L7 Acquisition of Fear Memory Flashcards
General info about anxiety disorders
- Anxiety disorders are very common
- 1 out of 14 people worldwide affected by anxiety disorder
- Prevalence of specific phobias 3-15%
- Major impact on daily life
Which part of the diathesis-stress model do we focus on?
The early learning experiences which relate to associative learning where you learn the relationships between stimuli, the environment and occurances
Picture 1
Who was Pavlov?
Most famous for his work on classical conditioning
- dog, bell, salivating
- Based on the principles he introduced, negative form of conditioning emerged - classic fear conditioning
Who was the leading figure in classic fear conditioning? What did he do?
John B. Watson conducted the earliest work in which he experimentally showed that fear conditioning is a mechanisms through which humans can acquire fears of intially neutral objects
- experiment ‘Little Albert’
What is the ‘Little Albert’ experiment?
Highly unethical experiement with little Albert who is 2 year old. John Watson and his assistant Rosalie Rayner made a terrifyingly loud noise whenever Albert began to play with a white rat. In this way, they trained the infant to fear the rodent.
What are the elements of conditioning in the ‘Little Albert’ experiment?
- Neutral stimulus (NS) – white rat
- Orientation response (OR) – looking, feeling (innate, initial, natural response in the case of seeing something new)
- Unconditioned stimulus (US) – aversive loud noise
- Unconditioned response (UR) – startle reflex, fear (natural response to the US – very normal in case of something bad happening- nothing to do with learning)
- Conditioned stimulus (CS) – white rat - stimulus paired with or perceded by the aversive outcome (US) and we learn that this stimulus is consistently paired with the outcome which results in the development of the CR
- Conditioned response (CR) – startle reflex, fear (often very similar response to the UR)
- The CR generalizes to other animals and objects that share feauters with the initial CS (rabbit, furry seal)
- Picture 2
The basics of fear conditioning
The very normal and natural response (UR) to an aversive outcome (US) transfers to a neutral object (CS) that was previously paired with that outcome and becomes CR
- Sometimes even after one occurrence of US and NS is enough to produce CR
What is learned during conditioning?
Fear response (UR) shifts from aversive stimulus (US) to previously neutral stimulus (CS)
- Original view: This happens through the CS-CR association
- Contemporary view: This happens through the CS-US association - CS is linked to a mental representation of the US (aversive event) = this is what drives fear conditioning
What is deconditioning?
Exitinction learning
- Precursor of exposure therapy
- Pioneered by Mary Covers Jones who reasoned that if we can learn that the neutral stimuli are linked to the aversive stimuli, we can probably also unlearn this
- exposed to CS presented without the US
What is counterconditioning?
Giving a reward after the CS so that the CS is now paired with something positive
- M. Jonas experimented with ‘Little Peter’ who was also conditioned to feat the white rat. However, to demonstrate counterconditioning, another experiment was done where he was repeatedly rewarded after seeing the white rat (CS) to link it with something positive. His approach behaviour increased over several exposures (fear decreasing)
What is a model that was proposed to explain why do people not learn that CS is harmless hence making the fear persist?
People have multiple opportunities where they encounter the CS without the negative events but despite that the fear still persists
- Mower developed the Two-process model of learning which states that the acquisition of fear (learning part - neutral stimulus → feared stimulus) is occuring through the process of classical conditioning, whereas the maintainance of fear is driven by operant conditioning
What is operant conditioning?
Operant conditioning is similar to classical conditioning but with behaviour
- behavior that is rewarded is likely to be repeated, while behavior that is punished is prone to happen less
How is fear maintained through operant conditioning?
Avoidance of CS which results in fear reduction (reward = feels good to not be scared) which then results in more avoidance of CS and this creates a vicious cycle
- great for reducing fear but it prevents the person to do the things they want to do but most importantly it prevents them from learning that the CS is harmless
How does fear conditioning look like in animal research?
- Fear learning CS -> US
- Animal in a cage with floors that give shocks to the animal which scares it and there are loud speakers, light and other environmental stimuli which become the conditioned stimuli
- They watch/listen to the environmental stimuli (CS) and then they get a shock (US)
- Often one pairing of CS and US is enough for the animal to have the CR after a second presentation of the CS
- index of the CR - freezing (behaviour animals do when they are scared)
How does fear conditioning look like in human research?
- People are showed different images (CS) which are initially neutral (or fear relevant stimuli, e.g. spider - easier to condition) and then given electrical shock (US) which is not always painful but aversive
- Very similar procedure to the animal research which is great because it’s comparable
- CS - US elicits the CR which can be measured in two ways
How is the conditioned response measured in humans?
- Subjective: US expectation (scale 0-100) - not necessarily fear response, rather conscious cognitive learning
- Physiological: startle response, skin conductance (on fingers - how much you sweat), heart rate
What is the startle response?
- Startle response is universal startle reflex to loud noise (104 dB, startle probe)
- Protective function with motor responses, e.g., eyeblink reflex - natural response
- Eye blink reflex measured with electromyography (EMG) of muscle activity under eye - the muscles contracts and the contractions we can measure and hence quantify the startle response
- Startle reflex is strengthened when someone is tense, i.e. when we are expecting danger
↪ Anticipation of something bad happening, threat system activated = eye blink (startle reflex) gets more intense - picture 3
- The startle probe is the loud sound (not the US, the US is the shock) and the startle reflex is the eye blink
What is differential fear conditioning?
- Show two different pictures and one has high probability that it will be followed by US (e.g. shock)
↪ You can play around with the probability – to increase/decrease uncertainty - CS+ followed by the US and CS- never followed by the US
Why do we need the differential fear conditioning?
To make sure that it’s not a natural response to the spider but rather a feared response to the specific picture = so learned response
- to make sure that there is CS - US association
- we want to be able to compare CS+ and CS-
- picture 4 - if we only looked at the CS+ line, we wouldn’t say this person developed fear, but if we compare it to the CS-, we see that CS- decreases due to habituation but CS+ stays high due to learned fear achieved by CS-US association = differential response
- Differential response is operationalised as the difference between the CS+ and the CS- response → this difference is the fear response
How does a fear conditioning trial look like in the lab?
Picture 5
- The loud sound is presented with every picture (CS+ and CS-) because that is to induce the startle response that we want to measure and it’s also presented in between the trials for really baseline response to the sound
- After first CS- then CS+ is presented and paired with a shock (US) etc…
- Always multiple presentation of CS and US because just one pairing is not enough = the shock (US) is not as negative as in rats or in naturalistic setting
What is an example of a design of the fear conditioning process at the trial level?
- Picture 6 - this design can be tailored to what exactly you want to measure
- CS+ presented for 8 sec + US expectancy (ask them) and skin conductance are measured, then right before they get the US startle probe is presented
↪ right before because there is a temporal expectation - they expect it to be at the end after 8 sec and so that when they are at their most tense, the startle probe is presented so that they can measure their startle reflex - After maybe 10 sec, the CS- is presented or another CS+
- And then the reaction to CS+ vs the reaction to CS- is the fear response
- Sometimes the startle probe is presented out of nowhere = noise alone trial
How does the US expectancy look like in CS+ and CS-?
- Picture 7
- They ask them how much they expect it and they rate it on a scale from 100 (absolutely) to -100 (absolutely not), 0 = I don’t know
- At the beggining 0 because they don’t know what to expect
- For CS+ it goes up → how high it goes depends on how often you let the US follow the CS+
- For CS- it goes down
- We can see that it’s a learning progress → CS- is safe since it’s never followed by the shock
- After conditioning, the US expectancy is higher for CS+ than for CS-
- The expectancy of US after CS+ doesn’t neccessarily have to increase but if you don’t do anything (CS-), it decreases → this means that the fear response to CS+ has developed
How does the startle response to the startle probe (noise alone) look like in a graph?
- Picture 8
- Present the startle probe when there is nothing on the screen
- Only reaction to the sound - it starts low because there is no expectation of US at all and then at the end CS- and noise alone (NA) are on the same level which means that there is no fear for CS- and NA but there is fear for CS+
- From this we can causally infer that learning took place between CS and US
How does fear conditioning look like in the brain?
- Amygdala is crucial for fear conditioning (at least in animals)
- Picture 9
- The activity that the CS generates is bound to the activity that the US generates in the lateral amygdala (LA)
- The behavioural response is also driven by the central nucleus of the amygdala
- The process: LA receives input from the CS and US and the LA integrates these signals meaning that US driven activity automatically spikes the CS driven activity in the amygdala
What other parts of the brain are included in fear conditioning?
- Hippocampus - contextual fear conditioning
- Pre frontal cortex - regulating the fear response
↪ the biggest differences between healthy and unhealthy people - PFC important for inhibiting the automatic associations that are formed which may be functioning to a lesser extent in clinical population
How can fear conditioning be studied in the human brain?
- Through fMRI
- While in the scanner, they perform a fear conditioning task and you can see that the amygdala gets activated more when they for example view the CS
What are the limitations of the fMRI studies?
- Despite observing higher activation of amygdala during fear conditioning, the fMRI scanning is very expensive so it was always done on a limited number of participants
- When meta analysis of the fMRI data was done they found no convincing evidence that the amygdala is involved in human fear conditioning
What does the unconvincing evidence from the meta analyses of the fMRI studies mean?
- We are not as similar to animals in regulating/experiencing fear in the brain as we initially thought
- We don’t have the equipment and technology yet to measure the activity in the amygdala related with fear conditioning that is actually taking place
↪ In animals, we can measure the excitatory and inhibitory neurons that get active/deactive when CS is shown
↪ In humans, we can only measure the average activity of the amygdala and not the precise neurons
What does the traditional learning theory posits about the etiology of anxiety disorders?
Traditional learning theory considers classical conditioning as a model for the etiology of anxiety disorders
- Stimulus learning - the UR transfers to the CS
- Is this correct?
What are the criticisms/problems of the traditional learning theory?
- Direct US experience is not necessary for fear learning
- US is not sufficient for fear learning
- Selectivity of phobias
- CR is not equal to UR
What is the first problem of the traditional learning theory?
Direct US experience is not necessary for fear learning
- The traditional learning theory posits that the CS-US pairing experience that results in fear
- However, this is not always the case
- Many phobics cannot remember a traumatic event (US) - which isn’t that big of an issue since the pairing could have happened, they just don’t remember it
- But the real issue is that some phobics fear stimuli they have never encountered (e.g. snakes)
How does the contemporary learning theory respond to the first problem?
There are different pathways of fear learning
What are the different learning pathways of fear learning?
- Direct learning (Trauma) - experiencing that CS results in US and the US is unpleasant and therefore the CS becomes unpleasant or fearful
- Vicarious learning - seeing something bad happening to someone else (observational learning) or copying fearful behaviour even though not knowing why (modelling)
- Information transfer (learning by instruction) - reading about a bad experience
What is the experimental evidence of these three pathways of fear learning?
Conditioning phase with CS+ and CS- → done in three groups:
- Receiving shock (US) after CS+ (direct learning)
- Watching video with other person receiving shock (US) after CS+ (observational)
- Instruction that shock (US) will follow CS+ (information transfer)
In part 2 - test for CS+ and CS- without any further instructions
What are the results from the experiment on the three pathways of fear learning?
Picture 10
- The response is larger for the CS+ in the direct learning = fear developed
- The overall response for the observational learning and the information transfer is lower than the direct learning but they still develop skin conductance response to the CS+
- Conclusion: fear learning not only through direct experiences but also indirectly
What are the clinical implications from this experiment?
- Fear also learned by observing fearful behaviour of parents or significant others
- Children of parents with anxiety disorders may be vulnerable to develop anxiety disorders
- Just by reading something, you could become fearful of something
What is the second problem of the traditional learning theory?
US not sufficient for fear learning
- Many people who undergo a traumatic experience (US) do not develop a phobia or anxiety disorder
How does the contemporary learning theory respond to the second problem?
Individual differences effect what is experienced and learned about traumatic events
What individuals differences are there that are related to fear learning?
- Individual differences in genetic predisposition and psychological traits
↪ Many genes but especially: low expression of 5-HTT gene
↪ High trait anxiety = measure of how anxious you generally are (this can be impacted by previous experiences)
↪ Behaviourally inhibited (excessively timid, shy…) -
Latent inhibition - prior neutral experiences with a stimulus (CS) reduces amount of fear conditioning when paired with an aversive event (US), i.e., protective factor
↪ E.g. more neutral experiences with a dog makes it less likely to develop fear of dogs after getting bitten than when only the aversive event is present and no prior experience with dogs
↪ Applied at dental practice - neutral experiences at the dentist reduce risk of dental anxiety when having a subsequent aversive dentist experience - Higher sense of control as a child, stronger sense of mastery over their environments → less frightened by novel and frightening events = protective factor
- Having control over the traumatic event → far less fear is conditioned when the aversive event is escapable than when it is inescapable
What is the third problem of the traditional learning theory?
Selectivity of phobias - Fear of spiders/heights much more common than fear of something that would be more rational e.g., cars
How does the contemporary learning theory respond to the third problem?
-
Preparedness: fear for prepared stimuli (important for evolution) easier to learn and more difficult to extinguish
↪ It doesn’t mean that the fear is innate but because they are evolutionarily important, we are more likely to develop fear of such - Evidence showing that conditioning with “fear-relevant” stimuli (e.g., picture of spider) is faster than fear-irrelevant stimuli
-
Belongingness - Certain CS-US combinations (between sensory modalities) more easily associated than other combinations
↪ Taste (CS) and nausea (US) easier than taste and pain
↪ Sound (CS) and pain (US) easier than taste and pain
What is the experimental evidence of belongingness of certain stimuli??
Rats exposed to tone (CS1) and fresh water (CS2) in two separate groups
- G1: sound and water → nauseating radiation (US1)
- G2: sound and water → shock (US2)
Part 2 (test) - Test of avoidance response (CR)
- G1: avoidance (CR) of water (but not sound)
- G2: avoidance (CR) of sound (but not water)
Purely on conditioning you would expect the same level of fear response to both stimuli because it’s not like one is perceeding the other = match between sensory modalities matters
What is the fourth problem of the traditional learning theory?
Traditional learning theory on what is learned during conditioning:
- Association between conditioned stimulus (CS) and conditioned response (CR): CS – CR association
- Conditioned response (CR) often identical to initial unconditioned response (UR)
Problem… the CR does not always equal UR!
- E.g. During conditioning with rats: CS+US (shock) the UR is jump, heart rate rises which is probably a manifestation of pain but after conditioning when CS is alone, the CR is freezing, sitting still, muscle tension, heart rate decreases which is probably a manifestation of fear
→ This makes it really difficult to conclude that it’s really the CS-CR association
How does the contemporary learning theory respond to the fourth problem?
- Association between CS and cognitive representation of US (CS-US) → by viewing the CS we mentally represent the US
- Feared stimulus (CS) predicts occurence of catastrophe (US)
- CR (e.g., freezing) prepares the body for impending catastrophe (US)
- Makes sense from the cognitive perspective however we don’t always see that it’s the expectation of US that is the same as the CR = can have very low expectation but still a very high CR and CR can also occur without any cognitive expectation of what’s gonna happen
- Most likely, there are actually more pathways for fear learning
↪ more automatic driven: stimulus-response associations
↪ more cognitive: stimulus-outcome association in which the CR prepares the body for the US which may or may not come
What is an example where CR doesn’t equal UR?
In panic disorder, CR can be two different reactions
- panic: intense fear, palpitations, sweating, feeling of suffocation
- anxiety: anticipation anxiety of having another panic attack
One can lead to the other and one can occur sometimes and the other not but they are two different responses to the enviornmental stimuli or the intraceptive bodily stimuli
Model of panic attacks - picture 11
- The cues (CSs) can result in two types of responses: anticipation anxiety (CS activates the cognitive representation of the US) and panic reaction (automatic stimulus response learning)
CS-US association between cues (CSs) and anticipated catastrophe (US; going mad, loss of control etc.)
- Anxiety disorders: Feel, think, and act as if feared stimulus (CS) is followed by a catastrophe (US)
Fear learning can be very adaptive. When may initially appropriate fears turn into maladaptive or irrational fears?
When…
… they persist when threat has passed
… generalize to other safe situations or stimuli
… when accompanied by avoidance behaviour that interferes with daily functioning
This is what happens in anxiety disorders
How can we use fear conditiong to better understand the development and maintenance of fear?
Fear acquisition in the lab is a strong and adaptive manipulation but we can also model or measure more maladaptive learning processes:
- Reduced safety learning
- Generalisation
- Avoidance
- Resistance to extinction
Modelling maladaptive processes can help to better understand individual differences in fear learning and how to change these processes
Studies that measured these maladaptive processes
How do people high on anxiety trait react to safety learning?
Interestingly, reduced differential fear learning is linked to anxiety disorders
- Participants that score high on anxiety often show similar rates of fear acquisition to CS+ compared to controls → might be understood as less fear but it’s not actually the case because they show elevated responses to the CS- (picture 12)
- This means they are worse at differentiating safe and threat stimuli = not good at safety learning → impared discrimination learning
- This differentiation predicts anxiety development six months later (prospective validity)
How does generalisation of fear conditioning look like in anxious participants when modelled in a lab?
Generalization measure → picture 13 where the circles from CS- get more and more similar to CS+
Phase 1: conditioning with CS+ and CS-
- Shock (US) after one circle (CS+) but not after another circle (CS-)
Phase 2: generalization test
- Presenting circles of intermediary sizes (GSs), as well as CS+ and CS
- Results: In all participants, the more the stimulus will start to look like CS+, the more fear they will show (picture 14)
- However, anxious participants have a steeper generalisation slope
- Elevated responding to generalization stimuli that resemble CS- was associated with higher levels of anxiety at follow-up
What is the conclusion from the study on differential learning and generalisation in axious participants?
- Individuals who failed to differentiate between the CS+ and CS− during acquisition may have lacked discrimination during the subsequent test for generalization, which in turn resulted in diminished responding to stimuli resembling the CS+ and heightened responding to stimuli resembling the CS−
- Individual differences in discrimination learning and generalization are vulnerability factors for the development of anxiety
How does avoidance of conditioned fear look like in anxious participants when modelled in a lab?
Phase 1: typical process of conditioning with CS+ and CS-
Phase 2: Allow participants to avoid the US - press a red button to avoid the shock (avoidance response - still adaptive and everyone will press it)
However, to model the avoidance behaviour:
- Offer reward for not avoiding (e.g., money)
- No longer present the US even if not avoiding - extinction phase
Results: (picture 15)
- Everyone will press a button if no reward is offered
- Anxious participants are more likely to avoid US even when offered reward therefore less likely to learn that US no longer occurs - they never reach the extinction phase
What is extinction learning?
After undergoing conditioning, you start to present all CS without the US
- Do this multiple times and people learn that the US no longer follows the CS+
How does extinction learning look like in anxious participants?
- Retain higher responses to the CS+ during extinction
- Show increased differentiation between CS+ and CS- throughout extinction
↪ although not all studies show this
Practice question: Suppose you have been bitten by a dog and you develop a dog phobia. How can your phobia be explained in terms of fear conditioning?
Learned association between dog (CS) and bite (US) through direct learning
Practice question: To what extent would your phobia then be irrational?
- Generalization of fear to other potentially safe stimuli (other dogs)
- Avoidance behaviour (e.g., no longer going to the park)
- Not adapt behaviour after more pleasant encounters with dogs
Learning objectives
- Explain and compare early learning approaches (classical conditioning and operant conditioning) to the development and maintenance of anxiety disorders.
- Describe the underlying neurobiological processes of fear learning.
- Describe and compare research methods for studying fear learning in animals and humans.
- Describe the criticisms of classical conditioning and explain how these criticisms are addressed by a contemporary learning theory approach.
- Explain and compare principles (phenomena) of a contemporary learning theory approach to fear learning (e.g., observational learning, safety learning).
- Assess the role of contemporary learning theory principles (phenomena) in various anxiety disorders and apply them to a clinical case.
