Cognitive Psychology Lecture Notes

Flashcards generated from lecture notes to help review and prepare for the exam.

What is proactive interference?

Old learning interferes with new learning.

What is retroactive interference?

New learning interferes with old learning.

How can you stop proactive interference?

By changing the category being interfered with, as STM is organized around concepts stored in LTM (by category).

What was the Loftus & Palmer (1974) experiment about?

Participants watched an accident and different expressions were used to describe the contact (hit, contact, smash); participants estimated different speeds of the car, and those in the 'smashed' group were more likely to say they saw broken glass.

What individual differences make people at higher risk of misinformation?

Poor general memory, high scores on imagery vividness, and high empathy scores.

What is mental imagery?

Mental representation sharing certain properties with pictures, preserving metric spatial information, changing with viewpoint, explicitly representing empty space, and experienced using spatial attention.

What is the modality-specific or analog viewpoint of mental imagery?

Visual mental images are analogous to pictures in the head.

What is the amodal or propositional viewpoint of mental imagery?

Mental images are not pictures, but rather non-pictorial abstract concepts.

What are the characteristics of language (Arbitrary, Structured, Generative, Dynamic)?

Sounds have no inherent connection with word meaning, organized within words and sentences, can create new utterances, and changes over time.

What are phonemes?

The most basic sounds in language.

What are morphemes?

The smallest unit of meaning in language.

What is syntax?

Rules for putting together words in a sentence.

What is semantics?

Meaning of words/sentences.

What is the critical period for language learning?

Period where language develops readily, and after which language acquisition is more difficult and less successful (birth to 5 years).

What is Noam Chomsky's Universal Grammar theory?

Children can acquire natural language as their native language through an innate blueprint we are all born with.

What is transitional probability in statistical learning?

Likelihood of one syllable following another, used in visual and auditory perception, and used by non-humans.

What is the Sapir-Whorf Hypothesis?

Language determines or influences one’s thought.

What is linguistic determinism (strong interpretation)?

Thoughts/behavior is determined by language.

What is linguistic relativism (milder interpretation)?

Thoughts/behavior is influenced by language.

Who is an expert?

Someone who can reliably demonstrate superior performance on tasks relevant in that domain.

What are the differences between expert and novice performance?

Experts create larger chunks of information, have abstract representations, and have perceptual differences.

What is perceptual learning?

Increase in the ability to extract information from the environment as a result of experience and practice.

What is deductive reasoning?

Draw conclusions that must hold given a set of facts (premises), starting with a general rule applied to a specific case.

What is inductive reasoning?

Conclusions are plausible given the premises, starting with specific examples and forms a general, broader rule.

What is belief bias (associated with deductive reasoning)?

Tendency to accept invalid conclusions if they are believable and reject valid conclusions if they are unbelievable.

What is bounded rationality?

Reasoning is bounded by the available information, cognitive limitations, and time.

What is confirmation bias?

We are typically more responsive to evidence that confirms our beliefs.

What is abduction?

Starts with an observation and tries to find the best possible explanation.

What are the components of a problem?

Initial state, goal state, operators, and path constraints.

What are Well defined problems?

All four components are well specified.

What are Ill-defined problems?

One or more of the components aren’t specified.

What is the Trial and error problem solving strategy?

Trying various possible solutions at random until one works.

What is the Algorithms problem solving strategy?

Logical, step-by-step procedure of trying solutions until one works.

What is the Heuristics problem solving strategy?

Mental shortcuts that allow us to find a solution more quickly.

What is Means-end analysis?

Divide problem into parts and direct search to each subgoal, making progress towards the goal with each step and reducing the difference between current state and goal state.

What is the availability heuristic?

Judging the probability of an event by how easily it can be brought to mind.

What is the representativeness heuristic?

Making the assumption that what is true of one instance of a category must be true of the whole category, judging probabilities based on the degree of similarity to representativeness of stereotypes of knowledge.

What is satisficing?

Choosing the first option that meets your criteria, even if it's not the best possible option.

What is maximizing?

Searching for the absolute best option out of all possibilities, but it takes more time and mental energy.

What are insight problems?

Involves a sudden change in the mental representation of the problem space rather than a search of paths within one problem space.

What is fixation?

Inability to see a problem from a fresh perspective.

What is mental set?

Tendency to approach a problem in one particular way, often one that has been successful in the past.

What is incubation?

Overcoming fixation by a period of not thinking about the problem for a while.

What is chunking (improve problem solving)?

States represented by meaningful arrangements rather than individual pieces.

What is composition (improve problem solving)?

Several separate steps in a process can be consolidated into a single step.

What is analogical problem solving?

Using analogies from similar problems to solve more challenging problems.

How does Availability affect Judgement?

Judged probability of an event is related to how easily that event can be brought to mind.

How does Representativeness affect Judgement?

People judge probabilities based on the degree that a situation is representative of their knowledge/stereotypes.

What is the conjunction fallacy?

When A and B occurring can't be more likely than just A alone.

What is Gambler's Fallacy?

After a long run on red, people expect that black is due to happen since it hasn’t in a while.

What is Anchoring & adjustment?

When making an estimate, people start with a figure and adjust this figure up or down to obtain their estimate, but adjustment usually isn't enough, so estimate is biased toward it.

What is confirmation bias (judgement)?

A tendency for people to look for evidence that confirms their hypothesis.

What is Illusory correlation?

People tend to see correlations that are consistent with their hypotheses.

What are Risk attitudes as defined by Prospect theory?

Risk-seeking with losses and risk-averse with gains.

What is Loss aversion as defined by Prospect Theory?

Losses hurt more than gains feel good.

What is the sunk cost fallacy?

We tend to stick with those things we've already put a lot of time, energy, or money into.

What are Framing effects?

Changing how an issue is presented can change people’s decisions.

What is Additive Difference (multi-dimensional decisions)?

List all the factors, weigh the factors by importance, give a rating for each factor, then add up the totals where the highest point total wins.

What is Elimination by aspect (multi-dimensional decisions)?

Choose most silent dimension, check to see which options have that dimension, last one standing wins.

What is satisfice (multi-dimensional decisions)?

Minimal criteria, randomly pic an option and see if it satisfied the minimum, first one to satisfy wins.

What are Recognition primed decisions?

Try and remember a previous situation that was similar to the current one and choose what you chose then.

What is Bayesian Reasoning?

Existing model of the world that can update based on the evidence currently available to us.

Break down Bayes Rule

Posterior = updated belief, Prior = likelihood that a particular situation occurs at all, Likelihood = how likely the evidence is given the hypothesis, Evidence = Overall probability of seeing the evidence, Posterior= Likelihood×Prior / Evidence.

Where does the Brain vs computer metaphor fall apart?

Brains are slow, good at learning, hard to break, serial vs parallel processing, Local vs distributed representations.

How do local representations work as a part of the brain as computer metaphor?

Code a concept with one node (0 or 1 that represents an idea).

How do distributed representations work as a part of the brain as computer metaphor?

Code information by a pattern of activation across multiple nodes, with features distributed among various nodes, so a concept like grandma or giraffe is distributed among them.

What are Network Science Characteristics?

Centrality—to what extent does your network have certain important nodes, Hierarchy—organization in layers, Ordered vs random networks, Egalitarian vs aristocratic networks, Small-World Networks.

What are Semantic Networks?

Nodes are concepts and links are relation ships between.

What are Layers (neural network)?

Neurons are arranged into layers grouped at same level.

What are Links (neural network)?

Connection between two nodes, passing signal, like synaptic connection.

What is Activation (neural network)?

Output of a node after it processes input, like a neuron firing.

What is Weight (neural network)?

Connection strength of the link, where some nodes weigh more, have more influence in activation, which factor into activation.

What are Neural networks good at capturing about human behavior?

Biological plausibility is not really because human neurons exhibit hebbian learning , where link between neurons increases in strength when fired together, and decrease when not and neurons that fire together, wire together therefore, means learning is implemented at the level of the synapse.

What are Distributed representations (Neural networks)?

Information represented across nodes with a particular activation pattern.

What is Graceful degradation (Neural networks)?

You can take a couple nodes and it will function all right, just if you lose a neuron IRL.

What is the Stability-plactivity Dilemma (Neural networks)?

How to learn new info while retaining old info, where the system can fall apart when new info is learned.

Structural vs functional definitions of the mind

Structural kind= medium dependent - what is it made of and Functional kind = medium independent - what does it DO.

What is AI?

Science of getting computers to behave in an intelligent manner.

What are Turing machines?

Machine that can compute anything that it is logically possible to calculate, where it Capture computing, has Four components , machine table, machine state, Tape, read/write head.

What is Turning Test/Imitation game?

Judge is connected to a computer in one room and a human in the other, if the judge cant tell which one is the computer and which one is human, then the computer is intelligent.

What is the Chinese Room Thought Experiment?

If you know all the rules of inputs and outputs but never understand chinese the way humans do.

What is Weak AI?

Idea that computers can be useful to simulate mental processes.

What is strong AI?

Idea that an appropriately programmed computer really IS a mind is a mind it understands the world and that the mind is really just a program.

What is the Rodney Brooks approach to emergent AI?

Intelligence is an emergent property of complex systems and that intelligence grows naturally out of basic processes.

SIGNIFICANCE of Shepard & Metzler: Mental Rotation of 3D Objects?

For every extra degree of rotation needed, it took a consistently longer amount of time to respond.

What is Linguistic Universality Theory?

Pushback against universal language theory, some languages dont follow these rigid, universal blueprints/guidelines,Lack of empirical, testable support,How would a child link universal guidelines to a specific language.

What are the factors that contribute to Expertise?

Repeating task leads to automaticity, encode information more deeply and meaningfully than novices, use Chunking, retrieval practice, & pattern recognition,Selective attention,Expertise takes time, Knowledge is domain-specific: doesn't always transfer to others, Feedback and motivation are crucial for developing expertise.