Section 10.1
• What is the default hypothesis about linguistic understanding? State some of the different ways of thinking about it?
Section 10.2
• What is the language of thought?
• How does Fodor argue from the rule-based conception of language learning to the language of thought hypothesis?
• How convincing do you find Fodor’s argument?
Section 10.3
• What do proponents of neural network models of language acquisition think that we can learn from them? Do you agree?
• What do you see as the principal strengths and weakness of neural network models?
Section 10.4
• What is the poverty of the stimulus argument?
• What are the basic ideas behind Bayesian approaches to language acquisition?
• What do you see as the principal strengths and weaknesses of Bayesian approaches?
The birth of a word (TED talk by Deb Roy)
The linguistic genius of babies (a TED talk by Patricia Kuhl)
Bayesian Methods for Unsupervised Language Learning (talk by Sharon Goldwater from YouTube Microsoft Research channel)
10.1 Language and Rules
Language (entry from Scholarpedia)
The acquisition of language by children (paper by Saffran, Senghas, and Trueswell, 2001, in Proceedings of the National Academy of Sciences)
10.2 Language Learning and the Language of Thought: Fodor’s Argument
The language of thought hypothesis (entry from Stanford Encyclopedia of Philosophy)
Innateness and Language (entry from Stanford Encyclopedia of Philosophy)
Innateness and contemporary theories of cognition (entry from Stanford Encyclopedia of Philosophy)
Poverty of the Stimulus Revisited (paper by Berwick et al., 2011, in Cognitive Science)
The poverty of the stimulus argument (paper by Laurence and Margolis, 2001, in British Journal for the Philosophy
of Science)
Poverty of Stimulus Arguments and Behaviourism (paper by King, 2015, in Behavior and Philosophy)
10.3 Language Learning in Neural Networks
CHILDES (Child Language Data Exchange System; downloadable transcripts of children’s language-use)
Neural net language models (entry from Scholarpedia)
Overregularization in language acquisition (paper by Marcus et al., 1992, in Monographs of the society for research in child development)
On language and connectionism: Analysis of a parallel distributed processing model of language acquisition (paper by Pinker and Prince, 1988, in Cognition)
Overregulatization in English plural and past tense infelectional morphology: a response to Marcus (1995) (paper by Marchman, Plunkett, and Goodman, 1997, in Journal of Child Language)
Rules or connections in past-tense inflections: What does the evidence rule out? (paper by McClelland and Patterson, 2002, in Trends in Cognitive Sciences)
The past and future of the past tense (paper by Pinker and Ullman, 2002, in Trends in Cognitive Sciences)
‘Words orRules’cannot exploit theregularity inexceptions (paper by McClelland and Patterson, 2002, in Trends in Cognitive Sciences)
Letting Structure Emerge: Connectionist and Dynamical Systems Approaches to Understanding Cognition (paper by McClelland et al., 2010, in Trends in Cognitive Sciences)
Gold’s theorem and cognitive science (paper by Johnson, 2004, in Philosophy of Science)
Language acquisition in the absence of explicit negative evidence: how important is starting small? (paper by Rohde and Plaut, 1999, in Cognition)
10.4 Bayesian Language Learning
English-learning Infants’ Segmentation of Verbs from Fluent Speech (paper by Nazzi et al., 2005, in Language and Speech)
Statistical Learning of Syntax: The Role of Transitional Probability (paper by Thompson and Newport, 2007, in Language learning and development)
The Bayesian Reader: Explaining word recognition as an optimal Bayesian decision process (paper by Norris, 2006, in Psychological Review)
Statistical Learning, Inductive Bias, and Bayesian Inference in Language Acquisition (paper by Pearl and Goldwater, 2016, in The Oxford Handbook of Developmental Linguistics)
Probabilistic models of cognition: Conceptual foundations (by Chater, Tenenbaum and Yuille, 2006, in Trends in Cognitive Sciences)
Statistical Learning by 8-Month-Old Infants (paper by Saffran, Aslin, and Newport, 1996, in Science)
Indirect Evidence and the Poverty of the Stimulus: The Case of Anaphoric One (paper by Foraker et al., 2009, in Cognitive Science)
Word learning as Bayesian inference (by Xu and Tenenbaum, 2007, in Psychological Review)
Human-level concept learning through probabilistic program induction (paper by Lake, Salakhutdinov, and Tenenbaum, 2016, in Science)
