Chapter 10 : Observational learning
Theories of Observational Learning
Social Cognitive Theory: > Who is associated with this theory? B A N D U R A > Claims: observational learning is EXPLAINED BY 4 PROCESSES THAT OCCUR DURING or SHORTLY AFTER observation of a model > ATTENTIONAL : have to do with the ORGANISMS OBSERVING RELEVANT ASPECTS of the MODEL'S BEHAVIOR & its consequences > RETENTIONAL : an observer's actions to RECALL MODEL'S BEHAVIOR > MOTOR REPRODUCTIVE : MUST HAVE THE MOVEMENT CAPACITY to be able to REPLICATE the OBSERVED BEHAVIOR > MOTIVATIONAL : observer must EXPECT THAT A REPLICATED BEHAVIOR WILL PRODUCE REINFORCEMENT...& INTERESTED IN /want such reward
Spence's Theory
Spence's Theory: > The tendency TO RESPOND TO ANY GIVEN STIMULUS is the RESULT OF THE INTERACTION of the increased & decreased TENDENCIES TO RESPOND, as reflected by the EXCIATORY GRADIENT and the INHIBATORY GRADIENT > Therefore, after discrimination training, the highest rate of responding... WILL NOT BE FPOR THE S+ because it RESEMBLES the S- MORESO THAN DOES A STIMULUS EVEN FARTHER AWAY > Known as a ...PEAK SHIFT PEAK in responding > Benefit : EMPRICAL ....DATA DRIVEN
Proactive Inference
A phenomenon whereby OLD LEARNING interferes with RECALL of RECENT INFORMATION
Concept Formation
Book defines concept as... CLASS OF THINGS that share ONE or MORE defining feature > Notion that THROUGH TRAINING, we LEARN TO GENERALIZE WITHIN CONCEPTS and DISCRIMINATE BETWEEN concepts > Other experiments show that animals can discriminate between...RELATIONAL CONCEPTS
Preparedness and learning
Breland & Breland (1961): "The MISBEHAVIOR of ORGANISMS" > Operant PROCEDURES FOR ANIMAL TRAINING Occasional manifestation of NON-REINFORCED BEHAVIORS THAT GET IN THE WAY OF LEARNING otherwise basic behaviors WHY? INSTINCTIVE DRIFT : the tendency of an animal to revert to a FIXED ACTION PATTERN (sequence of innate behaviors) - IF an action is INCOMPATIBLE or CONFLICTS with a FIXED ACTION PATTERN, the ORGANISM WILL HAVE DIFFICULTY LEARNING it > Continuum of Preparedness: the idea that organisms are GENETICALLY PREDISPOSED to LEARN SOME THINGS & not others
Kinds of Memory
Declarative > deals with INFORMATION THAT CAN BE EXPRESSED (which is usually done with words) > Sometimes called "EXPLICIT" MEMORY > ENDEL TULVING: Proposes DECLARATIVE be divided into: > SEMATIC Memory: the kind of memory that MAKES IT POSSIBLE for organisms to ACQUIRE, STORE, AND USE information about the world in which they live - "KNOWLEDGE OF THE WORLD" > ___________ Memory: memory for personally experienced events, the "__________" of our own lives > Possibly unique to ______, & a later __________ development
MORE Variables in Forgetting
Degree of Learning > Better something is learned, LESS LIKELY it is to be forgotten > OVERLEARNING: the continuation of TRAINING BEYOND the POINT REQUIRED to produce one errorless performance > Relationship with remembering/forgetting? **** The greater the the OVER LEARNING, THE LESS FORGETTING - THE GREATER THE OVERLEARNING, the LESS forgetting. Diminishing RETURNS. > FLUENCY: the number of CORRECT RESPONSES per minute Relationship with remembering/forgetting? > PRIOR LEARNING > PROACTIVE INTERFERENCE
Generalized Imitation
> An observing organism MAY IMITATE A MODEL'S BRHAVIOR even if that BEHAVIOR IS NOT REINFORCED - EVOLVED/ SURVIVAL "Following the crowd" it WORKED BEFORE , it will WORK AGAIN > Developmental/Learning history: possible to reinforce not only the IMMITATION OF SPECIFIC ACTS, but to reinforce a GENERAL TENDENCY TO IMITATE "Following the Crowd" > Imitation ITSELF IS A KIND OF BEHAVIOR that CAN BE STRENGTHENED or WEAKENED depending on its CONSEQUENCES
Pavlov's Theory of Discrimination
> Believed that discrimination training produced...PHYSIOLOGICAL CHANGES in the brain > CS+ becomes associated with an EXITATION (reflex action will be evoked)area, while CS- becomes associated with an INHIBITION (suppressed) area > Limitation: These physiological events are... ONLY INFERRED from OBSERVED BEHAVIOR, NOT EMPIRACALLY DEMONSTRATED
The Power of imitation
> Bobo Doll study - Imitative, AGRESSIVE BEHAVIOR - "novel aggression"
Introduction
> Can one individual learn by observing the experiences of another organism? > Thorndike's contribution: > first to really look at observational learning - Watson - THEIR RESEARCH >Absence of findings slows down research in this area for about a quarter century Carl Warden (1930's) : One of the FIRST people to DEMONSTRATE OBSERVATIONAL LEARNING OF ANIMALS . > 1960's...Albert Bandura: Social Learning Theory
Variables affecting Observational Learning
> Consistency : >CONSEQUENCES FOLLOWING observed behavior > Observers will do what is most useful... FOR THEM > Human observers tend to LEARN MORE FROM MODELS WHO are COMPETENT, ATTRACTIVE, LIKABLE, & PRESTIGOUS than from models who are not
Stimulus Control
> DEF: > When DISCRIMINATION TRAINING brings BEHAVIOR UNDER the influence of DISCRIMINATIVE STIMULLI > may be the result of a... COMPLEX COLLECTION OF STIMULLI rather than just ONE SINGLE stimulus > Benefit of: increases... EFFICIENCY of behavior
Forgetting
> DEF: a DETERIATION in LEARNED BEHAVIOR following A PERIOD WITHOUT PRACTICE > And NOT due simply to INJURY OR CHANGE IN PHYSICAL CAPACITY to demonstrate a behavior > "DETERIORATION" - because behavior is often LESS PRECISE and/or LESS EFFICIENT than it was during retention > In the context of this course, we focus on DETERIRATION OF BEHAVIOR, rather than of a NEUROLOGICAL RECORD OF EXPERIENCE
Semantic Generalization
> DEF: when learned behavior generalizes on the basis ... OF AN ABSTRACT FEATURE > This phenomenon demonstrates that, at least in humans, generalization can be based on... - ABSTRACT CONCEPTS AS WELL AS PHYSICAL PROPERTIES > FURTHERS concerns ABOUT PREDJUDICE EX. > Having a great experience in another country, when you hear the name of that country or anything affiliating with that country you AUTOMATICALLY STAN
nondeclarative memory
> Deals with memory that CANNOT BE EXPRESSED; IMPLICIT KNOWLEGE - Classically conditioned response - POTENTIALLY CONFUSING
Reminiscence
> Def: A phenomenon in which PERFORMANCE IMPROVES with the PASSAGE OF TIME > WHY? > When PHYSIOLOGICAL CONDITIONS present DURING TRAINING AGAIN BECOME PRESENT during 'testing.' > STATE INDEPENDENT LEARNING: a BEHAVIOR THAT IS LEARNED when the ORGANISMS IS IN A PARTICULAR PHYSIOLOGICAL STATE is LOST WHEN THAT STATE PASSES; though MAY EMERGE once that state is once again achieved
Basics of "Observational Learning
> Definition: a change in behavior due to the experience of OBSERVING THE CONSEQUENCES THAT A MODEL EXPERIENCES for its OWN BEHAVIOR - model does a behavior and HAS CONSEQUENCES that other organisms learn from > Two Basic Procedures - VICARIOUS REINFORCEMENT : behavior has an INCREASE IN STRENGTH due to observation. > VICARIOUS PUNISHMENT : behavior has a DECREASE IN STRENGTH due to observation. observer has a change in behavior BASED ON MODEL'S behavior getting REINFORCED
Neurological Damage & Learning
> Environment Fetal Alcohol Syndrome (FAS) & Fetal Alcohol Effects (FAE) LEAD EXPOSURE/poisoning (after BABY IS BORN) Head Injury Enriched environments:
Background on memory and forgetting
> Evolution of metaphors used to understand & explain memory > Memory, in general, as the PROCESS OF HOW "REPRESENTATION" of an experience are ENCODED,STORED, AND RETRIEVED > EBBINGHAUS (late 1800s): - some of the EARLIEST STUDIES OF MEMORY - Worked on MEMORIZING LIST of NONSENSE SYLLABLES
Assessing Memory
> FREE RECALL : organism is GIVEN OPPORTUNITY to PERFORM A PREVIOUSLY LEARNED BEHAVIOR FOLLOW RETENTION INTERVAL LIMITATION: " ALL OR NOTHING" form of measuring > Cued recall: approach that HINTS/PROMPTS to INCREASE LIKELIHOOD that a behavior will be produced - Gets at the more SUBTILE FORMS OF learning that are often NOT CAPTURED BY FREE RECALL > Recognition : individual only has to IDENTIFY THE MATERIAL that WAS PREVIOUSLY LEARNED
Heredity and Learning Ability
> GENETIC DIFFERENCES BETWEEN SPECIES of organisms that AFFECT THE CAPACITY FOR LEARNING. > Frank & Frank (1982): Problem-solving of DOGS vs. WOLVES > Genetic difference WITHIN A SPECIES also has a LARGE INFLUENCE on LEARNING ABILITY >Tryon (1940): "GOOD" vs. "BAD" problem solvers - SELCTIVELY BREEDING WITHIN A CERTAIN SPECIES
Generalization Following Extinction and Punishment
> Generalization is not limited to ...REINFORCED BEHAVIOR > EXTICTION and PUNISHMENT ,may TRANSFER TO OTHER SITUATIONS > The DEGREE OF SIMILARITY is once again an IMPORTANT FACTOR > Results will form a generalization gradient quite like those produced for reinforcement. - spreading out of effect
Generalization
> Generalization: the TENDENCY FOR BEHAVIOR TO OCCUR ... IN DIFFERENT SITUATIONS from the ONE WHICH BEHAVIOR WAS LEARNED > Alternative DEF: tendency to respond ... to STIMULLI NOT PRESENT DURING TRAINING EX. Classical Conditioning Thorndike
Critical Periods
> HARLOW (late 50s & early 60s) Acquisition of SOCIAL SKILLS In Humans? Book suggests that evidence is UNCLEAR FOR EXISTENCE of CRITICAL PERIODS in humans Case of "Genie Wiley" - > Notion of "SENSITIVE PERIODS"
Nonheritability of Learned Behavior
> Learned behavior is NOT INHERITED Blank Slate? > Versus REFLEXES & MODEL ACTION PATTERNS > GENETICS Notion of HERITABILTY of AQUIRED CHARACTERISTICS : > LAMARCK - First THEORY of EVOLUTION > Suggested that EVOLUTIONARY PROCESSES occurred because of A SPECIES' EFFORT TO ADAPT to its environment
Regions of Memory Formation & Storage
> Makes intuitive sense that different kinds of memory would get stored in different places > HIPPOCAMPUS: NEURAL STRUCTURE involved in the FORMATION of long-term memories > Case example of H.M. illustrates > HIPPOCAMPUS SEEMS more involved in EPISODIC > PROCEDURAL MEMORY may be INDEPENDENT of HIPPOCAMPUS > > If a memory has ANY STRUCTURE AT ALL, it is not a UNIFIED SPACE but a COMPLEX WEB OF NEURONS throughout the brain
Discrimination (cont)
> Operant Learning Application "Payoff" > Pretty astonishing results possible in basic life forms > Porter & Neuringer (1984): pigeons & Music - got PIDGEONS TO DISCRIMINATE music , PECKING PRODUCED FOOD or NO FOOD -Trained to discriminate CLASSICAL MUSIC BASED ON COMPOSER > Watanbe et al. (1995): pigeons & Art - Trained them to 90% ACCURACY - ACTUALLY DISCRIMINATED AGAINST THE STYLES of these artist - NOT JUST those used in training but OTHERS NEVER SEEN BEFORE
Can increase generalization by...
> Providing Training in a WIDE VARIETY OF SETTINGS > Survival Value > Dangers ? - The EXTENSION OF PSYCHOPATHOLOGY EX. A certain Dog breed bites you, now everytime you see a dog, heart rate goes up because of that one experience -ANIMAL ENCOUNTERS - PREJUDICE/ STEREOTYPES
Measuring Memory
> RELEARNING METHOD: assesses the AMOUNT OF TRAINING that MUST BE PROVIDED in order to REACH A PREVIOUSLY LEARNED LEVEL of PERFORMANCE > Also known as the SAVINGS METHOD > Gradient Degradation: a behavior is TESTED FOR GENERALIZATION BEFORE and AFTER a RETENTION INTERVAL. > A after passage of time indicates that has occurred
Among Children
> Reinforcement of a MODEL'S BEHAVIOR ADDS substantially to the EFFECTS OF DIRECT REINFORCEMENT > DIFFERIENTAL EFFECTS of reinforcement & punishment > Praising or reprimanding of adult's behavior > Similar effects for adults (see research by Kanfer & Marston on college students)
in the presence of others
> Replication ' > Non Replication - can happen even if observer was not paying attention - IF YOU SEE REPLICATED BEHAVIOR that turns out to have AN AVERSIVE CONSEQUENCE.....MOST LIKELY IMITATION - If REPLICATED BEHAVIOR has a POSITIVE CONSEQUENCE .....it is OBSERVATIONAL LEARNING
Carl Warden
> Results show that OBSERVERS CAN/DO BENEFIT > focused ON MONKEYS - watched another monkey SOLVE A PROBLEM and get REINFORCEMENT FOR DOING SO > Takes observer LESS TIME than would be expected if NO OBSERVING > Even FAILED PROBLEM SOLVING SUGGESTS learning HAS OCCURED > The MORE OBSERVING the BETTER THE LEARNING
Remembering & Forgetting
> Retrograde Ammesia: loss of memory prior to an event Antero Amnesia : INABILITY to FORM NEW MEMORIES of FOLLOWING AN EVENT
Background
> Semester in Review... >LEARNING!!! >Limitations > Learning is NOT AN INFINITE CAPACITY
Generalization Gradient
> Similarity to the TRAINING SITUATION > Relationship between novel stimulus & training stimulus: - More SIMILAR IT IS MORE LIKELY to RESOPOND TO TRAINING STIMULUS > Generalization gradient is the data of plotted results Generally, a function of ...SIMILARITY
Coming back to Metaphors
> So...memories are NOT "STORED" or "RETRIEVED" in the MANNER in which we COMMONLY PICTURE THEM > Not actually "FILING THEM AWAY" or "pulling them up" > Nothing actually moves > What we talk about as "memory" actually just seems to be PSYCHOLOGICAL EVENTS that involve CHEMICAL REACTIONS and NEURONAL ACTIVATION
Discrimination Training Procedures
> Successive Discrimination: the S+ & S- typically... ALTERNATE RANDOMLY > Simultaneous Discrimination: stimuli presented...AT THE SAME TIME
Miller-Dollar Reinforcement theory
> The changes in an observer's behavior are due to... THE CONSEQUENCES OF BEHAVIOR, NOT the model > Considers observational learning as merely A VARIATION OF STANDARD OPERANT LEARNING > Perhaps biggest difference between 2 theories is that... - Bandura's theory explains through INTERNAL/MENTAL PROCESSES Miller-Dollard theory looks to the SITUATION and the OBSERVER'S LEARNING HISTORY
Learning to remember
> The old 7 plus/minus 2 rule > Definite genetic variations in this capacity > However, also seems to be a learned skill Ericsoon & Chase (1982): Participant "S.F." sets new record of how many random digits he could remember = 82!
Errorless Discrimination Training
> Training will USUALLY INVOLVE A NUMBER OF ERRORS before ONE CAN TRULY DISCRIMINATE which option is correct > But...ERRORS CAN BE PUNISHING > Errorless Discrimination Training: a form of discrimination training in which the S- is introduced... IN A VERY WEAK form and GRADUALLY STRENGHTHENED while the S+ is CONSTANTLY REINFORCED. - By the time S- reaches the SAME STRENGTH as S+, there are FEW ERRORS MADE ****who is credited with this approach?
Closing Thoughts
> Why do we forget at all? > Given that where we are today is the result of eons of natural selection > BROKEN RECORD: once again SURVIVAL VALUE & FUNCTIONALITY > Almost none of the things that we are asked to remember today would have any survival value in the time of our ancient ancestors > Forgetting may also allow us to DISCARD PREVIOUS UNSUCESSFUL ATTEMPTS at PROBLEM SOLVING rather then STAYING STUCK ON THEM
Variables Affecting Observational Learning
> YOUNGER observers seem to be MORE LIKELY TO PURELY IMITATE the behavior of a model than OLDER OBSERVERS > BUT the AMOUNT OF BENEFIT OF OBSERVATIONAL LEARNING corresponds to COGNITIVE CAPACITY - CORROSPONDS WITH AGE > Observer's learning history
Discrimination
> the tendency for behavior to occur in situations that ... CLOSELY RESEMBLE the one WHICH the behavior WAS LEARNED, but NOT in SITUATIONS THAT DIFFER FROM IT > _INVERSE relationship between generalization & discrimination > Discrimination Training: involves procedures intended to get animals to be able to ...DISTINGUISH BETWEEN SPECIFIC STIMULLI > In Classical Conditioning: CS & CS > In Operant Conditioning: S+ S-
Observational Learning vs. Imitation
Caution! > Imitation: COPYING AFTER SEEING action. > Again...Observational Learning is A CHANGE IN BEHAVIOR due to observation. > Basically...take home message is that IMITATION IS NOT SYNONOMOUS WITH OBSERVATIONAL LEARNING > Imitation may even suggest that LEARNING HAS NOT OCCURED EXAMPLES? > One kid burns his hands on stove, runs away crying - One kid says "oooh that looks so cool" and TRIES IT like a dummy (imitation) - Other kid recognizes the pain the initial kid was in, does not try it > so we need to DECIDE WEHTHER REPLICATED BEHAVIOR is TRULY OBSERVATIONAL LEARNING or IMITATION What would you classify the Bobo Doll experiment as?
Habits & Stimulus Control
Environmental stimuli that have BEEN PAIRED REINFORCING EFFECTS of a HABIT ACQUIRE a degree of STIMULUS CONTROL > For example, the course of life itself is discrimination training for a smoker, and many cues become discriminative stimuli that take on stimulus control > Thus, in order to truly stop a habit...What are the options???, NOT WILLPOWER.....but CHANGING INFLUENCE of ENVIRONMENT
Observational Learning
Learning by OBSERVING EVENTS and THEIR CONSEQUENCES > Observation OF OTHERS > ASOCIAL OBSERVATIONAL LEARNING - EXPERIENCE LEARNING based on the OBSERVATION OF EVENTS without A MODEL - watching EVENTS THAT HAPPEN
Subsequent Learning
Learning can also IMPACT OUR RECALL OF PREVIOUS EXPERIENCE
Applications &Interpretations
Loftus et al . "CONFABULATION" - MADE UP & inaccurate MEMORIES Eyewitness Testimony Wording nuances Loftus & Palmer (1974): "hit" Loftus & Zanni (1975): "the" vs. "a" Serious implications for... WHAT WAS OR WAS NOT PRESENT - SUSPECT INTERIGATION BY POLICE, WITNESS CROSS EXAMINATION BY LAWYERS AND JURY DELIBERATIONS
Prior Learning
Meaningful INFORMATION is easier to remember than that WHICH is RANDOM. Ex. Chess board arrangement
Kinds of Memory
Short and Long Memories: > Has only to do with the AMOUNT OF TIME...NOTHING at all to DO WITH the TYPE or NATURE OF INFORMATION > REJECTION INTERVAL is the PERIOD BETWEEN a LEARNING EXPERIENCE and its RECALL > SHORT TERM MEMORY the capacity to REMEBER something over a SHORT RETENTION INTERVAL - Primary Memory (HOLDING TANK for memory) - working memory (able to MANIPULATE) and work with the information from PRIMARY MEMORY > LONG TERM MEMORY the capacity to REMEBER SOMETHING OVER LONGER PERIODS of time
Retroactive Interference
When NEW INFO LEARNED interferes with EARLIER LEARNING > New learning "PUSHES OUT" old learning
Variables in Forgetting
Why do we forget anything? > Passage of TIME . Does our neurological record DECAY over time? > Research base demonstrates correlation: LONGER the RETENTION INTERVAL, the MORE THE FORGETTING > BUT...does PASSAGE OF TIME CAUSE the forgetting??? > NO need to EXAMINE FACTORS that CAN ACCOUNT FOR the forgetting that OCCURS OVER TIME
Mnemonics & Mnemonic systems
a system FOR AIDING RECALL
Method of Loci
an approach in which each PARTICULAR FACT IS FIXED TO A SPECIFIC LOCATION > Typically a MENTAL REPRESENTATION OF A FAMILIAR PALCE
Cue- Dependent Forgetting
forgetting that RESULTS FROM THE ABSENCE OF CUES that were PRESENT DURING TRAINING "Retrace your steps" "Passwords"
Procedural memory
knowing HOW TO PERFROM A PROCEDURE ex Accidently writing down anquity when you meant to write anxiety