposed thinking processes? How close can they get to ‘thinking’?

Helen C. Reed (30 juni 2014). Mathematical Thinking, Learning and Performance. Insights and Interventions for Primary and Secondary Education. Vrije Universiteit. isbn 9789462591868

A remarkable statement; also, the second sentence is a homunculus-formula, the it that is doing the understanding and reasoning.

L. B. Resnick (1987). Education and Learning to Think. National Academy Press. (Zie hier voor download van pdf.)

Het is een boekje geschreven door Lauren Resnick; een aantal mensen hebben meegelezen en kritiek geleverd; een uitgelezen gezelschap (zonder dollen): Carl Bereiter, John Bransford, Ann L. Brown, Jerome S. Bruner, Susan Carey, Allan Collins, Robert H. Ennis, David Perkins, & Roger Schank, Wat dus niet betekent dat Bereiter en anderen instaan voor de uitspraken van Lauren Resnick! Ik ben het stuk aan het lezen en moet zeggen dat het een interessant overzicht en ook een tijdsbeeld geeft. Het is een kritische bespreking, wat niet wegneemt dat de ondergrond is dat Lauren Resnick graag ziet dat die hogere denkprocessen meer plaats krijgen in het onderwijs. Op zich, zou je denken, is er niets op tegen om zo’n wens te hebben. Iets anders is om daar dan ook voor te gaan pleiten, ook al is de empirische evidentie van een en ander vooral afwezig. Lauren danst een beetje om dit dilemma heen, althans in 1987. Ik zal nog proberen het precies te krijgen. Het stuk lijkt nog steeds springlevend te zijn: Dai (2012) verwijst er prominent naar.

Generally speaking, people rely on powerful but only narrowly applicable thinking methods in domains in which they are expert and use broadly applicable but weak methods for learning and thinking in fields they know little about. Good thinkers need both the powerful but specific and the general but weak kinds of skills.

p. 46

Ik ben ervan overtuigd dat de eerste uitspraak vandaag ook ondersteund wordt in de theorie van Ohlsson (2011). Het is bijna een open deur, zou je zeggen. Ook A. D. de Groot beschreef dit fenomeen in zijn proefschrift over het denken van de schaker: zodra de grootmeester het zonder zijn expertise moet doen, bijvoorbeeld bij de opgave om de positie van willekeurig geplaatste stukken op het schaakbord te onthouden, doet hij het niet beter dan de nieuweling in het schaken. Of zoiets. Oké. Maar neem nu de tweede uitspraak van Resnick: dat is gibberish. Ik weet niet wat die uitspraak betekent, als er al een precieze bedoeling achter ligt. En daarmee is in een notendop geschetst wat het probleem is met het overzicht van Resnick: op zich correcte conclusies uit de wetenschappelijke literatuur, maar vervolgens praat zij in onzorgvuldige termen over wat er in het onderwijs mogelijk of gewenst zou zijn. En dat is toch problematisch, omdat ideologische hoogvliegers er argumenten aan ontlenen om hun betogen voor meer ‘probleemoplossen’ in het onderwijs kracht bij te zetten.

André Tricot & John Sweller (2014). Domain-specific knowledge and why teaching generic skills does not work. Educational Psychology Review preview & concept

John Dewey (1912). How we think. Dover. [eBook KB]

Thinking about thinking, there is no ‘actor’ ‘doing’ the thinking (as is introspecitively clear). I am looking for terms to describe this. tweet thread

Robert Fisher (1990 6th). Teaching children to think. Cheltenham: Stanley Thornes. isbn 0748722351 homepage

The first sentence tells it all. Fsher has tried to connect his ideas to psychology — that does not mean he succeeded in doing this, and surely not that psychology supports his ideas. Well, my interest in the book: a 1990 protagonist of creativity in the school tells it all, and a lot better, I think, than Sir Ken does it a few decennia later. Nevertheless, I am afraid it is misleading. Thinking is an autonomous process (see Stellan Ohlsson's book, 2011), there isn’t a homunculus in the head that can learn how to do the thinking.

David N. Perkins & Tina A. Grotzer (1997). Teaching intelligence. American Psychologist, 52, 1125-1133. download

Ron Ritchhart David N. Perkins (2005). Learning to Think: The Challenges of Teaching Thinking. Ch 32 in K. J. Holyoak & R. G. Morrison (Eds.), The Cambridge Handbook of Thinking and Reasoning. Cambridge University Press. pdf

Lacking a theoretical framework. The theory being, of course, that there is no generic thinking skill, there are domain-specific thinking skills.

12 februari 2017 \ contact ben at at at    

Valid HTML 4.01!