Thinking

Annotations

Ben Wilbrink

Thinking, of course, is a major theme in philosophy, psychology, the neurosciences, and education. My selection of articles and books simply is: is this article or book relevant to questions of design of achievement test items?

For example, Thorndike's association theory was very influential in educational design in the early twentieth century, especially so in mathematics. His associationism is back again, in the highly sophisticated guise of Parrallel Distributed Processing models of 'thinking' in the later twentieth century.

These PDP-models promise to be adequate to the physical structure of the brain, and therefore to its thinking processes. Let me explain in a few words the issue here. Most models in cognitive psychology in the second half of the twentieth century assume thinking to be a linear process, much like the way programmed computers 'think.' However, it is obvious from a little introspection that the brain does not function like a computer. Literally looking inside the brain reveals a multitude of brain cells, all of them being more or less active all of the time. Something heavily 'parallel' is happening here. Thinking about thinking, it is evident that there is not a little thinking man inside our skull, nor are there replica's of the things and events that might be the 'objects' of our thinking. There simply are no brain cells dedicated to storing and giving back a particular name, picture, event, and not other names, pictures, or events. The question then is, how does the brain function in the process that we somehow observe to be the retrieval of information or the construction of new information on the basis of items of information somehow available to or in the brain? What is happening between 'input' to and 'output' from the brain? Between the two a neural network does the job; the network is able to do theat job because in previous learning its multitude of connections have been tuned to deliver the right output. What is more, the same network also is fine-tuned to deliver adequate output to a lot of other possible inputs. PDP-models are about the mechanisms that make this kind of multiple use of the same connections possible. No homunculus, programmer, genii or god is necessary to make this kind of functioning of neural networks possible. The impact of this kind of theory about thinking on didactics should be enormous, once the educational commuty wakes up to its importance. Carl Bereiter's (2002) book is a wake-up call, for example. Thanks to him, I finally realized that PDP models are not simply fascinating in themselves, but will be highly relevant to instructional science also, complementary--at the micro-level--to much of cognitive psychology's--macro-level--achievements of the seventies and eighties of the last century.

Vision is an area that lends itself to some illustrations of what is happening in this Parallel Distributed Processing.

Frank Werblin and Botond Roska (April 2007). The movies in out eyes. Scientific American, 54-61.

"The retina processes information much more than anyone has ever imagined, sending a dozen different movies to the brain."
film of images being sent to the brain. The remarkable thing is that the 'images' almost literally 'sent to the brain' are extremely coarse, meaning the brain's neural networks have to work very hard to construct the images the rabbit 'sees.' Humans are related to rabbits, remember? The hard work of the brain = Parallel Distributed Processing. The film is about several different images being sent to different parts of the brain, take a peak. Somehow or other the brain gathers that distributed information together again, producing the wonderful world we think we see 'directly.' You do not believe this? Try looking through a stereoscopic viewer to some old fashioned stereoscopic photographs of the Liberty Statue or whatever: it is not easy to 'see' only one picture in depth, but suddenly it is there! In a split second the brain's neural networks have figured it out! Nothing 'you' can do about it!
Of course, in the case of vision, the retinal receptors take the input, the world you 'see' is the output of the neural network or the parrallel distributed processing in between. Some of this brainy hardware must already come pre-programmed at birth, but most of the delicate structures of this neural network have been 'learned' or 'tuned' during your lifetime.

"By encoding each piece of knowledge as a large set of interactions, it is possible to achieve useful properties like content-addressable memory and automatic generalization, and new items can be created without having to create new connections at the hardware level."

Rumelhart & McClelland 1987 p. 108

It is easy to see that the citation given is highly relevant to the learning of categories and concepts, very much the basics of what learning is in life as well as in school. The implication is that the design of achievement test items at the level of instances of concepts etcetera, has to reckon with the characteristics of the underlying hardware, in this case the hardware being the student's brainy neural networks. A special case here undoubtedly will be that of the kind of 'wrong' answers in multiple choice questions; the risc being that the wrong kind of wrongness will teach students wrongness itself. There is indeed recent research documenting adverse effects, see chapter two of 'Toetsvragen ontwerpen' html.

Concepts in their turn figure in artificial intellegence's cognitive schemas. How about using PDP models at the schema-level? Rumelhart and McClelland discuss the possibilities in their chapter 14.

Carl Bereiter (2002a). Education and Mind in the Knowledge Age. Erlbaum. questia

My summary and annotations in a special page
Een sterk statement over de stand van zaken in wat cognitieve wetenschappen en hedendaagse filosofie over de inrichting van de leeromgeving te melden hebben.

Parallel Distributed Processing - Connectionism

How the brain's hardware might handle cognition. Ultimately, achievement testing should be valid to the brain's microprocessing of information given and information asked back. Not many specialists in educational measurement ever even offered this suggestion. Carl Bereiter? He is not especially a 'measurement specialist.' Any readers having suggestions for recent research that might be especially relevant to this issue, please let me know.

David E. Rumelhart, James L. McClelland, and the PDP Research Group (1986). Parallel distributed processing. Explorations into the microstructure of cognition. Volume 1: Foundations, 2: Psychological and biological models.. The MIT Press.

James L. McClelland and David E. Rumelhart (1988). Explorations in parallel distributed processing. A handbook of models, programs, and exercises. The MIT Press.

Rogers, T. T. and McClelland, J. L. (2004). Semantic Cognition: A Parallel Distributed Processing Approach. Cambridge, MA: MIT Press. Not (yet?) in questia.com

Paul Grobstein Simple networks, simple rules: Learning and creating categories. Serendip site. html

David L. LaBerge, & S. Jay Samuels (Eds) (1977). Basic processes in reading: perception and comprehension. Lawrence Erlbaum. (i.a. James A. Anderson: Neural models with cognitive implications - Eleanor J. Gibson: How perception really developes: A view from outside the network - Herbert H. Clark: Inferences in comprehension - David E. Rumelhart: Understanding and summarizing brief stories - John R. Anderson: Computer simulation of a language acquisition system: A second report)

Christian Lebiere, Marsha Lovett, Paul Munro, Christian Schunn (2004). Proceedings of the Sixth International Conference on Cognitive Modeling: 6th ICCM 2004 Integrating Models, July 30-August 1, 2004, Carnegie Mellon University, University of Pittsburgh, Pittsburgh, Pennsylvania, USA. Erlbaum. questia

a.o. Yasuaki Sakamoto, Toshihiko Matsuka and Bradley C. Love: Dimension-Wide vs. Exemplar-Specific Attention in Category Learning and Recognition - David Danks: Constraint-Based Human Causal Learning - Helmar Gust, Kai-Uwe Kuhnberger and Ute Schmid: Ontological Aspects of Computing Analogies - Alexei Samsonovich and Kenneth DeJong: A General-Purpose Computational Model of the Conscious Mind - Lael J. Schooler and Ralph Hertwig: How Forgetting Fosters Heuristic Inference - M. Afzal Upal: A Computational Model for Acquisition of Counterintuitive Concepts

Philip T. Quinlan (Ed.) (2003). Connectionist Models of Development. Developmental Processes in Real and Artificial Neural Networks. Psychology Press.. questia

Gary F. Marcus (2001). The algebraic mind. Integrating connectionism and cognitive science. The MIT Press.

Cognitive psychology

P. N. Johnson-Laird (2004). The history of mental models. In Ken Manktelow and Man Cheung Chung: Psychology of Reasoning. Theoretical and Historical Perspectives. Erlbaum questia

Paul Bloom (2004). Descartes' baby. How child development explains what makes us human. London: William Heinemann.

Carey, S. (1992). The origin and evolution of everyday concepts. In R. Giere (ed.), Cognitive Models of Science (Minnesota Studies in the Philosophy of Science, Vol. XV). Minneapolis: University of Minnesota Press, 89-128. pdf

Instruction - Education

Lieven Verschaffel, Filip Dochy, Monique Boekaerts and Stella Vosniadou (Eds) (2006). Instructional psychology: Past, present, and future trends. Sixteen essays in honour of Erik de Corte. Elsevier.

from the contents 2. Mathematics in the Mind: Architecture, Development, and Educational Implications. Andreas Demetriou and Areti Panaoura 3. Attentional Processes, Abstraction, and Transfer in Early Mathematical Development. Erno Lehtinen and Minna M. Hannula 4. Examining Mathematics Learning From a Conceptual Change Point of View: Implications for the Design of Learning Environments Stella Vosniadou and Xenia Vamvakoussi pdf 5. Reasoning With Mental Tools and Physical Artifacts in Everyday Problem-Solving. Roger Saljo, Ann-Charlotte Eklund, and ?sa Makitalo 6. Modelling for Life: Developing Adaptive Expertise in Mathematical Modelling From an Early age. Wim Van Dooren, Lieven Verschaffel, Brian Greer, and Dirk De Bock 8. Student Learning in Context: Understanding the Phenomenon and the Person Noel Entwistle, Velda McCune, and Max Scheja 12. The Difficult Marriage Between Education and Technology: Is the Marriage Doomed? Gavriel Salomon and Dani Ben-Zvi 15. From Individual Learning to Organizational Designs for Learning Lauren B. Resnick and James P. Spillane

Ann L. Brown (1992). Design Experiments: Theoretical and Methodological Challenges in Creating Complex Interventions in Classroom Settings. The Journal of the Learning Sciences, 2, 141-178. pdf

The story of a personal tour of discovery. Introduction to the literature on thinking in education. Use its title to find more recent expositions. Use its list of references to locate the classics in the field.

E. Fischbein (1975). The intuitive sources of probabilistic thinking in children. Dordrecht: Reidel.

Deanna Kuhn (1991). The skills of argument. Cambridge University Press.

Deanna Kuhn (2005). Education for thinking. Harvard University Press.

This is an exposition of what thinking in education should be. Use it to find out what the shortcomings are of the many attempts to introduce 'thinking' in education. Do not be mistaken about this text: it is evidence based, and the evidence in the book itself consists of highly revealing reports of classical teaching of 'thinking' as it typically is practiced by teachers everywhere in the U.S., or the world, for that matter. I my opninion, this is a key publication on education for the early twentyfirst century. I will use it to lay a secure fundament under my design technology for achievement test items.
My summary and annotations in a special page

Marlene Scardamalia (2002). Collective Cognitive Responsibility for the Advancement of Knowledge. In Barry Smith: Liberal Education in a Knowledge Society. Open Court. [" This volume looks at the thinking of educational theorist Carl Bereiter, who has tackled the problem of the liberal education canon in a new way."] pdf

related on my website

See the sitemap

links

Michelene Chi's publications

April 13, 2007 \contact ben apenstaartje benwilbrink.nl

http://www.benwilbrink.nl/literature/thinking.htm