Origins of Mind (Milan):
Philosophical Issues in Cognitive Development
--- by [email protected]
A course at Università degli Studi di Milano about how humans come to know about objects, causes, words, numbers, colours, actions and minds.
Slides and Handouts
You can find slides and handout below, together with an outline of each lecture.
Please note that these may be revised.
Lecture 01
Date given: Tuesday 23rd September 2014
The Question (001)
Introduces the question around which this module is organised
From Myths to Mechanisms (021)
A quick, scene-setting discussion of nativism and empiricism.
Inbetween mindless behaviour and thought (031_short)
Why does the question about the origins of knowledge require philosophical inquiry?
Two Breakthroughs (051)
Two scientific breakthroughs that have recently furthered understanding of how knowledge might emerge in development. The first breakthrough is the discovery that preverbal infants enjoy surprisingly rich social abilities, abilities which may well be foundational for later linguistic abilities and enable the emergence of knowledge (e.g. Csibra & Gergely 2009; Meltzoff 2007; Tomasello et al. 2005). A second breakthrough is the use of increasingly sensitive—-and sometimes controversial--methods to detect sophisticated expectations concerning causal interactions, numerosity, mental states and more besides in preverbal infants (e.g. Spelke 1990; Baillargeon et al. 2010).
Lecture 02
Date given: Tuesday 23rd September 2014
Objects vs Features (161)
Knowledge of objects requires abilities to (i) segment objects, (ii) represent them as persisting and (iii) track their interactions. What this involves can be illustrated by contrasting objects with features.
Segmentation and the Principles of Object Perception (171)
Humans, adult and infant, segment objects in accordance with four principles: cohension, boundedness, rigidity and no action at a distance (Spelke 1990).
Permanence (181)
There is evidence that abilities to represent unperceived objects as persisting appear early in infancy, from four months of age or earlier. It appears that a single set of principles might explain both abilities to represent objects as persisting and abilities to segment objects.
Causal Interactions (201)
Knowledge of objects requires being able to track their causal interactions. How do humans achieve this?
Recap and Questions (206)
A brief recap of what we've learnt so far about infant representations of objects and their causal interactions.
A Problem (207)
Some principles describe infants' abilities to segment and represent persisting physical objects and their causal interactions. What is the relation between these principles and the mechanisms of object perception? Are they, for example, things that infants know?
Like Knowledge and Like Not Knowledge (211)
There are principles of object perception that explain abilities to segment objects, to represent them while temporarily unperceived and to track their interactions. These principles are not known. What is their status?
Lecture 03
Date given: Tuesday 23rd September 2014
Perception of Causation (251)
Can humans perceive causal interactions?
Object Indexes and Causal Interactions (261)
To understand the launching effect we need to consider how perceptual systems segment and track objects.
Object Indexes and the Principles of Object Perception (266)
The Principles of Object Perception are not knowledge, nor do they feature in inferences directly yeilding knowledge. Rather they characterise the operation of those perceptual systems whose job is to segment and track physical objects.
Perceptual Expectations (271)
This is an introduction to the notion of a perceptual expectation. We draw on studies of representational momentum to show that some perceptual expectations concern the ways forces act on objects.
Lecture 04
Date given: Tuesday 23rd September 2014
Knowledge of Colour: a Question (110)
How does knowledge that these things are blue (say) but that is not emerge?
Categorical Perception of Colour (111)
Categorical perception is a pervasive and useful feature of human experience much neglected by philosophers despite illuminating psychological research on this topic. Here we focus on colour. The aim is initially just to understand what categorical perception is.
Categorical Perception in Infancy (121)
Categorical perception of colour emerges early in infancy, from around four months or earlier, as has been demonstrated using habituation (Bornstein, Kessen and Weiskopf 1976) and visual search tasks (Franklin, Pilling and Davies 2005).
Categorical Perception and Knowledge (131)
How is infants' categorical preception of colour related to adults' knowledge of colour?
Core Knowledge (601_milan)
Why do we need a notion of core knowledge? What is core knowledge and what is its relation to knowledge?
Appendix: Categorical Perception in Infants and Adults (Optional) (141)
How are infants' and adults' categorical preception of colour related?
Lecture 05
Date given: Wednesday 24th September 2014
Communication with Words: A Question (501)
How do humans first come to communicate with words? An utterance is an action with a complex structure. Recovering this structure requires several transitions: from continous sounds and bodily movements to phonetic gestures; from phonetic gestures to words; and from words to clauses and sentences. We shall ignore most of these transitions and focus on one word utterances.
Preview: Shipwreck Survivor vs Lab Rat (506)
There are two main views about how humans first come to communicate with words. Both are wrong.
Does being able to think depend on being able to communicate with language? (511)
Some philosophers including Donald Davidson claims that if someone can think, she can communicate by language. What are the consequences of this claim for an account of how humans come to be able to communicate with language?
Training (521)
Many philosophers of language (Russell, Wittgenstein, Quine, Dummett and Davidson among them) hold that humans are able to communicate with words because they are trained in the use of words.
Understanding (531)
One potential problem for the view that children learn to communicate with words through training in when to utter them does not appear sufficient for understanding.
Creativity (541)
Some children acquire language without training, and any significant linguistic experience. And children in mundane linguistic environments use words creatively from the outset. What does this tell us about the view that humans come to use words through being trained in their use?
Mapping words to concepts (551)
Paul Bloom and others suggest that acquiring a language is in large part a matter of working out which concepts correspond to which words. How plausible is this view?
Summary (561)
Our question was, How do humans first come to communicate with words? We are left with a problem: none of the answers considered plausibly provide anything remotely approaching a complete answer to this question.
Appendix: Grice/Tomasello (optional) (571)
Can we learn anything from Grice's views on meaning about how humans come to acquire abilities to communicate with language? Michael Tomasello has suggested, in effect, that we can.
Lecture 06
Date given: Wednesday 24th September 2014
Pointing (661)
From around 11 or 12 months of age infants spontaneously point to request, inform and initiate joint engagement. Here we examine evidence for this claim. We then ask, by contrasting humans with nonhuman primates, what is involved in producing and comprehending pointing.
A Puzzle about Pointing (663)
Humans point in ways that no other animal does. Why? Also, humans first point to request, inform or initiate joint engagement months after they can produce pointing gestures. Why? (These questions are taken from Tomasello, Carpenter & Liszkowski.)
What is a communicative action? (671)
To understand what follows from infants' abilities to comprehend and produce pointing gestures, we need to ask what a communicative action is. One view hinges on shared intentionality, which is related to Grice's attempts to analyse meaning. Here we consider implications of that view and contrast it with two alternatives.
Words and Communicative Actions (681)
Here we pause to connect our studies of non-linguistic communication back to linguistic communication. Reflection on non-linguistic communicative actions (in particular, pointing gestures) makes apparent that understanding a communicative action involves more than identifying which thing it picks out. This may matter for understanding abilities to communicate by language too.
Lecture 07
Date given: Wednesday 24th September 2014
Knowledge of Mind (401)
The challenge is to explain the emergence of awareness of others' mental states; here we focus on awareness of others' beliefs.
Mindreading: First Puzzle (411)
The first puzzle concerns apparently conflicting findings about when and how humans acquire awareness of others' beliefs.
Mindreading: Second Puzzle (421)
The second puzzle concerns apparently conflicting findings about whether ascriptions of belief to others is automatic or not.
Modules and Cognitive Efficiency (431)
We can resolve the two puzzles by conjecturing that there are multiple kinds of mindreading, including one that is modular and another that is not. But to make sense of the conjecture that mindreading sometimes is modular, we need to understand how it could be cognitively efficient.
Minimal Theory of Mind (441)
The construction of a minimal theory of mind enables us to understand how mindreading could be automatic in a limited but useful range of cases.
Signature Limits Generate Predictions (451)
In response to the two puzzles, we have made a conjecture. This puzzle implies that infant and automatic reasoning is modular and involves minimal theory of mind. But minimal theory of mind has signature limits. These allow us to generate predictions to test the conjecture.
Lecture 08
Date given: Thursday 25th September 2014
Action: The Basics (701)
When do human infants first track goal-directed actions and not just movements? A variety of evidence suggests that the answer is, from around three months. Is this competence is related to adults’ abilites to track goal-directed action? Is it a manifestation of cor knowledge?
How Do Infants Model Actions? (706)
We have seen that from around three months of age infants can track the goals of actions, and that their abilities may involve a form of core knowledge. What model of action underpins their abilities to track the goals of actions? And what is a model of action anyway?
Does Infants’ Model of Action Involve Intentions? (711)
What model of action underpins infant cognition of action? Is it a model on which actions are related to goals by intentions, as Premack and others have suggested? Probably not.
Pure Goal Ascription: the Teleological Stance (712)
Pure goal ascription is goal ascription which occurs independently of any knowledge of mental states. Here we consider an account of pure goal ascription involving the teleological stance.
From Action to Communication and Joint Action (741)
How does what we’ve learnt about how humans come to know truths about actions illuminate issues about social interaction and communication?
Lecture 09
Date given: Thursday 25th September 2014
Conclusion and Outlook (999)
What progress have we made in trying to understand how humans come to know about--and to knowingly manipulate--objects, causes, words, numbers, colours, actions and minds?
Appendix: Theoretical Background
Date given: Thursday 25th September 2014
Core Knowledge and Modularity (231)
What Is Core Knowledge? How Is It Related to Modularity
Core Knowledge (601)
Why do we need a notion of core knowledge? What is core knowledge and what is its relation to knowledge?
Syntax / Innateness (641)
Infants' abilties to detect syntactic structure are surprisingly sophisticated (relative to their performance in producing utterances). Here we consider a case study. We also use this case study to examine how poverty of stimulus arguments work. Some argue that infants have innate abilties to learn about syntax.
Syntax: Knowledge or Core Knowledge? (646)
Given that humans, infant and adult, do represent facts concerning the syntax of languages, what grounds are there to deny that these representations are knowledge?
Computation is the Real Essence of Core Knowledge (291)
What is core knowledge (= modular representation)? We have seen (§231) that we cannot characterise an expalantory notion of core knowledge merely by listing features. A better way of characterising core knowledge requires idenityfing the kind of process in which core knowledge occurs.