Learning Psychology PSYC-2502 PDF
Document Details
Uploaded by RichDryad2480
Işık University
Dr. Emel Erdogdu
Tags
Summary
These are lecture notes for a course on Learning Psychology, specifically covering classical conditioning. The lecture notes describe various aspects of classical conditioning, such as extinction, spontaneous recovery, stimulus generalization, and stimulus discrimination. The notes also include information on experimental neurosis and the study of learning without an unconditioned stimulus.
Full Transcript
Learning Psychology PSYC-2502 Dr. Emel Erdogdu [email protected] 1 Classical conditioning so far.... CS, CS, UR, US Response acquisition Extinction Spontanious recovery Disinhibition Stimulus generalization / Stimulus di...
Learning Psychology PSYC-2502 Dr. Emel Erdogdu [email protected] 1 Classical conditioning so far.... CS, CS, UR, US Response acquisition Extinction Spontanious recovery Disinhibition Stimulus generalization / Stimulus discrimination Extinction Reduction of the CR when CS is presented without US According to Pavlov, extinction is not deletion but another thing is learned: NOT to increase salivation Extinct CR can recover fast, as soon as CS is paired with US again Spontaneous recovery Reappearance of the CR after extinction İf the recovery was endogenous the response will decrease again each time the CS is presented without US Disinhibition Disinhibition is the temporary increase in strength of an extinguished response towards a newly introduced /unrelated stimulus effect For example: if your anxiety while giving a speech in class gradually fades, it may suddenly recover when a noisy ceiling fan starts up or someone walks in late. Stimulus generalization When CR are elicited by other stimuli that show some similarity with the original CS 2000 Hz sound , US saliva 1900, 2100, 4000 Hz Little Albert; White coat + nurse Semantic generalization previously conditioned to the word "wildcat" may be elicited by the word "tiger« Stimulus discrimination The ability to differentiate between a conditioned stimulus and other stimuli that have not been paired with an unconditioned stimulus 2000 Hz sound , US saliva 1900 → no CR Experimental neurosis «Abnormal behavioral condition» produced by a problem scenario the lab animal cannot solve because it is too difficult or impossible. The stress of not being able to solve the experimental problem causes altered behavior which mimics a mental disorder (artifically produced psychosis) Pavlov noticed when dogs were very confused and unable to distinguish between stimuli they growl and act aggressive towards experimenters which was abnormal behavior for the well trained dogs. Classical Conditioning: Mechanisms A. What Makes an effective Conditioned and Unconditioned Stimulus? B. What Determines the Nature /influences the Conditioned Response? C. How do Conditioned and Unconditioned Stimuli Become Associated? A - What Makes Effective Conditioned and Unconditioned Stimuli? 1. Initial Responses to the Stimuli (Pavlov) (reflexive) 2. Novelty of Conditioned and Unconditioned Stimuli 3. CS and US Intensity and Salience 4. CS–US Relevance, or Belongingness 1 - Initial Responses to the Stimuli What is, or how is the first reaction changing with the presence of the US and CS ? İdentifying potential CSs and USs requires comparing the responses elicited by each stimulus before conditioning → all is relative The US in one experiment might be the CS in an other Sugar water paired with sickness → CS (taste-aversive) Presentation of lever followed by sugar water (US) (sign- tracking) 2 - Novelty of Conditioned and Unconditioned Stimuli If either the CS or the US is highly familiar, learning occurs more slowly than if the CS and US are novel. A: The Latent-Inhibition or CS-Preexposure Effect Preexposure phase: repeated CS without US produces habituation. CS becomes familiar and insignificant Preexposure disrupts or retards learning US with CS → effect called late inhibition → limits processing and attention to inconsequential stimuli İn schizophrenia inability to suppress attention to irrelevant stimuli (disrupted late inhibition !) 2 - Novelty of Conditioned and Unconditioned Stimuli B: The US-Preexposure Effect Conditioning of CS proceeded faster for unfamiliar US than for familiar US One food type is given before conditioning procedure Lever (CS) and new food (US) is faster learned than CS and familiar food (fUS) 3 - CS and US Intensity and Salience Most biological and physiological effects of stimulation are related to the intensity of the stimulus input Intensity → Salience: significance or noticeability of stimulus İncrease salince by making it: more attention-getting more relevant to the biological needs similar to likely environmental encounters (naturalistic) Sexual conditioning is more robust with the naturalistic CS. 4 - CS–US Relevance, or Belongingness extent to which the CS is relevant to or belongs with the US Innate tendency to associate certain stimulus types Garcia & Koelling (1966) → how rat learn about pain or illness Natural, likely to: - get sick after eating poisoned food - Feel pain after beeing bit by predator - food (flavored water - taste CS) and predator representing cues (audio-visual (sound/light) CS) - Shock or illness US taste-sick A.V.-shock CS-US relevance/belonging phenomenon can be genetically related Already present in very young infants Faster learning of Fear of snakes vs fear of flowers “Fear conditioning occurs most readily in situations that provide recurrent survival threats in mammalian evolution” Mineka and Öhman (2002) Learning Without an US So far: Clas.Cond with US → stimulus that has a large unconditionally behavioral impact, without prior training But Pavlovian CC works not only for sex, shock or food... Two forms of classical conditioning without US A. Higher-order conditioning/second order cond. B. Sensory preconditioning A: Higher-Order Conditioning (HOC) Irrational fears are often learned through H.O.C H.O.C occurs in two phases Lady → learned fear of crowds (CS1) (injury US?)→ crowded theater → fear of theater → movie theater became aversive stimulus (CS2) Conditioning of CS2 in absence of any US Similar to the standard procedure for inhibitory conditioning First – order second –order conditioning HOC is a form of learning that occurs when a neutral stimulus is repeatedly paired with a conditioned stimulus, that is already associated with a desired response through conditioning, to become another conditioned stimulus itself Why does one procedure lead to conditioned inhibition in one condition but to excitatory second order conditioning in others? Timing simultaneous CS1 & CS2 → conditioned inhibition Sequential CS1 & CS2 → excitatory (HOC) Also used in advertisement industry (pairing product with things you already like/desire) (remember evaluative conditioning) B - Sensory Preconditioning You already had an association of vanilla & cinnamon. When learning aversion towards cinnamon, very likely you will also have aversion towards vanilla as it was already paired involves a two-stage process as in HOC remember evaluative conditioning B - Sensory Preconditioning Sensory preconditioning Second phase Now also aversion phase towards vanilla Cinnamon/Vanilla → US (illness) is (CS2) CS1/CS2 paired with CS1 Does not cause any (Cinn.) obvious behavior Aversion: CR B - What Determines the Nature of the Conditioned Response? 1. The US as a Determining Factor for the CR 2. The CS as a Determining Factor for the CR 3. The CS–US Interval as a Determining Factor for the CR 1 - The US as a Determining Factor for the CR The US nature of the conditioned response influences the CR The form of the CR resembles the form of the UR Jenkins and Moore (1973) Light + food vs light + water Do food and water cause different CR? Yes! CR is US dependent 1 - The US as a Determining Factor for the CR Stimulus substution model: according Pavlov, CS is assumed to activate neural circuits previously activated only by the US and elicit responses similar to those elicited by the US Solid arrow indicates preexisting neural connections. Dashed arrow indicates neural connections established by conditioning. Because of these new functional connections, the CS comes to elicit all responses previously elicited only by the US. Not true for all situations → shock/jump but tone freeze 2 - The CS as a Determining Factor for the CR The nature of the CS also influences the CR Timberlake and Grant (1975) CS for food (US) was presentation of another rat through a flap door. Expectation based on stim.subst.model: gnawing and biting when rat (CS) is present But Contrary: CS elicited social affiliative responses instead 3 - The CS–US Interval as a Determining Factor for the CR İnterval between the conditioned stimulus and the unconditioned stimulus influences CR Fast or slow approaching car → different behavior Fast: panic, jump, freeze Slow: concerned but planned Short interval: good for activating immediate responses Long interval: activates responses that prepare the organism for the US over a longer time horizon Example of CR difference due to CS-US interval Akins (2000) Sexual conditioning (bird, quail) (US: access to female) CS: light CS-US either 1 min or 20 min 1min: Sign tracking behavior (CS approach) 20min: greater locomotor behavior (pacing between halfs of the experimental chamber) Behavior Systems Theory Can explain different behavior due to interval length and activated behavioral system CR does not only rely on US but also on CS and CS-US interval → Pavlovs Stimulus substitution model cant explain everything Presentation of a US will activate the behavior systems relevant to the US (here sexual US → sexual behavior system) Domjan, 1997; Timberlake, 2001; Rau & Fanselow, 2007 İs it S-R vs S-S learning ? How does the CS produce the response ? Produces S-R connection between CS and UR ? Or Produces S-S connection between CS and US ? Test this with US devaluation Reduce the value of the US → when value is decreased the CR will decrease too US devaluation: Reduction in the attractiveness of an unconditioned stimulus, usually achieved by aversion conditioning or satiation stimulus–response (S–R) learning The learning of an association between a stimulus and a response, with the result that the stimulus comes to elicit the response directly. stimulus–stimulus (S–S) learning The learning of an association between two stimuli, with the result that exposure to one of the stimuli comes to activate a representation, or “mental image,” of the other stimulus. Hence participants respond to the CS not because it elicits a CR directly but because the CS activates a representation or memory of the US. And the CR is assumed to reflect the status of the activated US representation. Food deprived rats Phase 1: normal conditioning (CS-US assoc.) Phase 2: devaluation of US in experimental group by giving the rats sufficient free food to completely satisfy their hunger. Test: if the CR is elicited by the US representation, devaluation of the US representation should reduce responding to the CS. Result: yes, we see reduced CR (less active) after US devaluation X Evidence of S–S learning in a wide range of classical conditioning situations; not all Conditioning can modify the response to the US - Conditioned diminution of UR - A reduction in the magnitude of the UR to an US caused by presentation of a CS that had been conditioned with that US Examples Conditioned analgesia: less feeling of pain (UR) to electroshock (US) after several presentations of the CS release of endogenous opiates due to cues (CS) which reduces pain sensitivity CS was paired with electro shock (natural response) https://link.springer.com/article/10.1007/s11916-019-0766-0 Conditioned Drug tolerance Conditioned Drug tolerance: Cues (CS) accompany administration of the drug are paired with the pharmacological effects of the drug (US) Cues are conditioned to elicit conditioned compensatory responses One consequence of this learning is that the CS elicits physiological processes that counteract the drug effect. Dual process theory: Process a (initial sedation) and process b (hangover)! Cues activate Process b which prepares the body for the coming drug → less effective Unfamiliar environment without cues → overdose C - How Do CS and US Become Associated? 1. The blocking effect 2. The Rescorla–Wagner Model 3. Attentional Models of Conditioning 4. Timing and Information Theory Models 5. The Comparator Hypothesis 1 – The Blocking Effect Interference with the conditioning of a novel stimulus because of the presence of a previously conditioned stimulus. The established association between CS1 and US prevents learning of CS2 and US Phase 1: Light (A) followed by food (US) (paired / unpaired) Phase 2: Light simultaneous with tone (B) followed by food (US) Test effect of tone seperately Little effect for B in experimental group B-US association is blocked because A-US association already exists In control group B was conditioned Little CR stronger CR Some children have difficulties to learn to read new words together with the picture when they learned the pictures before The picture-word association blocks reading -words What is the difference between HOC and blocking effect ???? İn HOC no US present during second phase (only CS1 & CS2) Why is the association of CS1/CS2 sometimes working and sometimes not? → US has to be surprising to be effective in producing learning. If the US is signaled by a previously conditioned stimulus (CS1), it will not be surprising, hence no learning What is surprising? An event is surprising if it is different from what is expected. →effectiveness of a US is determined by how surprising it is → forms basis of formal mathematical model of conditioning (Robert Rescorla and Allan Wagner) 2 - The Rescorla–Wagner Model formal mathematical model as basis for modern learning theories İnvolves importance of «US surprise» Unexpectedly large stimulus → excitatory conditioning Unexpectedly small stimulus → inhibitory conditioning → Effectiveness of a US, depends on how different the US is from what the individual expects 2 - The Rescorla–Wagner Model ΔV = k(λ – V) ΔV: change in value K: konstant λ: US V: value of λ Surprise factor: (λ –V) First trial: λ > V V increases with time until (λ =V) → Surprise decreases with trials İncreased surprise to unexpected US