CGSC Exam 3
Infants knowledge of objects
-
organization of cognition
-
Spelke's principles: solidity and continuity constraints
- every object moves on a single trajectory through space and time - and it is impossible for these paths to intersect in a way that would allow more than one object to be in one place at a time - drawbridge experiment good example of reasoning according to solidity constraint, since it shows that infants are sensitive to impossibility of there being more than one object in a single place at one time - for continuity violation -- infants know items can't just disappear - for solidity violation -- infants know it is impossible for there to be more than 1 object in the same place at one time
Spelke's principles: principle of contact
- only surfaces in contact with one another move together
Spelke's principles: principle of cohesion
- surface belong to a single individual if and only if they are in contact - evidence: infants do not appear to perceive boundary between two objects that are stationary and adjacent, even when differ in shape and texture - infants show more surprise when object comes apart, even if object doesn't have the Gestalt properties of homogeneous color and figural simplicity
organization of the mid- 2 general approaches
-Horizontal organization= mind organized in terms of general cognitive abilities/capacities Perception Memory Attention -Vertical organization= mind organized in specialized peripheral systems + central systems in the middle Perceptual systems Central processing Motor systems
What are some of the problems with the 1978 hexagonal model that was proposed?
-Identifies connections between different cognitive sciences, but sheds no light on why there need to be such connections -Doesn't help us to understand what each discipline contributes -Doesn't give a very accurate picture of contemporary cognitive science
What principles govern infants' knowledge of objects?
1)principle of cohesion 2) principle of contact 3) solidity constraints 4) continuity constraints
evidence for massive modularity hypothesis
Arguments from specialized deficits to specialized modules General arguments against the idea of domain-general central processing
puzzle of altruistic behavior
Cooperative behavior presumably has a genetic basis But how did the genes coding for cooperative behavior ever get established in the gene pool? natural selection seems to favor "selfish"behavior -free riders can always exploit altruists
characteristics of Fodor's modules.
Dedicated processing systems that are 1. domain-specific -modules are specialized for specific tasks 2. informationally encapsulated -cognitive faculties can only call upon a very limited range of information to carry out specific tasks 3. cognitively impenetrable -modular processing remains unaffected by what is going on elsewhere in the mind 4. fast-transform inputs to outputs quickly and efficiently 5. mandatory-cognitive modules respond automatically to appropriate stimuli, rather than being under executive control 6. Possibly have fixed neural architecture -regions of the brain associated with particular types of modular processing (e.g. fusiform face are) 7. specific breakdown patterns -modular processing can fail in highly determinate ways
Argument from Statistics
Domain-general systems only have access to what can be inferred from perceptual processes by general cognitive mechanisms However, the world has "statistically recurrent domain-specific structure" - certain features hold with great regularity in some domains, but not in others
How does TIT-FOR-TAT explain the evolution of altruism/cooperation
Evolutionary pressure for selection of cheater detection module According to Cosmides and Tooby, this module explains the pattern of choices made in conditional reasoning tasks
darwinian modules vs fodor modules
Fodor Modules: -domain-specific -informationally encapsulated -mandatory -fast -neurally constructed -specific breakdown patterns only the first fodor module applies to darwinian modules. None of the rest apply completely Relative helping example: - not informationally encapsulated, because other things may come into play other than the calculations of a Darwinian Kin Selection module, not mandatory b/c may not be applied every time, etc
Fodor on central (non-modular) processing
Fodor suggests that central processing has two distinguishing features- Quinean and Isotropic ***Central processing has to be non-modular*** - Quinean-holistic view of knowledge - Isotropic-not informationally encapsulated "The more global (more isotropic) a cognitive process is, the less anybody understands it. Very global processes, like analogical reasoning, aren't understood at all". He argues that cognitive science is really best suited to understand modular processes. It can tell us very little about central processes.
Fodor's objection to the Massive Modularity Hypothesis
Fordorean modules are domain-specific, informationally encapsulated, mandatory, and fast (4 characteristics) may also have fixed neural architecture, and specific patterns of breakdown. Massive modularity theorists think all information processing is essentially modular. They understand modules in a much less strict way than Fodor. "No such thing as central processing, as fodor discusses."
massive modularity hypothesis
Gives a picture of the overall organization of the mind -each module in our mind is evolved to solve a particular adaptive problem -each module exploits specialized rules that are DOMAIN-SPECIFIC
How do the predictions of vertical and horizontal approaches to cognition differ in terms of the relationship between perception and cognition?
Horizontal: No real boundary between perceptual processes and central cognition. Vertical: Fundamental differences between perceptual systems and central processing
Affirming the consequent example
If it is a car, then it has wheels. It has wheels. Therefore, it is a car. (incorrect)
Denying the antecedent example
If it is a car, then it has wheels. It is not a car. Therefore, it does not have wheels. (incorrect)
Modus Tollens Example
If it rains tonight (a), the baseball team will lose (b). It does NOT rain tonight (not a). Therefore, they will NOT lose. (not b)
Modus Ponens Example
If it rains tonight (a), the baseball team will lose. (b) It rains tonight. (yes a) therefore They will lose. (yes b)
Where did Fodor's idea for the modularity of the mind come from?
In his book he defends the idea that the mind contains autonomous cognitive sub-systems. (Tries to defend neuroanatomist Franz Joseph Gall - who tried to pin specific mental functions down to particular locations in the brain) He argues that Gall was basically correct to think the mind was made up of semi-autonomous cognitive facilities.
interpretations of this
Infants "knew" the box was there and that the drawbridge could not pass through it Object permanence (objects exist even if not perceived)
Spelke's 1993 experiment.
Infants can parse the visual array into spatially extended and bounded individuals that behave according to certain basic principles of physical reasoning Principle of cohesion -surfaces belong to a single individual if an only if they are in contact Infants do not appear to perceive the boundary between two objects that are stationary and adjacent even when the objects differ in color, shape, and texture
infants vs adults
Infants place more importance on spatiotemporal continuity than on featural continuity (movement is more important than whether object looks the same) Infants less surprised if an item disappears behind a screen and another completely different object reappears on the other side Adults less surprised if the same object re-appears in an impossible location (magic trick)
What is phrenology?
It essentially assigned different functions to cortical areas. Scientists who studied phrenology believed that "bumps" on the skull signified an enlarged cortical area.
example of a darwinian modules
Massive modularity theorists hold that the human mind is a collection of specialized modules, each of which evolved to solve a very specific set of problems that were confronted by our early ancestors - Hunters & Gatherers. Cheater detection • Folk psychology/mindreading • Kin selection • Intuitive physics Number sense
4 conditioning rules
Modus ponens: If p then q. p. Therefore q Modus tollens: If p then q. Not-q. Therefore not-p Affirming the consequent: If p then q. q. Therefore p Denying antecedent. If p then q. Not-p. Therefore not-q
Wason Selection Task
Most famous task of psychology of reasoning. Ability to represent cost and reasoning. If there is a vowel on one side there has to be an odd # on the other side. Much easier if placed in the context of a social rule.
Franz Joseph Gall
Neuroanatomist that tried to pin specific mental functions down to particular locations in the brain • Character traits can be read of shape of the skull --not correct • Specialization for performing particular cognitive tasks -yes!
Argument from Error
Organisms tend to learn by getting things wrong. Learning required feedback, and negative feedback is often easier to come by than positive feedback. "Domain-general cognitive mechanisms could not have been selected by natural selection because they would have made too many mistakes. This is why they say there must be a distinct cognitive mechanism for every domain that has a different definition of what counts as a successful outcome."
Baillargeon vs Piaget
Piaget -object permanence at 8-9 months Baillargeon-object permanence at 4.5 months
cassava root studies (Cosmides and Tooby)
Social Exchange Two versions of cassava root story Descriptive: married men live on the side of the island where cassava roots grow, while unmarried men live on the side where the molo nuts grow Social exchange: only married men have the right to eat cassava roots
arguments for massive modularity in false belief tasks
Some neuropsychological deficits are thought to provide evidence for Fodor modules autism as evidence for a dedicated theory of mind module
4 stages of balance beam problem
Stage 1. Number of weights Stage 2. Distance from fulcrum Stage 3. Both weight and distance are used, but not if both vary at the same time. Stage 4. Weight and distance used properly and accurately (achieved usually in adolescence)
switched selection task
Standard social exchange selection task • If BENEFIT (p) then COST (q) • violation = p and not-q Switched social exchange selection task • If COST (p) then BENEFIT (q) • violation = q and not-p • Subjects typically give the logically correct answer on the standard version, but not on the switched version
What are three areas of child cognitive development that can be modeled by neural networks
Tense learning, object permanence, and balance beam problem
Cosmides and Tooby in massive modularity hypothesis
The Cosmides/Tooby experiments seem to show specialized skills for cheater detection not simply specialized skills for conditional reasoning involving social exchanges These experimental results are integrated with the massive modularity hypothesis via an evolutionary explanation of why there needs to be a cheater detection module: EVOLUTION OF ALTRUISM (cooperative behavior)
What is the integration challenge?
The challenge of providing a unified account of cognition that draws upon and integrates the whole space. Many regions within the "space" of cognitive science remain little studied & the "space" is not organized by discipline 3 approaches to this challenge: - Local Integrations - Global Models of integration - The mental architectures approach
at what age do children pass the false belief tasks
Young children typically pass at 4 years Failed by 80% of a sample of autistic children aged between 6 and 16, with mean verbal and nonverbal mental ages > 5
Prisoner's Dilemma
a particular "game" between two captured prisoners that illustrates why cooperation is difficult to maintain even when it is mutually beneficial
Habituation in experiments with infants
allow us to measure what infants know about the world Violation of expectation: infants will show a greater degree of surprise (as measured by looking time) when an unexpected event occurs •Habituating infants to specific events provides a baseline level of expectation
Which two(-ish) disciplines are being integrated in the first example of local integration (with the Wason selection task and Cassava root studies)
cheater detection module and cooperative behavior ?
Balance Beam Problem
children go through stages to tackle the problem of whether the beam will balance or not
examples of possible modules
color perception shape analysis face recognition grammatical analysis of heard utterances recognizing voices
Baillargeon 1987 test with infants
dishabituation: infants look longer at objects they find surprising drawbridge experiments involve habituating infants to a rotation of an object, then presenting them with an impossible 180 degree rotation illusion
what modification improves human performance?
framing the question in such a way that what is being checked is a condition that has to do with permissions and entitlements
Describe habituation
learned behavior that allows the animal to disregard meaningless stimuli
Mcclelland and Jenkins
modeled balance beam problem with a neural network
Munakata 1997 Model
modeling object permanence designed a network to indicate the ball when it is put of the field of view
Solidity and Continuity constraints on movement
objects move in a certain way and it would be impossible for them to intersect trajectories
What is the best strategy for one-shot prisoner's dilemma or an iterated prisoner's dilemma with a set number of repetitions?
one shot: best strategy is to not cooperate iterated: best strategy is tit for tat, because the other player can retaliate
switched form
results in incorrect answers
example of local integration
specific cases where cognitive scientists have built bridges across levels of explanation and between disciplines Local Integration 1: Inspired by experiments in the psychology of reasoning showing that people are systematically prone to very specific types of fallacies in deductive reasoning • Evolutionary psychologists suggest that these patterns of error are by-products of highly specialized cognitive mechanisms that emerged to solve adaptive problems • massive modularity thesis • cheater detection module
how do people perform when 1: there isnt a social context
subjects fail and pick the affirming the consequent
2: there is a social context with cheating
subjects seek out the cheater - tell when someone has taken a benefit without paying the price
3: there is a social context without cheating
success comparable to the test without context
Synchronous Movement
the objects move together, infants assume they must be one connected object
Do Cosmides and Tooby's Cassava Root studies support domain-general or massive modularity theories of cognition?
they support domain-specific theories, They gave two arguments for thinking that there is nothing more to the mind than a collection of specialized sub-systems: The argument from error, & The argument from statistics and learning Basic assumption is that the human mind is a product of evolution, and it works by natural selection. Any mental architecture that we have today must have evolved because it was able to solve the adaptive problems that our ancestors encountered.
false belief task
type of task used in theory of mind studies in which children must infer that another person does not possess knowledge that they possess. For example, children shown that a candy box contains pennies rather than candy are asked what someone else would expect to find in the box.
how does tit for tat show up in the real world?
voting patterns in senate
according to wason, why do people fail this task?
when the problem is not put into the right context, subjects fail. Explained by cheater detection module: when the selection task is framed in terms of permissions and entitlements it engages the cheater detection module
