Week 6 Escape and Avoidance learning PDF

Preview

Summary

PSYU2236 lecture notes...

Description

[2]: Behaviour Consequences: Escape & Avoidance Learning

Escape and Avoidance: Escape: Getting away from an aversive stimulus in progress - Escape behaviour results in termination of aversive stimulus Avoidance: Behaviour occurs before aversive stimulus which prevents delivery of it - Negative contingency between response and aversive stimulus - Results in an increase in behaviour that is maintained by negative reinforcement

Escape & Neurosis and Learned helplessness: - Are made in presence of aversive stimulus - Make a response - Aversive event terminates - Action is negatively reinforced

But what if there is escape? - Better still, you CAN anticipate the aversive event before it affects you and evade it? - Avoidance learning

Avoidance: Active avoidance: Actively making a response to avoid event (e.g running away) Passive avoidance: Learning NOT to make response in order to avoid event (e.g NOT speaking up)

Active avoidance: In a typical avoidance experiment… - Rat is trained in a shuttle box with a hurdle in the middle - Tone is presented for 10 seconds then an electric shock is delivered through floor of cage - If rat jumps over hurdle while shock is on, shock is immediately terminated - If rat jumps before shock comes on, then tone is turned off and scheduled shock is cancelled - Depending on when rat jumps over barrier, it can either escape from shock or avoid it altogether - Called signalled avoidance because experimenter provides signal to indicate when shock is imminent

Discriminative Avoidance:

Measuring avoidance learning: - Avoidance learning is measured in latency/time to respond to signal

- This subject avoided shock on the 9th trial and continued to do so thereafter

Can learn to use SD to avoid US: Brogden, Lipman, and Culler (1938) Guinea pigs: - CS = tone; US = shock (stimulated running) UR = pain; US = Running Classical conditioning group: CS was followed by US Avoidance group: - CS > run > no US - CS > don’t run > US Note: In classical conditioning had NO opportunity to escape

Passive Avoidance: Passive avoidance uses shuttle-box procedure - Is a form of operant conditioning where the person/animal MUST abstain from act/reaction which will otherwise result in a negative outcome - Animal learns to suppress their normal dark seeking reflex because their entry into dark chamber is paired with a foot shock - Thus, they learn to stay in bright side of box to avoid shock

Obsessive Compulsive Disorder vs Phobia: - OCD typically involves an active avoidance response - Phobic behaviour typically involves a passive avoidance response Example: - A person with OCD will clean frequently or compulsively check things - A person with a phobia will avoid object of their fear (e.g dogs or avoid units with presentations)

Avoidance Paradox: - A behaviour is made before aversive stimulus occurs Note: Avoidance Behaviour clearly increases, so is reinforced BUT what is taken away to reinforce it? Mowrer and Lamoreaux (1942): - “Not getting something can hardly, in and of itself, qualify as rewarding” - Problem for behaviourists who NEED to be able to specify a specific stimulus Note: Not getting punished is ONLY rewarding if punishment is expected - That is ONLY if subject is anxious or fearful and if this expectation in some ways gets reduced

Theories of Avoidance: Two process/factor theory: Classical and operant learning experiences are involved in avoidance learning

Mower’s Two Process Theory of Avoidance: Explains avoidance learning in terms of two necessary processes: 1: First, subject learns to associate warning stimulus with aversive stimulus - This is a classical conditioning process - The warning stimulus of light is the CS, the aversive stimulus of shock is the US

2: Subject can be negatively reinforced during warning stimulus; is second operant conditioning process - Two-process theory reduces avoidance learning to escape learning - The organism NOW learns to escape from the CS and the fear which it elicits

Avoidance Conditioning & Phobia: Definition: Irrational fear of specific object or situation where fear is disproportionate to real threat Acquisition which involves Classical Conditioning - Elevator (CS): Feeling Trapped (US) > Fear (UR) - Elevator (CS) > Fear (CR) Maintenance: Avoidance (i.e negative reinforcement ) - Elevator (SD): Avoid Elevator (RR) > Reduced Fear (SR)

Support for Two Factor Theory: - Two-factor theory predicts that avoidance responding will be learned ONLY to extent that warning signal terminates when response is made Kamin (1957) trained four groups of rats in a two-chamber avoidance apparatus… - Terminate signal; avoids shock - NO termination of signal; avoid shock - Terminates signal; does NOT avoid shock & CONTROL - A significant amount of avoidance responding occurred in the first group ONLY Note: Response terminates signal and enables animal to avoid shock - Avoidance responding was poor in group that could avoid shock BUT could NOT terminate signal

Futher Support for Two Factor Theory: - Delaying tonset of reinforcement reduces effectiveness of reward, Thus it should be possible to reduce level of reinforcement by introducing delay between avoidance response and termination of feared stimuli After avoidance response, the CS was terminated either…. (1): Immediately (2): 2.5 seconds after response (3): 5 seconds after response (4): 10 seconds after the response - As predicted, animals in zero delay condition successfully avoided shock on over 80% of trials. - Animals in 10s delay condition avoided shock on fewer than 10% of trials - Effectiveness of CS termination to support avoidance was decreased by increasing delay - Results suggest source of reinforcement in avoidance conditioning was reduction of fear generated by termination of CS

Sidman Free-Operant Avoidance: Avoidance can be learned without a warning CS: - Shocks at random intervals - Response gives safe time

- Extensive training BUT rats learn avoidance (i.e errors and high variability across subjects)

Support for Two Factor Theory: Sidman Free-Operant; bit of a problem for Two Factor theory BUT… 1: Rescorla and Lolordo (1965) trained dogs using the Sidman avoidance procedure - NO warning stimulus and shock is programmed to occur at fixed time intervals 2: Shocks were programmed to occur every 10 seconds BUT every time dogs jumped over hurdle in shuttle box they ensured a shock-free period of 30 seconds 3: By jumping at least once every 30s dogs could ensure they would NEVER receive a shock

Rescorla, R.A., & LoLordo, V.M. (1965): - Once dogs learned avoidance response they were confined to one half of shuttle box and given discriminative fear conditioning trials - One tone (CS+) was followed by shock and another tone (CS-) was NOT - Sidman avoidance training was resumed and once dogs were responding reliably, CS+ and CS- were occasionally presented for 5 seconds - When the CS+ was presented, rate of jumping doubled - When the CS- was presented, rate of responding fell to almost zero Rescorla & LoLordo: - CS+ can amplify avoidance & CS- can reduce avoidance - Evidence that conditioned stimuli have acquired drive properties Note: Supports Two Factor Theory’s position that it is CS that drives avoidance response

Evidence against Two Factor Theory: Conditioned avoidance responding in dogs: Solomon, Kamin & Wynne (1953) - Light (CS): Shock (US) > Fear (UR) - Light (CS) > Fear (CR) (i.e fear response elicited by CS) - Light (SD): Crossing barrier (R) > Reduced Fear (SR) Shock then disconnected: - Dogs jump barrier for 100s of trials to avoid shock BUT response should extinguish because CS occurred without US Herrnstein & Hineline (1966): - Rats were placed in Skinner box - Electric shock delivered randomly (i.e probability = 0.3 for every 2 second period that elapsed) - Probability of shock reduced from 0.3 to 0.1 if lever was pressed - Rats could NOT avoid or escape shock BUT just reduce number of shocks received - Most rats learned task and kept lower rate of shock probability Note: Is a problem for two-factor theory because avoidance learning is occurring in absence of CS - Avoidance learning can be explained by JUST the reduction in shock rate

One-factor Theory: - Avoidance is negatively reinforced by lower rate of aversive stimulation to which it is associated

- Reduction of aversive stimulation accompanying avoidance maintains avoidance Note: Which theory is correct depends on nature of the situation - Several processes seem to be involved in avoidance learning

Problems for Theory: - Fear is a necessary component BUT fear reduces with experience

The Fearlessness Problem: Fear and avoidance are NOT as firmly linked as the theory believes: - According to theory, fear provides motive to perform avoidance response - Early in experiments dogs would exhibit various signs of fear when tone was presented - BUT once avoidance response is well learned subjects responded without apparent fear

Kamin, Brimer & Black (1963): 1: Rats taught lever press in operant chamber for food 2: Auditory CS+ for shock > make avoidance in shuttle box until: 1, 3, 9, 27 avoidances in a row 3: Back in operant chamber CS+ (i.e from shuttle box) was presented 4: Measured suppression of lever press

Alternation of Behaviour (Yo-yo): Two-Factor theorists came back with this scenario: - Every successful avoidance puts CS on extinction - With extinction, fear drops so motivation to avoid decreases - Resultsin more shocks, strengthening of CR again and increasing avoidance response Note: BUT, we don’t tend to see this….

Persistence of Avoidance: - Cognitivists believe avoidance responding is based not on fear but on subjects’ expectation that response will avoid shock - During initial training, when warning stimulus is followed by shock, it is assumed that subjects form an expectation that shock will occur when stimulus is presented - When animal eventually jumps over barrier to avoid shock, a new expectation forms i.e The shock does NOT occur if response is made - Next time warning stimulus is presented, animal will recall both expectations and because it prefers NOT to be shocked it will perform response that produces this outcome

Cognitive explanation of avoidance learning based on expectations: - Thus, fear has little role in this theory as it can therefore account for shortcomings of two factor theory Note: First, regarding disappearance of fear during training - Two factor theory assumes once avoidance response is learned and warning stimulus is NO longer followed by shock, fear conditioned to this stimulus will extinguish - But animals continue to jump because it still expects shock to occur if they don’t jump and prefers to avoid this outcome

Cognitive explanation of avoidance learning: - The difficulty of extinguishing avoidance behaviour this directly from a cognitive analysis The theory says that avoidance depends on two expectations: - In absence of a response, shock WILL occur BUT if response is made shock will NOT occur - Early in extinction, the dog holds BOTH of these expectations and therefore responds - When shock does NOT occur its expectation that responding will NOT be followed by shock is confirmed and it will continue to jump - With each new trial this expectation receives further confirmation, so tendency to jump should be strengthened

Response Prevention > extinction: - If analysis is right, then animals continue to respond during extinction because they never get chance to learn what would happen if they didn’t respond Katzev and Berman (1974): Trained rats to avoid shock in a shuttle box and then gave them 50 extinction trials during which the CS was still presented but shock No longer followed Control Group: Were allowed to jump over the barrier during this phase to terminate CS Response prevention/flooding group: Had barrier placed above the hurdle so that they could NOT jump

Katzev & Berman, 1974: - Rats that could control the termination of the CS during treatment responded significantly more often during extinction than rats that could NOT - The probability of responding during extinction was also a decreasing function of the duration of CS exposure during treatment

Cognitive theory of avoidance: - The fact fear plays minor role in this theory allows it to explain continuation of avoidance responding in the absence of fear Note: BUT for same reason it has difficulty explaining evidence that fear does influence avoidance - In Rescorla and LoLordo, presenting a stimulus which had previously been paired with shock doubled subjects rate of responding (i.e even though they were already avoiding shock effectively)

Other factors to consider: The response problem One difficulty arose when experimenters tried to teach rats to press a bar to avoid shock: - We know rats are very good at pressing bars to obtain food - They can also learn to jump over a hurdle or run down an alley to avoid a shock - So if bar pressing is an easily learned response and avoidance of shock is a very powerful reinforcer then you may think that it is easy to train rats to press a bar to avoid shock Note: Yet hundreds of trials are NEEDED to learn to press a bar to avoid shock, and many never do

Bolles and SSDRs: - Each species has different innate responses for coping with danger, he called these Species Specific Defense Reactions (i.e SSDRs)

The particular SSDR that occurs depends on: 1: The nature of the aversive stimulus 2: The response opportunities provided by the environment

SSDRs: Innate responses, evolved: - If an effective means of escape is available animal is most likely to flee for cover when it encounters stimulus - Without a familiar escape route freezing will be predominant defensive response - SSDRs predominate in initial stages of avoidance Hierarchy: - If SSDR works keep it, if NOT try the next one

Summary: - Positive punishment motivates escape behaviours which are strengthened via negative reinforcement - Avoidance learning involves avoiding versive event before it occurs - Learned avoidance can take a passive or active form Note: There are a number of theories of avoidance Two Factor theory of avoidance: - Classical and operant conditioning are involved in avoidance learning - Avoidance is driven by an escape from fear NOT prevention of aversive event Cognitive theory of avoidance - Avoidance responding is NOT based on fear BUT on subjects’ expectation that response will avoid an aversive stimulus Species Specific Defense Responses (SSDRs): - Innate responses to threat/fear - Some avoidance behaviours are more easily learned than others...