Mathematical Cognition 1 A Parallel-Distributed Processing Approach to Mathematical Cognition1

How should we think about the nature of our knowledge of mathematical concepts, and the mechanisms we use to learn and use these concepts when we do mathematics? Here we describe a perspective on the answers to these questions and a future research program to address them that is grounded, in part, in the parallel distributed processing [PDP] approach to cognition and learning [1,2] implemented in artificial neural networks. We begin with a more basic question, namely the nature of mathematics and mathematical reasoning, and proceed from there to consider mathematical knowledge, learning, and thinking, stressing the roles of culture and experience in the creation and learning of mathematics. We then review the PDP perspective on the nature of knowledge and learning, and consider how it can address many findings in the mathematical cognition literature, and how it provides alternative ways of understanding what nature may provide and what nurture may create. Next we discuss the exciting challenges facing the approach and how they might be addressed, organizing the discussion around the question: How might a neural network-based approach meet the challenge of learning to achieve a level of competence in algebra and geometry sufficient to pass a highschool proficiency exam in geometry, in a human-like way, from experiences similar to those of human learners? We conclude with a discussion of the implications of the approach for learners and teachers of mathematics and for the processes of teaching and learning.

[1]  Gerald Tesauro,et al.  TD-Gammon, a Self-Teaching Backgammon Program, Achieves Master-Level Play , 1994, Neural Computation.

[2]  Ha Hong,et al.  Performance-optimized hierarchical models predict neural responses in higher visual cortex , 2014, Proceedings of the National Academy of Sciences.

[3]  Nicole M. McNeil U-shaped development in math: 7-year-olds outperform 9-year-olds on equivalence problems. , 2007, Developmental psychology.

[4]  Nikolaus Kriegeskorte,et al.  Deep Supervised, but Not Unsupervised, Models May Explain IT Cortical Representation , 2014, PLoS Comput. Biol..

[5]  James L. McClelland,et al.  Nature, nurture, and connections: Implications of connectionist models for cognitive development. , 1991 .

[6]  Pierre Pica,et al.  Exact Equality and Successor Function: Two Key Concepts on the Path towards Understanding Exact Numbers , 2008, Philosophical psychology.

[7]  James L. McClelland,et al.  U-Shaped Curves in Development: A PDP Approach , 2004, Journal of Cognition and Development.

[8]  S. Dehaene,et al.  Exact and Approximate Arithmetic in an Amazonian Indigene Group , 2004, Science.

[9]  James L. McClelland,et al.  A connectionist model of a continuous developmental transition in the balance scale task , 2009, Cognition.

[10]  R. Linsker,et al.  From basic network principles to neural architecture , 1986 .

[11]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[12]  D. Klahr Patterns, Rules, & Discoveries in Life and in Science , 2012 .

[13]  Geoffrey E. Hinton Learning and Applying Contextual Constraints in Sentence Comprehension , 1991 .

[14]  S. Dehaene,et al.  The mental representation of parity and number magnitude. , 1993 .

[15]  Mitchell J. Nathan,et al.  Embodiment in Mathematics Teaching and Learning: Evidence From Learners' and Teachers' Gestures , 2012 .

[16]  Yee Whye Teh,et al.  A Fast Learning Algorithm for Deep Belief Nets , 2006, Neural Computation.

[17]  Samy Bengio,et al.  Show and tell: A neural image caption generator , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[18]  Geoffrey E. Hinton,et al.  An Efficient Learning Procedure for Deep , 2010 .

[19]  Shane Legg,et al.  Human-level control through deep reinforcement learning , 2015, Nature.

[20]  Hilary C Barth,et al.  The development of numerical estimation: evidence against a representational shift. , 2011, Developmental science.

[21]  Justin Halberda,et al.  Developmental change in the acuity of approximate number and area representations. , 2013, Developmental psychology.

[22]  David C. Plaut,et al.  Quasiregularity and Its Discontents: The Legacy of the Past Tense Debate , 2014, Cogn. Sci..

[23]  Robert M. Capraro,et al.  Changes in Equality Problem Types Across Four Decades in Four Second and Sixth Grade Textbook Series , 2012 .

[24]  Martin H. Fischer,et al.  Finger counting habits modulate spatial-numerical associations , 2008, Cortex.

[25]  Patrick W Thompson,et al.  Quantitative reasoning, complexity, and additive structures , 1993 .

[26]  Marco Zorzi,et al.  Emergence of a 'visual number sense' in hierarchical generative models , 2012, Nature Neuroscience.

[27]  G. Lakoff,et al.  Where mathematics comes from : how the embodied mind brings mathematics into being , 2002 .

[28]  A. Henik,et al.  Is three greater than five: The relation between physical and semantic size in comparison tasks , 1982, Memory & cognition.

[29]  Oren Etzioni,et al.  Solving Geometry Problems: Combining Text and Diagram Interpretation , 2015, EMNLP.

[30]  Michael Cole,et al.  Culture and thought; a psychological introduction , 1974 .

[31]  James L. McClelland,et al.  Learning hierarchical category structure in deep neural networks , 2013 .

[32]  Zenon W. Pylyshyn,et al.  Connectionism and cognitive architecture: A critical analysis , 1988, Cognition.

[33]  Eric A. Jenkins,et al.  How Children Discover New Strategies , 1989 .

[34]  Robert S. Siegler,et al.  Representational change and children’s numerical estimation , 2007, Cognitive Psychology.

[35]  Justin Halberda,et al.  Developmental change in the acuity of the "Number Sense": The Approximate Number System in 3-, 4-, 5-, and 6-year-olds and adults. , 2008, Developmental psychology.

[36]  Steven Pinker,et al.  On language and connectionism , 1988 .

[37]  Mitchell J. Nathan Rethinking Formalisms in Formal Education , 2012 .

[38]  James L. McClelland,et al.  Why there are complementary learning systems in the hippocampus and neocortex: insights from the successes and failures of connectionist models of learning and memory. , 1995, Psychological review.

[39]  Michael C. Frank,et al.  Number as a cognitive technology: Evidence from Pirahã language and cognition , 2008, Cognition.

[40]  James L. McClelland,et al.  Semantic Cognition: A Parallel Distributed Processing Approach , 2004 .

[41]  S. Carey,et al.  On the limits of infants' quantification of small object arrays , 2005, Cognition.

[42]  M. Alibali,et al.  The function of gesture in learning to count: more than keeping track * , 1999 .

[43]  James L. McClelland,et al.  On the control of automatic processes: a parallel distributed processing account of the Stroop effect. , 1990, Psychological review.

[44]  Pierre Pica,et al.  Education Enhances the Acuity of the Nonverbal Approximate Number System , 2013, Psychological science.

[45]  James L. McClelland,et al.  Understanding normal and impaired word reading: computational principles in quasi-regular domains. , 1996, Psychological review.

[46]  J. Shaughnessy,et al.  CHARACTERIZING THE VAN HIELE LEVELS OF DEVELOPMENT IN GEOMETRY , 1986 .

[47]  Quoc V. Le,et al.  Sequence to Sequence Learning with Neural Networks , 2014, NIPS.

[48]  Elizabeth S. Spelke,et al.  Toward exact number: Young children use one-to-one correspondence to measure set identity but not numerical equality , 2014, Cognitive Psychology.

[49]  L Girelli,et al.  The development of automaticity in accessing number magnitude. , 2000, Journal of experimental child psychology.

[50]  Geoffrey E. Hinton Learning multiple layers of representation , 2007, Trends in Cognitive Sciences.

[51]  James L. McClelland,et al.  Parallel Distributed Processing at 25: Further Explorations in the Microstructure of Cognition , 2014, Cogn. Sci..

[52]  James L. McClelland,et al.  Letting structure emerge: connectionist and dynamical systems approaches to cognition , 2010, Trends in Cognitive Sciences.

[53]  Jeffrey L. Elman,et al.  Finding Structure in Time , 1990, Cogn. Sci..

[54]  S. Carey,et al.  How counting represents number: What children must learn and when they learn it , 2008, Cognition.

[55]  James L. McClelland,et al.  A neural network model of learning mathematical equivalence , 2014, CogSci.

[56]  Rafael Núñez,et al.  No Innate Number Line in the Human Brain , 2011 .

[57]  K. Davidson,et al.  Does learning to count involve a semantic induction? , 2012, Cognition.

[58]  S. Carey,et al.  The Epigenesis of mind : essays on biology and cognition , 1991 .

[59]  Geoffrey E. Hinton,et al.  Transforming Auto-Encoders , 2011, ICANN.

[60]  S. Carey Knowledge Acquisition: Enrichment or Conceptual Change? , 1991 .

[61]  Rochel Gelman,et al.  Enabling constraints for cognitive development and learning: Domain specificity and epigenesis. , 1998 .

[62]  James G. Greeno,et al.  Number sense as situated knowing in a conceptual domain , 1990 .

[63]  Demis Hassabis,et al.  Mastering the game of Go with deep neural networks and tree search , 2016, Nature.

[64]  C. Mervis,et al.  Patterns of object–name extension in production , 1989, Journal of Child Language.

[65]  P. Gordon Numerical Cognition Without Words: Evidence from Amazonia , 2004, Science.

[66]  Geetha B. Ramani,et al.  Taking it to the classroom: Number board games as a small group learning activity , 2012 .

[67]  Colin M. Macleod,et al.  Training and Stroop-like interference: evidence for a continuum of automaticity. , 1988, Journal of experimental psychology. Learning, memory, and cognition.

[68]  H. Margolis Paradigms and Barriers , 1990, PSA: Proceedings of the Biennial Meeting of the Philosophy of Science Association.

[69]  W. A. Martin,et al.  The MACSYMA system , 1971, SYMSAC '71.

[70]  Charles Kemp,et al.  How to Grow a Mind: Statistics, Structure, and Abstraction , 2011, Science.

[71]  James L. McClelland,et al.  Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations , 1986 .

[72]  E. Spelke,et al.  Large number discrimination in 6-month-old infants , 2000, Cognition.

[73]  Alex Graves,et al.  Recurrent Models of Visual Attention , 2014, NIPS.

[74]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[75]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[76]  C. Gallistel,et al.  The generative basis of natural number concepts , 2008, Trends in Cognitive Sciences.

[77]  Andrea Facoetti,et al.  Developmental trajectory of number acuity reveals a severe impairment in developmental dyscalculia , 2010, Cognition.

[78]  Korbinian Moeller,et al.  Unbounding the mental number line—new evidence on children's spatial representation of numbers , 2014, Front. Psychol..

[79]  Roger N. Shepard,et al.  The Step to Rationality: The Efficacy of Thought Experiments in Science, Ethics, and Free Will , 2008, Cogn. Sci..

[80]  Elizabeth M. Brannon,et al.  Evolutionary Foundations of the Approximate Number System , 2011 .

[81]  Han L. J. van der Maas,et al.  Evidence for the Phase Transition from Rule I to Rule II on the Balance Scale Task , 2001 .

[82]  S. E. Mann,et al.  Number Words and Number Symbols: A Cultural History of Numbers. , 1970 .

[83]  Andrea Paula Goldin,et al.  From Ancient Greece to Modern Education: Universality and Lack of Generalization of the Socratic Dialogue , 2011 .