Learning and Cognition Flashcards
Cognitive
- process of learning and using knowledge
EXAMPLES:
- Perception: Interpreting sensory information to understand the environment.
- Memory: Storing and retrieving information for future use.
- Reasoning and Decision-Making: Analyzing information to make judgments or select actions.
Learning Definition 3
1) Set of biological, cognitive and social processes
2) Through which organisms make meaning from their experiences
3) Producing long lasting changes in their behaviour, abilities and knowledge
EG.
1) PROCESSES
2) ANALYSE MISTAKES (Make meaning from their experiences)
3) CHANGE FUTURE BEHAVIOUR (changes in their behaviour, knowledge, abilities)
Why is learning important?
- Predict the future from our past experiences
- Help with survival - PAVLOV
Mental Representation, function
WHAT IS IT?
- Having a picture, idea, emotion, in your mind, response (eg. salivation) about something
- Even when it’s not right in front of you!
- Typically triggered by a stimulus (eg. owner getting up and walking toward the kitchen)
FUNCTION?
Help you make decisions, solve problems
Dog’s mental representation vs ours // what is his mental representation of the food?
1) ours more complex
-> think of type of treat, nutritional value, planning to buy more treats later
-> dogs more focused on the present moment
2)
- emotionally excited/happy
- image of food
- salivating
2 forms of learning
1) Non-associative learning
- ie. AUTOMATIC
2) Associative learning
- ie. has to be LEARNT
Examples of non-associative/associative learning 2
ASSOCIATIVE
- Classical conditioning
NON-ASSOCIATIVE
- Sensitisation
- Habituation
Sensitisation - example type of stimuli, response
EXAMPLE
= sudden loud bang in the lecture room
STIMULI
= Potentially THREATENING stimuli in the environment
= Sudden, surprising
RESPONSE
- TEMPORARY state of heightened attention and responsibility
- Remain alert
- Increased response to subsequent stimuli (ie. another bang would cause even more response)
Habituation
EXAMPLE = constant rumbling of air conditioning in the back of a lecture room
STIMULI
= Persists, doesn’t go away
RESPONSE
- Gradual diminishing of attention to the stimuli
Classical conditioning - definition 3
- Learning a predictive relationship
- (CONDITIONING) = Between a NS (neutral stimulus) and an UCS and its unconditioned response
- (AFTER CONDITIONING) = so that the NS becomes a CS (conditioned stimulus) that can cause a CR (conditioned reflex)
Classic conditioning - GENERAL UNDERSTANDING/DEFINITIONS OF KEY TERMS
**conditioned = learn
1) Neutral stimulus = no response
2) Unconditioned stimulus = unconditioned response (autonomic, reflex)
eg. salivation
3) Conditioned stimulus = conditioned response
eg. bell - except this time it PREDICTS food
NS, UCS, UCR, CS, CR - what does it stand for?
- NS = neutral stimulus
- UCS = unconditioned stimulus
- UCR = unconditioned response
- CS = conditioned stimulus
- CR = conditioned response
Classic conditioning - KEY TERM EXAMPLES (first 2)
1) Neutral stimulus = bell
2) Unconditioned stimulus = food, salivate
Conditioned stimulus, DEFINITION, EXAMPLE
Pairing NS with UCS
SO…
- Stimulus is still bell
- except this time it PREDICTS food to come out
IE. BASICALLY NS, EXACT SAME OBJECT, BUT THIS HAS A TWIST - HAS AN ASSOCIATION/PREDICTION ATTACHED
Conditioned response
LEARNED reaction to the CS
- unconditioned response has become conditioned response!
(note same response eg. salivation, however this time it is learnt, not reflex, due to CS!!!) - NOT the NS as that is the bell, in that context may be jumping up in shock
- So it is specifically bell, salivate!
3 phases of Classical Conditioning - PROCESS
1) BEFORE CONDITIONING
- NS -> no response
- UCS -> UCR
- UCS + UCR = Reflex
2) CONDITIONING
- NS + UCS -> UCR
(need a few trials to repeat this)
3) AFTER CONDITIONING
- CS = CR
Extinction. Definition, How is it done?
- WHAT WE WANT: CR to decrease over time
- Learned INHIBITION of the CS-UCS association
- Eg. play the bell 10x, but do not give food
NOT NS and UCS
-> NS has already transformed to CS through the conditioning process
-> want to disassociate CS (bell that predicts food) with the UCS (the food)
Spontaneous recovery
DEFINITION:
- Sudden, temporary return of the CR
- When there is a long REST period after extinction
- Introduce just the NS (bell)
HOW WOULD IT BE DONE?
- Let’s say extinction process has to be done many times
- 24 hours no trials, relax break
- Now the CS may cause a. brief CR again!
(ie. playing the bell may cause brief salivation)
WHY?
- Learn new idea: CS doesn’t predict UCS bell doesn’t equal food
- But previous association/predictive relationship is not forgotten - MEMORY IS NOT ERASED
HOW TO PREVENT?
- Space out extinction over multiple sessions - gradually prevent spontaneous recovery
Rapid re-acquisition
DEFINITION:
- Faster RELEARNING of the CR than the first time
- When the CS is re-paired with the UCS after extinction
- Reintroduce the bell + food UCS + NS
HOW WOULD IT BE DONE?
eg. after extinction, pair the CS (bell that predicts food) with the UCS (food) 2-3x, quickly learn again
Stimulus generalisation
- Organism responds to stimuli similar to CS
EG. dogs salivate not only to original bell but similar sounds like doorbell, whistle
RMB KEY WORD IS CS - because dogs won’t respond to NS in the first place!!
- its confusing because NS and CS are both bells…
Stimulus discrimination
- When an organism differentiates between the CS and other stimuli - only respond to the CS
EG.
- eg. only salivate to a specific bell tone and not other pitches etc.
Another name for UCS + categories for responses
ANOTHER NAME:
Biologically significant stimuli (ie. relate to survival, autonomic, reflex responses)
RESPONSES:
- Defensive = flight, fight, freeze
- Appetitive = approach, rewarding
- Aversive = punishing
Extinction - how to prevent spontaneous recovery
- Increase distance of the sessions
- Vary context of the sessions
Difference between spontaneous recovery + rapid reacquisition
RR =reintroduction of food + bell
SR = introducing just the bell
What do
spontaneous
recovery and rapid
reacquisition
suggest about the
extinction
process?
- ## Extinction doesn’t erase what is learnt, just previous associative relationship is supressed
Generalisation - how would trials work?
- Plot graph - increment tone to see drop off point where it gets less similar to the original sound
Operant conditioning
- Learning through an action’s consequences (ie. rewards and punishments)
Operant conditioning - reinforcement
Reinforcement = making a behaviour MORE LIKELY TO HAPPEN
- Positive reinforcement = ADDING something pleasant
-> smth = positive reinforcer - Negative reinforcement = TAKING AWAY something unpleasant
-> smth = negative reinforcer
Operant conditioning - punishment
Punishment = making a behaviour LESS LIKELY TO HAPPEN
- Positive punishment = ADDING something unpleasant
-> smth = positive punisher = unpleasant UCS stimulus - Negative punishment = TAKING AWAY something pleasant
-> smth = negative punisher = pleasant UCS stimulus
Reinforcement vs punishment
- rmb reinforcement = INCREASE of behaviour
- punishment = DECREASE of behaviour
-> always true, ignore words like negative and positive
Reinforcement example EG. Skinner had a Skinner box
- Target behaviour we want to INCREASE = pushing a lever
- Positive reinforcement = rat gets food after pushing the lever
- Negative reinforcement = electric shock stops after pushing the lever
Punishment example EG. Skinner had a Skinner box
- Target behaviour we want to DECREASE = pushing a lever
- Positive punishment = receive electric shock after rat pushes the lever
- Negative punishment = rat given no food if lever is pushed (ALSO KNOWN AS RESPONSE COST)
SO…
- electric shock = positive punisher = unpleasant UCS stimulus (wincing is natural response!)
Extinction - for reinforcement
- when we want to STOP the target behaviour we tried to increase
HOW IS IT DONE?
- stop reinforcement!
WHAT HAPPENS WHEN REINFORCEMENT IS STOPPED? 2 stages
1) FIRST TIME = the rat presses the lever even more frequently/AGGRESSIVELY for a short time
2) LATER = there is a GRADUAL DECLINE when reward is no longer presented
SAME FOR PUNISHMENT
Reinforcement schedules - 2
CONTINUOUS = rewarding every time behaviour is done
- Quick learning to do behaviour
- BUT rapid extinction if reinforcements stop
PARTIAL REINFORCEMENT = rewarding only sometimes when the behaviour is done
- Slower, persistent learning to do behaviour
- BUT slower extinction once reinforcements stop
Punishment schedules - PARTIAL PUNISHMENT
SIMILAR TO PARTIAL REINFORCEMENT!!
EXAMPLES = GAMBLING
- sometimes win (reinforcer)
- however hard for extinction
- hope that you could win that day!
Shaping behaviour - definition
- achieving difficult desired behaviour
- by REINFORCING SUCCESSIVE APPROXIMATIONS (small steps)
Shaping behaviour - steps
GETTING DOG TO ROLL OVER
- High Frequency Component
- find a behaviour the dog already does often + AND close to desired behaviour
- In this case, it might be lying down.
- Reinforce the dog for lying down - Drop Reinforcement and Observe Variability
- Once the dog consistently lies down, stop rewarding just for lying down.
- This might confuse the dog, causing it to try different movements while lying down - Reinforce Closer Approximations
- Wait for the dog to make a movement closer to rolling over (eg. legs up)
- When this happens, we immediately reward it - Cycle Through Closer Approximations
- We continue this process, rewarding movements that get progressively closer to a full roll
In what circumstances is punishment only effective? 3
1) CONTINGENCY
- Clear connection between behaviour and punishment
(ie. shouting = no more phone)
2) CONTIGUITY
- Punishment must immediately follow the behaviour
(ie. after shouting, immediately take phone, don’t wait 2 hours later)
3) CONSISTENCY
- Punishment must happen every time the behaviour shows up
(ie. don’t say no phone, then when lazy let the child keep the phone)
Why is punishment not effective? 4
1) Positive punishment doesn’t work for long term behaviour change
- only SUPPRESSES behaviour
- rarely effective due to the 3C’s
2) Doesn’t teach more desirable behaviour
- focuses on avoiding the bad behaviour
3) Produces negative feelings in the learner
- HENCE doesn’t promote new learning
4) Harsh punishment may teach the learner to use the same behaviour with others
(SOCIAL LEARNING)
Alternatives to punishment 3
1) Stop reinforcing problem behaviour on ACCIDENT
- find out reason for bad behaviour
- Eg. whining child -> reason = want to get candy as they know you will give it to them
- do not reinforce by ignoring their whining!
2) Reinforce alternative behaviour that is INCOMPATIBLE with the current behaviour
- Incompatible = bad and good behaviour can’t exist at the same time
- eg. whining + asking politely can’t happen at the same time!
- eg. “thanks for talking politely, here’s your candy”
3) Reinforce the non-occurence of the undesirable behaviour
- eg. reward person for not doing the behaviour in a SET TIME
eg. not whining for 30 mins, give them a candy
Antecedent
- A stimuli that acts as a CUE
- Signals the AVAILABILITY of a REINFORCER
Antecedent - examples, what do you notice?
EXAMPLE 1
- reinforcer (pleasant UCS stimulus) = scrolling on phone (dopamine release)
- antecedent = phone lighting up
EXAMPLE 2
-Reinforcer (pleasant UCS stimulus) = drinking coffee (yummy)
- antecedent = seeing your favourite coffee shop
NOTICE:
- Antecedent-reinforcer relationships are based on a CLASSICALLY CONDITIONED CS-UCS
-> Antecedents were previously NS, but combined with UCS, you predict reinforcer
ABC model of operant conditioning
Antecedent → Behavior → Consequence
How It All Works Together
1. The antecedent (seeing your phone light up) triggers a behavior (checking social media).
2. The behavior (checking social media)
3. leads to a consequence (feeling happy). If the consequence is good, you’re more likely to repeat the behavior when you see the antecedent again.
Discriminant stimuli
- a specific type of ANTECEDENT that signals the availability of reinforcement
- BUT only for a PARTICULAR behaviour!!
eg. animal training - hand up whales will jump + get a reward from one box, hand down whales will stay still + get a reward fro another box
Cognitive map - definition 2
- Mental representation of the spacial characteristics of a familiar environment
- Rather than a series of chained responses to external cues
Cognitive map example
FOR EXAMPLE…
- Eg. way to campus from one end to another i rmb
- i dont rmb the way as a series of instructions like turn left, turn right etc.
Latent learning 2
1) Learning can occur in the ABSENCE of rewards/punishments
2) Learning is only demonstrated once there is a motivation/reinforcement
Tolman’s rat maze running experiments - what was it, what does it prove? 2
1) COGNITION MAP
- Rats didn’t just memorize turns(“turn left, then right”).
- Instead, they built amental mapof the maze, used this to figure out where the foodshouldbe
2) LATENT LEARNING
-Different levels of food given
- Only when given food, learning was demonstrated
Tolman’s rat maze running experiments - how did it show cognition map? 3
- Training Phase:
- STRAIGHT PATH
- Rats learned to run from start to (food) in a maze shaped like a STRAIGHT PATH - The Surprise Test:
- MANY DIFF PATHS OPENED
- Tolman blocked the original path and opened a “sunburst” of 12 new paths
- The rats had to choose a new path to find the food. - What Happened:
CHOSE PATH WHICH POINTED TO ORIGINAL LOCATION OF FOOD (ie. slightly to the right)
- Instead of randomly picking paths or choosing the shortest one…
Tolman’s rat maze running experiments -results pointing towards latent learning
1) Red group = no food at end of maze
-> less mistakes over time, but they are not really getting better at finding final destination, pretty stagnant
2) Blue group = regular food (ie. Always present in maze)
-> drastically make less and less mistakes each trial
-> pretty much get perfect in finding the final destination
3) Green group = no food reward until day 11 (ie. no food, only present in maze after day 11)
-> up to day 11: look like group A, pretty stagnant
-> after day 11: once there is a food reward they become like blue group, almost no mistakes!
Tolman’s rat maze running experiments - how does it prove latent learning
PART 1
- actually ALL groups were learning and had a cognitive maze in their heads
- only when there is a reward like food, they will show the learning!!
-> they are motivated to do so/only when there is a reason
ALSO PART 2
- Learning can occur in the absence of directly experienced rewards and punishments!
Social Cognitive Learning Theory
- key component = OBSERVATIONAL LEARNING
- definition: learning through imitating others, copying what we observe (a model!)
Bandura test on latent learning - kids + doll - describe 4
- Setup:
- Children watched a short film with an adult acting aggressively toward a Bobo doll eg. Punching it
- There werethree versions of the film, each showing different consequences for the adult’s aggressive behavior:
* Reward Condition: The adult was rewarded for aggression.
* Punishment Condition: The adult was punished for aggression.
* No Consequence Condition: No consequences were shown.
* Each child watched only one of these videos - Observation Phase:
- After watching the film, each child was left alone in a playroom filled with toys, including the Bobo doll
- The children didn’t know they were being watched through a one-way mirror. - What Researchers Measured:
- They observed whether the children imitated the aggressive behaviors they saw in the film. - Incentive phase
- person comes in later, now gives them permission + encourages them to imitate behaviour shown in video)
Bandura test on latent learning -explain findings, refer to graph in GDocs
TIP:
- compare black between all males, compare all white between all females etc. - dont mix up categories
UNDERSTANDING:
- White = No incentive = left in a room to play with the doll
- Black = Positive incentive = person encourages them to imitate behaviour shown in video
- X axis, Model rewards = ie. watched video where punching was rewarded
- WHITE BARS - OBVIOUS
- reward + no consequences are roughly similar to each other
- Punishment - lowest due to learnt consequences
(OPERANT LEARNING) - BLACK BARS
- All increased compared to their associated white
-> proves LATENT LEARNING + ALL children learnt the behaviours EQUALLY (regardless of punishment in video or gender) - GENDER
- Boys higher than all girls to their associated bars - tend to be more aggressive
Bandura test on latent learning - what does it mean by vicarious reinforcement/punishment?
VICARIOUS = learn from others/observation
- here learn the consequences through the video!
Multi-store model of memory - SUMMARISED 3
1) Sensory registers
2) Short term memory
3) Long term memory
Multi-store model of memory - STEP 1, STORY
1) SENSORY REGISTERS
- these HOLD information about a perceived stimulus
- BUT only for half a second, 500 milliseconds)
- After this, the information is LOST (INFORMATION LOSS)
eg. a random OTP code
Multi-store model of memory - STEP 2, STORY
2) SHORT TERM MEMORY
- BUT if the information is PAID ATTENTION to…
eg. you CONSCIOUSLY focus on the stimulus
- this info is MOVED from the sensory register to short term memory!!
- it is held in memory for roughly 20-30 SECONDS
HOW IS IT MAINTAINED IN THE STM?
- you can make it STAY in your STM by REPEATING it over and over!!!
- but this is only temporary - once you stop repeating, the memory fades :(
Multi-store model of memory - STEP 3, STORY
3) LONG TERM MEMORY
- BUT if you keep repeating it over and over and over, it will be MOVED to your long term memory!!
eg. a poem, phone number
- This is called MAINTENANCE REHEARSAL
BUT…
- if the info is UNDERSTOOD, the memory will be stronger in the long term memory
- known as ELABORATIVE REHEARSAL
Retrieval?
this is recalling info so that it is brought from the long term memory BACK to the short term memory
2 types of sensory registers
1) ICONIC MEMORY
- Visual sensory register
- This HOLDS visual information for half a second
2) ECHOIC MEMORY
-Auditory sensory register
- This HOLDS auditory information for 3-4 seconds
Sensory registers - capacity + duration
LARGE CAPACITY, SHORT DURATION
- Can hold lots of info
- But only for a short period of time
Sperling’s experiment - what is it?
ON THE SCREEN THERE ARE 12 LETTERS - WILL APPEAR + DISAPPEAR
1) FULL REPORT
- Participants were asked to recall all the letters
- They could only recall about 3-5 items
-> This UNDERESTIMATES the CAPACITY of the visual sensory register/iconic memory!!
SO…
2) PARTIAL REPORT
- Participants were cued to recall only one row of letters after seeing the entire array by using a certain pitch of tone
- They were able to repeat back all 4 letters!!
-> Since they didn’t know which row would be cued beforehand, their ability to recall all 4 letters showed that all 12 letters are stored in visual sensory register!!
-> SO…This revealed that iconic memory has a very large capacity!!
* BUT duration is short - by the time they say all 4 letters, already forgot the other 8 letters
GENERAL 3 STEPS OF MEMORY
THINK OF LEARNING IN BIO CLASS
taking in info, understanding, interpreting it –> storing it in the memory –> retrieving it
1) Encoding
2) Storage
3) Retrieval
Encoding DEFINITION - 3
Acquiring info + attending to it + then interpreting it
- Acquiring info from experiences/stimulus
- AND attending to info through mental processes (ie. is it important?)
- THEN interpreted by comparing it to prior knowledge
Storage
- Information is stored in NEURONS in the brain
- CATEGORISATION: Diff kinds of memories are stored in diff neuron networks
FOR EXAMPLE: neurons in the visual cortex store information about sights, while
neurons in the amygdala store information about the emotions that were experienced
** - Storage capacity + duration differ between diff memory systems
Retrieval - what does it involve 3?
1) Remembering info = personal reminiscence, recalling facts
2) Doing = executing practical skills
3) Knowing = conditioned responses
Explicit vs implicit retrieval -meaning + examples
Explicit = on purpose // implicit = involuntary
- Explicit = eg. remembering what you had for dinner last night
- Implicit = eg. know how to ride a bike without thinking of the steps
What can trigger retrieval? 2
- Highly context dependent
- Environmental CUES (eg. Hearing a song, photograph might make you think of a vacation in the past)
- Internal (interoceptive) CUES = feeling hungry might make you rmb a memorable meal, period cramps reminds you of your period experience as a kid
Reconsolidating process - what is it, detail of the process?
DEFINITION:
- Recalling a memory
(eg. the structure of an amino acid)
DETAIL OF PROCESS
- This process is RECONSTRUCTIVE -> piece together various parts of the memory, fill in the gaps w/ existing knowledge -> therefore ERROR-PRONE
