TJ&DM Flashcards
1
Q
‘Gastrophysics’ borrowed its name from…
A
Psychophysics - the field that investigates the relationship between objective qualities of stimuli and experienced sensory qualities
2
Q
Preconceived perceptions (gastrophysics)
A
Pre-conceived perceptions can enhance the experience of food, but it can backfire if it doesn’t meet expectations
3
Q
Taste (gastrophysics)
A
○ Basic tastes come from the tongue: sweet, sour, salty, bitter (maybe umami = delicious?)
○ What about fruity, smoky, herbal, burnt, and all the many others?
○ These are flavours, and they emerge when smell is added
○ The classical tongue mapping is probably not true - there are relative differences, but great variability between people
4
Q
Oral referral
A
Smell adds to the taste experience, but we do not realise this but “project” the experience to the tongue
- orthonasal
- retronasal
5
Q
Orthonasal
A
Sniffing external aromas
6
Q
Retronasal
A
Aromatic odour is received through the back of the mouth into the back of the nose during food consumption
- this is where the rich perception of aroma comes from
7
Q
Key Gastrophysics message
A
® There are still individual differences and some people will prefer black and some will prefer white
® But when there appears to be a generalised trend that replicates, it can be useful in the real world
8
Q
Is the science of gastrophysics sound?
A
Pay attention to the methods and statistics:
§ What kind of statistical tests are used, and are they appropriate?
§ How many tests are conducted, and do the authors control for multiple comparisons?
§ How many participants were tested, and what is the reported effect size (if any is reported)?
§ Do the results replicate across studies?
What other factors could have biased participants (and the results)?
9
Q
Making decisions after consuming (Bode)
A
® Idea that if you are waiting for a large reward you will be impatient and think it is 2 secs quicker
Actually found that after a large reward, the estimate is longer but after a small reward it is shorter
□ Turns out high calorie consumption actually led to less patience, which is not what they expected
10
Q
Anxiety vs Fear response
A
They share many common physiological features, but they can be differentiated:
Fear responses are elicited by specific stimuli and tend to be short lived
Fear responses decrease when a threat has been removed or dissipated
Anxiety may be experienced in the absence of a direct physical threat and typically persist over a longer period of time
11
Q
Extinction of a conditioned fear response
A
e.g. presenting the tone repeatedly without the shock, resulting a gradual decrease in the conditioned fear response
Extinction is the basis for graded desensitisation training in psychological practice
12
Q
Systematic desensitisation
A
Developed in the 1950s by Dr Joseph Wolpe
attempts to replace an anxiety or fear response with a relaxation response through a classical conditioning procedure
You gradually associate, through repeated pairing, a fear-arousing stimulus with a state of relaxation, in a series of graded steps
Really important that you replace the anxiety or fear with relaxation response before upping the stimulus
13
Q
3 ways conditioned fear returns
A
§ After the passage of time (Spontaneous recovery)
§ Changes in context (Renewal effect)
§ Stress (Reinstatement effect) - one of the more common ones
interpreted as evidence that the original ‘fear memory’ is not deleted, or erased, but rather inhibited during the extinction phase of conditioning experiments
14
Q
Fear conditioning neural network structures
A
Include the amygdala, hippocampus, ventromedial PFC, dorsolateral PFC, and striatum
15
Q
Cognitive regulation (neurology and 2 strategies)
A
the dorsolateral prefrontal cortex (dlPFC) regulates fear expression through projections to the vmPFC, which in turn inhibits amygdala activity
Strategies include:
§ Reinterpret the significance of an event (cognitive reappraisal)
§ Focus attention on the less fearful aspects of a situation (selective attention)
16
Q
Active coping (basic idea)
A
argued to be the most commonly used ‘untaught’ method we use to regulate our emotions
strategies involve an awareness of the stressor, followed by attempts to reduce the negative outcome
Learning an action to avoid a fearful event or diminish the fear response
17
Q
Active coping (neurology)
A
○ During active coping, info from the LA (lateral nucleus) is routed not to the CE (central amygdala), which drives fear expression, but to the B (basal amygdala), which in turn projects to the striatum. The striatum is thought to reinforce instrumental action taken during escape- from- fear or avoidance learning
18
Q
Reconsolidation period
A
○ Diminishes conditioned fear expression through alteration of the original CS-US association stored in the LA
Linked to the idea that immediately after learning there is a period of time where a memory is fragile, less permanent
○ During the consolidation period, you actively seek to disrupt the formation of the memory
○ Or, if you are past the consolidation period, you look to modify or inhibit, but not eliminate, the memory, by actively retrieving the memory
○ The act or retrieval makes the underlying memory trace fragile again, called the reconsolidation period
- potentially allowing you to block the memory completely
19
Q
Kindt et al. (2009), Schiller et al. (2010) and Debiec and LeDoux (2004)
A
were the first to show that in humans you could use propranolol to block the return of fear
Still not clear what the mechanism for propranolol’s effects are
20
Q
Anxiety and Fear
A
Several prominent theories of Anxiety Disorder propose that dysregulation of the neurocircuitry of conditioned fear may be central to the disorder
trait anxiety is associated with heightened amygdala activation and elevated fear expression
Anxiety also impairs extinction learning and retention, as well as the regulation of emotional responses via cognitive strategies
○ These deficits appear to stem from impairments in the regulation by prefrontal cortex of the amygdala ○ For example, anxious patients exhibit reduced prefrontal activation during or before fear extinction, and require heightened prefrontal recruitment to successfully reduce negative emotion with cognitive reappraisal
21
Q
Cognitive effects of Anxiety (info processing biases)
A
two key cognitive information processing biases:
- A bias to attend toward threat-related information
- A bias toward negative interpretation of ambiguous stimuli
22
Q
Anxiety and Attention Threat Study (ambiguous faces + homphones)
A
For stimuli with more than one potential interpretation, people with anxiety have a tendency toward a more negative perception
§ Ambiguous facial expressions (Richards et al., 2002)
§ Face-voice pairings
§ Verbal homophones
□ Die or Dye
When evaluating the probability of future life events, individuals with anxiety will judge the probability of negative outcomes as far more likely than non-anxious controls
23
Q
Anxiety and economic decision making
A
hyperactivity of the amygdala, while attending to, evaluating or anticipating negative stimuli, contributes to heightened cognitive and affective responses to a potential threat stimuli
Prefrontally mediated cognitive and affective regulation processes also appear to be impaired in anxiety, reducing the ability to modulate these pre-existing tendencies
24
Q
Uncertainty
A
Across species, stimuli that are unpredictable will elicit greater anxiety
people with anxiety will show threat-related information processing biases that alter their decision making
Humans as a general trait will be risk averse
For all, the amount of attention given to the aversive choice option predicts the likelihood of avoiding that option
Importantly, when the decision paradigm requires the anxious participant to make a choice for another person, they are less risk averse
25
Risk (Dohmen et al. 2011)
§ Analysed the relationship between the response to the 10 point risk question and other variables in 22,000 participants
§ Women were less likely to take risks than men
§ Increasing age was associated with less risk tolerance
§ Greater height was associated with greater risk tolerance
§ Having a mother with higher academic achievement, and to a lesser extent a father, increased risk tolerance
26
Anxiety and risk avoidance (physiology)
Evaluating your own physiological response to a risky choice appears to be critical to risk avoidance in anxiety
If your evaluation of an option includes increase sympathetic nervous system responses (e.g. BP, HR, SCR increases), then anxious participants are more likely to avoid these option
Hence why casinos sometimes give free alcohol to increase likelihood of risky decisions
26
Ambiguity
multiple possible outcomes with unknown probabilities
§ Like with the attentional bias effect, decisions between ambiguities options will illicit greater anxiety, and people with anxiety will show greater avoidance of ambiguous decision making - e.g. choosing a clear urn to an opaque one
§ In ambiguous decision making situations, people with anxiety will overestimate the probability of negative outcomes, and their subjective cost
Car vs Train example
27
Framing & Anxiety
§ Anxiety, train and clinical, is associated with greater framing effects in their decision making
§ Argued to be driven by loss aversion, where the increased sensitivity to the certain loss engenders an avoidant response, increasing risk taking behaviour
28
Loss Aversion
§ The degree to which avoiding losses is prioritised in comparison to achieving equivalent gains
§ Anxiety, of all forms, is associated with increased loss aversion
29
Emotions shape decisions via the depth of thought
negative emotions increase the depth of information processing, signalling when a situation demands additional attention
§ Not always a good thing
Increased focus on anchoring effects
relative to neutral state participants, angry participants allocated less to welfare recipients, and sad participants allocated more
- an effect that could be eliminated when given greater unrelated cognitive demands
30
Intervening to reduce emotion effect (4 strategies)
Time delay (simplest)
Suppression (ineffective)
Reappraisal (one of the best strategies)
Increasing awareness of misattribution (can often be quite difficult)
31
Burden of health transition
□ 90's - most burden was from communicable diseases
□ 00's and towards 20's - burden is shifting from communicable disease to more non-communicable/behaviour-related diseases
® Behaviour is a huge component
32
Habit (definition)
§ The exhibition of a learned behaviour that was insensitive to changes in regarding outcomes
33
Common features of habit learning (human and animal)
§ Repeated responding that forms context-response associations in memory
§ Automatic habit performance that is relatively insensitive to changes in the value of the response outcome
- important distinction to consider is that humans engage in much repetition of behaviour
§ E.g. High levels of repetition in daily activities
□ Probably about 40% of what we do every day is habitual
34
Habit creation (vs. repetitious behaviour)
Typically arise due to an interface with goal-related behaviour
§ Goals direct human action by providing a definition of a desired outcome
§ Repetitious behaviour is usually, but not necessarily, goal related and can be distinguished from habit behaviour because it does not persist when the value of the repeated behaviour is absent
35
Automaticity
○ All habits have a level of automaticity, but not all automatic responses are habits
§ E.g. Startle reflex is an automatic response, but not a habit
§ E.g. Nail biting - negative reinforcement by self-soothing or distraction
□ Also alcohol at social functions easing social anxiety
36
Difference between habits and other automatic processes
§ Priming
§ Classical conditioning
§ Non-associative learning
○ Habit automaticity is specific to a particular response or behaviour
37
Implementation intentions
Often referred to as automated goals
Will only influence behaviour insofar as they are consistent with someone's motivation level or intention
§ Getting people to commit/set an intention will increase the likelihood of otherwise non-habitual behaviours
□ E.g. The dentist asking you when you would like to have your next appointment
□ Don't work with automatic/frequent behaviours like seatbelts
38
Habit representation
Perception of the relevant context cue automatically activates the mental representation of the habitual response
Habits can be triggered/cued in deliberate or inadvertent ways
It is generally easier to break a habit if you are conscious of it
A habitual response is the cueing of mental representation that contains both the features of your response, but also the features of perceptual information that cues your response
39
Habit insensitivity
Habit performance is characterised by the insensitivity to outcomes
One consequence for this finding is that repeated behaviour over time will become more habitual and less goal dependent
Habit strength will also interact with behavioural intentions
Dual Process models argue that habitual behaviour exists because it offers a greater level of efficiency
Habits are a ready default setting for behaviour that allow us to respond, unless we are particularly motivated and able to engage in more deliberate and specific goal pursuit
40
Habit formation
Habits develop through instrumental learning
§ Reward responses will be repeated - Thorndike
Habitual responding continues to be influence by motivational processes
- Pavlovian Context Cues
- The motivational effect of these cues is not related to the value of the outcome
Habits develop most readily when rewards are provided on an interval schedule
- Most 'checking' behaviours are like this - e.g. Checking your phone for messages
- Also gambling and 'arguably' social media algorithms - interval schedules
41
Habit formation pt II
○ Repetition will not always lead to habit formation
§ If the repetition of a behaviour includes decision making, it will generally prevent, or slow, the formation of habitual behaviour
§ Deliberative decision making protects against habit formation even when you respond repeatedly to the exact same cue
Human Research has typically not seen a relationship between the extent of training (or repetition) and greater expression of habits
42
Habit circuit
converging evidence indicates that habitual behaviour is mediated by circuits that link cortical brain areas and the basal ganglia
§ The basal ganglia is a subcortical structure, that houses a collection of subcortical nuclei
§ Dorsolateral striatum appears critical to acquiring new habits
§ Dorsomedial striatum appears critical to goal-directed control
□ This info comes from animal research
43
Habits > Goal Pursuit (balance factors)
The balance between habits and goal pursuit can be influenced by a number of situational factors
§ Distraction
§ Time pressure
§ Task ability (or lack thereof)
§ Self-control
Habits will trump goal-directed behaviour when they lack the ability and motivation to engage in deliberative decision making
Acute and chronic stress will increase the reliance on habits
§ Time pressure, reduced self-efficacy = more likely to rely on habits
44
Habit and Awareness
○ People have some awareness of their habitual responses, but they have poor awareness of the cueing mechanism that activates the habit
One aspect of this lack of awareness is that participants with stronger habits will have a greater tendency to infer that behaviour was guided by their behavioural intentions (or goals)
Over time, goals can be formed as a consequence of habitual behaviour, rather than preceding it
45
Changing Habits (2 aspects)
○ Behaviour change techniques have difficulty changing habitual behaviour because the responses do not reflect a persons stated intention or desire
○ The major challenge is that habitual behaviour continue to be activated automatically by recurring environmental cues
○ In response, behaviour change techniques have focused on two aspects
1. Impede the automatic cueing of old, unwanted habits
2. Encourage the repeated use of new, desired behaviours to the point they become habitual
46
Behaviour change techniques to reduce interference from old habits
- Invoked thinking
- Inhibitory plans
- Attempting to link specific inhibitory plans to time or environmental cues that are linked to the habitual behaviour
- Exposure Management or habit discontinuity
47
Robbins et al. 1973, 1993 - Vietnam Vets
Of those who had reported being dependent, only 5% had relapsed within 10 months of return
® For context, normal heroin relapse rates are around 70%....
§ Taking them out of the environment to a completely different one with removed cues and stressors was highly effective (just naturally)
48
Encouraging Formation of Desired Habits (past)
○ Behaviour change techniques to encourage the formation of desired habits have not always been built upon the concepts of habit formation: repetition, stable context, appropriate reward schedules
- Repetition required can be extensive
- Reminders can reduce automaticity
49
Neuroimaging is...
- Expensive
- Complicated
- Correlational
○ Prior to doing a brain imaging experiment, ask yourself if you have questions about the brain, behaviour, both, or something else
50
Neuroeconomics
○ Neuroscience = the neural underpinnings of behaviour and mind
○ Psychology = science of behaviour and mind
○ Economics = science of choice
51
Economics aims
Most focused on understanding the exchange of goods, services, information, etc.
Thing to keep in mind:
□ The value of things is effectively just what someone is willing to pay for it - not a fixed thing
® Arguably nothing has inherent value
52
Learning from observation (predicting choice)
§ What people have done tells us they might do
□ Past behaviour is the best predictor of future behaviour
□ People don't really change that much...
§ People don't always have good insight into why they are doing things - this is where experimentation is useful
53
Improving predictions of consumer choice
- Know the market
- Know the buyer/seller history
- Know the traits of the buyer/seller
- Know the state of the buyer/seller
○ Traditional economics only permits us to observe
○ Behavioural economics permits us to ask
○ Neuroeconomics permits us to look at brain regions that we think compute value
54
Kahneman & Tversky
○ Most prominent integration of Psychology and Economics
○ Basic idea
§ The choices that people make are often not actually rational (as previously assumed)
--> S-curve
- Really have to look at the notes for this
54
Prospect Theory - Gains:
§ Certainty effect - we value knowing the outcome even if the questionable choice is technically a bit better
§ Possibility effect - the possibility of that much money is often enough to make the irrational choice
- Need to look at notes for the rest of prospect theory
55
How do we model attitudes (model types)
○ Model-fitting (Traditional vs. Bayesian)
§ Basically, traditional modelling yields a single estimate/figure (like the mean or regression or something) but with a Bayesian models you are given a distribution of values - a whole probability of possible answers
§ The beauty of it is that Bayesian models are more realistic because no one is ever the value given by the traditional models
56
By bother with brain if we can model preference (neuroeconomics)?
○ The brain is ultimately making the choice and therefore, should be a more direct predictor of choice
§ We may be able to make better guesses about the behaviours if we know which areas of the brain are involved
§ You are getting much closer to where the decision is actually being made
57
Neuroeconomics is integrative
○ Remember, we are integrating across three disciplines:
1. Economics - values and perceptions
2. Neuroscience - how the brain operates in the process
3. Psychology - representations/theories of what people do and methods to experiment
* His position - academia does a really bad job of combining the three and just view them as equal to their parts rather than greater than the sum of it parts
○ The combination of approaches allows us to build models of decision-making
58
Better Models of Decision-Making (Rangel et al. (2008))
Really important graph in slides
Representation --> Valuation --> Action Selection --> Outcome Evaluation --> Learning ---> (feeds back to first 3)
§ Representation - mental model - what are my feasible actions, my internal states, my external states
* Model of what your options are
§ Valuation - how much do I value each of these options/actions - particularly the value of how badly I do or don't want something
§ Action selection - the process of choosing which option you want to pick
§ Outcome evaluation - how did the action go for you?
59
Reminder to revisit Lecture 4
Not easily made into flashcards
60
Drivers of Panic Buying
- Perception of scarcity
- Fear of the unknown
- Coping behaviour
○ May be a mechanism to regain a sense of control
- Psychosocial factors
61
Findings of panic buying study
Appeal to morals message had the largest impact on purchasing intentions, compared to the no message control
scarcity message also saw a smaller reduction in intentions to increase purchasing, but this difference was not significant at conventional levels
- appeared to be the best at reducing expectations that products might run out.
62
Summary Week 6 (Panic Buying)
○ Use a moral or norms-based appeal to combat panic buying
○ Continuing to encourage supermarket purchase limits
○ Consider further testing to refine messaging
63
Reminder to Revisit Week 6
Lots of info on slides/images
64
What is different about pokies?
Can gamble fast - rapid and continuous
Argument that if we made them more like lotteries where you have to wait it would help the addiction rates
Losses Disguised as Wins (LDWs)
65
Losses Disguised as Wins (LDWs)
○ When you win less than put in in a single spin - so you have made a net loss, but it is presented as a gain
- Closely resembles genuine gains
- Evidence that people do mistake these events as genuine gains
66
Non-Gambler LDWs Study Findings
§ People significantly overestimate their wins when there are LDWs
Had peopel watch someone gamble in either condition and track gains/losses
□ In control, about half were perfectly accurate
□ In LDWs condition, they tended to count each LDW as a win
67
Problem Gambler LDWs Study Findings
§ Theories - by gambling a lot they might get wise to LDWs and understand them better
§ OR - they might be more susceptible and succumb to them more
§ BUT - they actually found there was no difference in the size of overestimation of wins between them and non-problem gamblers
□ Appears to be a consistent phenomenon among everyone
□ Very good grounds for regulatory intervention
67
Could LDWs contribute to gambling addiction?
Pretty strong case for yes
OPERANT CONDITIONING
○ Random/variable reinforcement has a more robust influence on habitual responses
§ More likely to continue gambling in spite of consistent losses
§ When you add LDW's you are cranking up the positive reinforcement and lowering the extinction rate
CLASSICAL CONDITIONING
○ The celebratory animations and music that is played with a LDW is the same as a true win
○ The conditioned association between this fan fair and true wins could be a contributing factor to the tendency to mistake LDWs for genuine pay outs
§ They could be classically conditioned to think they won
○ The music and lights could also be a reinforcer by themselves (secondary reinforcer?) - CHECK THIS
§ Conditioned stimulus --> secondary reinforcer
This is all under the assumption that the LDWs are genuinely rewarding
68
LDWs EEG Study Results
Essentially measured reward positivity
- signal was tracked to the reward centres of the brain (medial pre-frontal cortex or anterior cingulate
Found clear evidence that LDWs were eliciting more positivity than regular losses
- Provides evidence that LDWs could provide a type of reinforcement during gambling
69
Do we need to change LDW regulation?
LDWs may or may not count as being misleading, legally speaking
§ But colloquially, they definitely seem to systemically inflate the perceptions of positive gambling events
Does it pass the pub test?
○ Do you agree - "to what extent are pokies likely to mislead or deceive consumers?'
§ Intervention condition - read a newspaper headline (made up but realistic) that explains what LDWs are
§ 40% of the control group agree with the statement
§ 55% of the intervention condition agreed - significant
○ So it might be legally misleading or not, but is certainly seems like the everyday person agrees that they are misleading once educated on it
70
What is Computational Psychiatry?
explanatory bridge between altered cognitive function and neurobiological mechanism associated with mental illness
Uses computational modelling to formalise the mathematical relationship between symptoms, environment and neurobiology
Provides tool to aid diagnosis and identify the neurobiological causes of specific symptoms in individual patients.
71
Brain Imaging Techniques used in Computational Psychiatry
□ EEG - good temporal res.
□ MEG (magnetic) - less common than EEG, record magnetic activity rather than electrical activity - good temporal res
□ (F)MRI - play sounds, show pics, etc. And see where it is being processed in the brain - good spatial res
Diffusion imaging - looks at brain connectivity across different areas - white matter pathways (?)
Then computational modelling - e.g. brain connectivity, behavioural models
72
Computational Psychiatry methodological developments
○ People used to identify which areas were responsible for which functions and name them - e.g. Fusiform face area
§ But now in the last 25 years or so, we have realised that it is not any single area that does a function, but networks of areas
73
Mismatch Error
specific term relating to EEG and Oddball paradigm, and specific to a time window of 100-250 milliseconds
Our brains try to predict what is coming next in a sequence by extracting regularity from the environment
MME occurs when there is salient difference between expected and actual stimulus
74
Prediction Error Response (PER)
Broader term (than MME) that is not locked to a particular time nor specific to a paradigm
75
Why should we care/Relevance for psychiatry (PER)
○ Index of sensory memory trace
○ Index of learning patterns or regularities
○ Index of detecting violations or irregularities
○ These identified responses are reduced in schizophrenia - implications for diagnosis
○ (also relevant for computational neurology as it's reduced in coma and predictive of awakening)
§ Importantly - this response is predictive of waking up
76
Predictive Errors in Psychosis
Graph in slides
Basically, the deeper into psychosis/psychotic diagnoses you are, the less PER you display
Also mitigated PE in Schizophrenia, in those who later transition, and an index of functioning
Useful as a potential warning/diagnostic tool in a clinical setting
NOTE - So far this is all just a proof of concept
77
Reminder to revisit Week 9 (computational psychiatry)
Second half is very dependent on slides and graphs
78
Self-intoxication legal defence
○ You cannot use the defence of defence of self-intoxication when charged with a violent crime
§ This is particularly about classifying murder/manslaughter
§ The reason is that otherwise you could just want to kill someone, get pissed and do it, and then it wouldn't be your fault
79
3 Circumstances for substance-related psychosis
Substance induced psychosis
- transient disorder, substance induced, short term (no previous mental illness)
Emergence of psychotic illness
- (same as previous, but long-term), psychotic illness doesn't resolve
Previously diagnosed mental illness (e.g. Schizophrenia) then start using drugs (e.g. meth)
- Is it the meth causing the psychosis, or the schizophrenic psychosis leading to meth use?
○ Note - many people self-medicating for mental illness are polypharmacy - use lots of different drugs at different times
80
Meth use and violent behaviour (animals)
○ Acute doses in animals have inconsistently produced violent behaviour, low to moderate doses appear to augment aggressive responses, but the effects are not consistent across species or paradigms
mixed results are possibly confounded by the acute effects of meth on increased vigilance and psychomotor behaviour
○ Chronic methamphetamine exposure in animal studies (Sokolov and colleagues) has consistently demonstrated increased aggression and fighting behaviour
81
Meth use and violent behaviour (humans)
Acute doses of dextro-amphetamine, which is an oral medication used to treat narcolepsy and other conditions, show no increases in aggression
○ Chronic use has been prescribed to successfully treat aggressive behaviour in children with ADHD
○ 60+% of heavy chronic Meth users report problems with aggression and problems controlling violent behaviour
Australian data indicates that prevalence of violent crime amongst regular Meth users is comparable to other illicit drug using groups
82
Intoxication and crime
Almost half of assault offences and three quarters of offensive behaviour incidents involve intoxication
A review of crime statistics from 11 countries, 62% of offenders convicted of a violent crime had been consuming alcohol shortly before committing the crime
Alcohol was more than twice as likely to be a contributing factor to violent crimes, compared to non-violent crimes
The magnitude of violence was related to the magnitude of consumption
83
Four pharmacological links between intoxication and aggression
Psychomotor stimulant effect
Interrupted threat detection
Alterations of the pain system
Alterations of cognition
84
Psychomotor stimulant effect (intoxication)
□ Euphoria accompanying intoxication is rewarding, which is accompanied by increases in nervous system stimulation
® E.g., increased BP, HR, respiration
® Also increase sensation seeking, approach behaviours
® Confrontational and provocative behaviours
Approach behaviour is greater for alcohol rather than soft drink in alcoholics
§ Evidence that over time alcohol systematically changes the approach behaviour towards alcohol - reinforcing to the alcohol dependence
□ You can use approach behaviour training to try and reduce this, but training them to approach soft drink rather than alcohol
85
Interrupted threat detection (intoxication)
§ Intoxication can result in diminishing appropriate levels of stress and or fear
§ Animal studies demonstrate that alcohol infrequently heightens aggression in lab animals
□ specifically characterised by increased proximity to a threat
® Where they otherwise would avoid the threatening situation, they now get to close for that to be an option and they (at least feel) they have to fight
§ Other drugs that have an anxiolytic effect (e.g., cannabis) but are not linked to violence, also have a CNS depressant effect
§ Hoaken et al 2003 – showed that the alcohol-aggression relationship was mediated by the cardiovascular stress response dampening by alcohol
86
Taylor Aggression Paradigm
§ Subjects are placed in a situation where electric shocks are received from and administered to a fictitious opponent during a competitive reaction-time task
§ Physical aggression is indexed as the intensity of shocks administered to the opponent
§ Given that aggression is highly related to provocation, the TAP has low and high provocation conditions
§ Opponents are introduced at the beginning of the experiment and sit opposite throughout, they are same gender and race
§ Pain thresholds are individually assessed and factored into shock intensity
Results from study:
§ Alcohol significantly increased aggressive behaviour in comparison to the placebo, for men but not women
§ The effect of alcohol expectancies on aggression in men was rendered nonsignificant when statistically controlling for dispositional aggressivity
* “intoxicated aggression, in men, is primarily the result of alcohol’s pharmacological properties in conjunction with an aggressive disposition.”
□ Alcohol is necessary but not sufficient to induce aggression
You need to have the underlying aggressive disposition for the alcohol to make you aggressive
87
Alteration of the pain system (intoxication)
* In the descending limb (when your BAC is lowering) of alcohol intoxication pain sensitivity is decreased
* In the ascending limb (increasing BAC) of intoxication humans sensitivity to pain increases (Gustafson, 1985, Gray, 1982)
* The effect of alcohol is therefore thought to be paradoxical
□ Subjective experience of pain heightens self-reported irritation, annoyance, anger and aggression towards another, even if the source of pain cannot be attributed to them (Berkowitz and Thome, 1987)
□ In the descending limb of intoxication the analgesia experienced can block the painful aspects of aggression
example of people fighting in a night club after an earlier minor irritation with each other
88
Alterations of cognition (intoxication)
This whole literature was started to find a legal BAC limit for driving
* Early research examined a broad range of tests such as reaction time, visual and auditory acuity, hand-eye coordination, gross body movement, short term memory
* Threshold or effective blood alcohol concentration (BAC)
Three elements:
- Cognition
- Processing Speed
- Divided Attention
89
Cognition (intoxication)
○ Reviews indicate that most behaviour experimentally tested have a threshold BAC of 0.08%
* E.g. Simple reaction time
○ Higher-order cognitive function, such as decision-making, judgement, memory have a reliable BAC threshold of 0.06%
○ More complex (and likely valid) divided attention tasks that emphasise dual task performance, have a reliable BAC threshold of 0.05%
90
Processing Speed (intoxication)
○ Simple reaction time tasks showed intoxication slowed behavioural responses to stimuli
* Motor issue
* Flicker fusion task
* Subtraction task
○ Processing speed is the conflation of three stages of processing
* Perceptual, central, and motor
□ Early perceptual stages have been found to be least effected by intoxication
□ Central processing stages are most sensitive to intoxication
* Speed accuracy trade-off
□ An intoxicated person will purposefully slow down in order to maintain or improve accuracy
□ Makes it difficult to assess at what stage of processing the intoxication is having an effect
91
Divided Attention (intoxication)
○ Trade-off problems involving performance speed are generally avoided by the use of divided attention tasks
* Divided attention tasks
□ Assess the ability to performance two or more activities at once
□ For example a participant may be required to maintain a verbal working memory load while performing a stroop colour naming task
□ Or manually track a rotating target (pursuit task) while performing some auditory discrimination task (detecting differences amongst tones)
○ Performance in these tasks has been found to be highly vulnerable to alcohol intoxication, with threshold BACs of 0.05% or less
○ Performance of the individual task can often be unimpaired during intoxication, but when required to perform the tasks simultaneously performance decrease dramatically
* Relevance to driving capacity
92
Reminder to revisit Week 10
Look back at the studies and their findings
93
Week 10 (Intoxication) Summary
○ Evidence demonstrates that alcohol consumption:
* is associated with increases aggression in men but not women
□ the increase in men can be largely predicted on the basis of pre- existing aggression characteristics
○ Cognition is significantly impaired by moderate doses of alcohol
* Particularly important processes such as inhibitory control and dual- task performance
* The level of individual impairment predicts subsequent alcohol consumption, with the greater impairment of inhibitory control leading to greater levels of consumption
* The impairment identified in published research has not exceeded BAC levels of 0.09% and as a consequence we do not know whether a point is reached at which individuals have no control over their behaviour or their aggression
94
What is 'Neuromarketing'?
Loosely refers to the measurement of physiological and neural signals to gain insight into customers' motivations, preferences, and decisions, which can help inform creative advertising, product development, pricing, and other marketing areas
Assumption - a person is just their brain, if we can understand the brain we can understand the person and what they will do.
95
7 P's of Services Marketing
People
Physical evidence
Process
Promotion
Product
Price
Place
All flow into the next
□ All things we may be able to modify or learn about that could influence how the customer engages
§ Idea of the customer being at the centre and thinking about what they will do
96
Specific periods of behaviour change (neuromarketing)
There are particular periods in your life where routines fall apart and buying habits are in flux
□ People are generally pretty stable over short-medium periods of time
□ But there are some specific periods where your needs and behaviours change
® E.g. Have a kid, a puppy, moving, new job
® During these times, you are susceptible to buy things you otherwise wouldn't
Key goals - getting the relevant stuff in front of you at the right time
97
Marketing 4 P's
○ Product
§ What is being sold. How does it meet consumer needs? How is it branded?
○ Pricing
§ What is the appropriate price for this product
○ Place
§ What's the best way to get the product to the consumer?
○ Promotion
§ How should the consumer learn about the product?
98
Product Policy (neuromarketing)
The heart of marketing - which are the products that make the most sense for me to offer and who do I offer them to?
§ How a company determines what products to offer in what context, to which individuals
§ Usual market research uses surveys. These are unreliable as they do not always track preferences (neuroscience may help)
99
Price Policy (neuromarketing)
Higher prices may be perceived as a disincentive to buy a product, but it could also be viewed as a market of product quality.
"Willingness to pay" is individually computed --> subjective value (neuroscience may help)
100
Communication Policy (neuromarketing)
The efficacy of ads is not especially clear (change in purchasing in markets where ads were run - neuroscience may help)
□ What is it about an ad that works? - not always clear why it works
Participants who received beta-blockers recalled less details about ads (drug blocked emotional arousal)
□ Premise that emotional memories are often more easily recalled than non-emotional memories
□ E.g. Where you were on 9/11
101
Distribution Policy (neuromarketing)
§ Framing effect is utilised to the advantage of the company
§ Bonuses for loyal customers
102
Geographical segmentation
Examples of the heatmap for what people in different USA locations call soft drink/MRI costs
Pepsi might want to have different ads calling it a different thing for the different areas
Now with streaming, this can be targeted on the individual level
MRI costs in Alaska/hospitals compared to MRI clinics are an example of segmentation gone awry
103
Psychological segmentation
§ Could be that you are a company that decides you can only have 3 or 4 versions of your ad
□ E.g. One where a spokesperson uses a different word for soft drinks each time
□ Psychologically segment the biggest groups (almost like a phenotype) and then figure out which group(s) are the most efficient to advertise to
® Also where influencer advertisements are good for reaching broader audiences
Cambridge Analytica - example of huge mismanagement
§ 87 million users had data exposed
□ Utilised info to advertise/influence people about an upcoming election
□ People obviously pretty unhappy with them trying to influence democratic processes
104
Surge Pricing
Example of Uber and Flight costs - changing price based on supply/demand in real time
Uber also uses graphical segmentation - will be more expensive in busy, high demand areas
105
How economics and marketing intertwine
○ Since companies are ultimately interested in influencing choice (marketing) and economics is interested in understanding choice itself, they are a natural pair
○ The inclusion of psychology provides a framework within which to conduct experiments towards the above goals
*Important chart in slides*
106
Where to exert influence (neuromarketing)
§ The different layers/levels that we can exert influence:
□ Individual - influencing the individual directly
® May be a good or bad option depending on the size of the organisation
□ Group - segment/target a specific group
® E.g. New families
□ Broader market/society
® Example of influencing public perception of Swedish engineering being as good as German engineering
107
Nudging Evidence (neuromarketing)
○ Evidence is good, effect size (d=0.43), unless you adjust for publication bias (d=0.04)
§ Placebo effect is general around d=0.3
Mertens et al., 2021:
§ When only looking a high quality studies - there is no longer an effect
□ Pretty concerning
□ Does this make a difference out in the world - don't know...
§ So, the data for nudging doesn't really point to it being that useful, but if the cost for you to do it is low, then you might as well give it a try
***more on publication bias problem in slides***
108
Neurally Informed Marketing
§ He finds it pretty crazy that people have even tried this
§ Looking at how people respond to different ads neurally - another basis for segmentation
□ But does the brain activity even relate to a behavioural activity? - if not, who cares?
109
Supply and Demand (Pepsi example)
○ Increasing the price to make it feel premium
§ Experiment of giving people and product and then ask how much they would buy it for, vs telling them how much it costs/them already buying it and then asking how much they like it
□ Key point is that it can really change how we feel/behave based on the sequence
□ Bit of sunk-cost can come in here
□ The data will be biased in either case, so you want to do both when running experiments to help cancel them out
110
How to acquire neuroimaging (neuromarketing)
○ “An EEG (electroencephalogram) reads brain-cell activity using sensors placed on the subject’s scalp; it can track changes in activity over fractions of a second, but it does a poor job of pinpointing exactly where the activity occurs or measuring it in deep, subcortical regions of the brain (where a lot of interesting activity takes place). EEG equipment, typically costs about $20,000.”
○ “An fMRI can peer deep into the brain but is cumbersome, and it tracks activity only over the course of several seconds, which may miss fleeting neural incidents…fMRI machines…typically cost about $5 million with high overhead”
111
Alternative options to neuroimaging (neuromarketing)
○ Other methods than brain imaging that are maybe not as reliable, but far easier to implement and use in the real world
§ We can't carry a brain scanner around with us
§ But eye tracking is way cheaper, easier to implement and still useful
○ Probably still need the brain experiments - but can we see if the brain scans correlate to these other methods, then use those methods in the real world
- May not be super reliable but could be useful
112
General consensus on neuromarketing
○ “The prevailing attitude…can be summarized as… ‘neuroscience either tells me what I already know, or it tells me something new that I don’t care about.’”(Ming Hsu - UC Berkeley) …do marketers really need to be told that people’s brains react differently to Coke and Pepsi to understand the importance of branding?
- Harrell again
○ What are we actually learning for the cost of doing all these experiments
- Maybe it is interesting and we are learning some stuff, but for an organisation with a specific goal the general consensus is that it probably isn't worth it
□ Apple or Coke probably don't mind throwing some money at it, but smaller organisations probably wont
Advice he got:
□ Before designing an expensive MRI experiment ask yourself whether this is a brain or a behaviour questions - and if it is a behaviour question, do we really care what the brain is doing?
113
Lessons from old/young woman illusion and Kanizsa illusion
Woman:
§ Perceiving is really more than just passively receiving sensory information from the environment - it is an active process
§ There is something more, an interpretation, that is happening in our brain and causing the image we see
Kanizsa
§ Good example of how what we perceive is not just about what is literally there
§ We are predicting/inferring information to see triangle
□ As if the line continues and is occluded by a white triangle on top of it
114
Lesson from black/white dog illusion
§ Once you see the dog on its own, you can easily identify it in the noisy image
□ Our perception is dependent on our previous experience/prior knowledge
115
Lesson from the feet imprint illusion
Perception defaults to hardwired beliefs
□ What is happening here:
® We have an ambiguous image of an imprint of feet in the sand
® The most likely explanation is informed by a hardwired belief that light comes from above - which is creating the shadows that we see
® Even when the image flips, we keep the hardwired belief that light comes from above, so we now believe the feet must be raised
□ Another prior belief could be that we know feet tend to imprint in sand, but it is not completely insane to have a sort of feet sand castle
® Both are significant, but sun probably more so
116
the 'Basic Framework' (computational psychiatry II)
Formalises the perception principles demonstrated by the illusions
- perception is more than just sensation, it is an active process
- perception depends on prior experience
- perception defaults to hardwired beliefs
Basic Framework - sensory info <---> prediction/priors
○ Perception is an active predictive process
○ According to the Bayesian Brain hypothesis, sensory information is represented probabilistically
○ Humans use prior knowledge and uncertainty to guide perceptions/decisions
§ E.g. if you are lost in the dark without glasses you can rely on either what you see, or your memory of where you left your glasses
□ The darker the room, the more likely you are to rely on your memory
117
Audiovisual integration
§ When determining where an object is in space we are really good at using vision
□ Audio is really good for timing
§ But if an object involves both (e.g. A mouth talking - listening and watching mouth) we integrate the two - e.g. It might be harder to understand someone if they cover their mouth
§ If one source is stronger than the other, we put more weight on that sense than the other
- We can represent these weightings of where things are in space through Gaussian distributions
***specific examples and more details in slides***
118
'Basic Framework' example paper (design)
Experiment with the two throwers and 'ripples' in a pond
○ Told there are two throwers
§ One really good one and one bad one
§ Told the thrower is aiming at the middle of the pond
□ Good thrower will be closer to the middle on average and bad thrower will be more varied
○ Light blue dots are the splashes the coin makes when it falls into the pond (or ripples)
○ Task manipulated both the prior and likelihood variance
§ Design - vary the decision, prior and likelihood info
§ Prior is determined by the thrower's skill at aiming for the middle with the coin
§ The likelihood is manipulated by how close or far the splashes are shown to be
119
'Basic Framework' example paper (results)
§ Slope - likelihood/prior reliance
§ Y axis plots where participants say the true position of the coin is
□ If people are just using the prior info (thrower is aiming at the middle) they will always put the estimated of where it fell over the horizontal line ('it is in the middle')
□ If they were just using likelihood info (spread of the dots) they would always say the coin fell in the middle of the dots and it would stay on the diagonal line
® What we see is something in the middle - they are using both sources of information
§ She explains each graph in the lecture
§ Generally - you can see how they shift their weighting of the information given based on their beliefs of the thrower's accuracy and the closeness of the splashes they can see
§ We can relate the slope of these regression lines with how much weight people place on one source of info than the other - through Bayesian equations
120
Autism and the Kanizsa illusion (theories)
§ More likely to see the reality of the shapes - no triangles
○ Why?
§ Stronger bottom-up (sensory) or weaker top-down effect (priors) on perception?
§ Are they placing more weight on sensory information in front of them or less weight on their previous experience?
§ One theory
□ The sensory information is being encoded very precisely (Precise-likelihood)
□ Therefore they rely more on sensory information because it is more reliable
§ Another
□ It is driven by blurry/highly variable prior experience (Hypo-priors)
□ Therefore they rely on sensory information because prior experience is not reliable at all
® But in this scenario, there is less confidence in the final decision - almost like they have no choice (?)
121
Reminder to look at Autism (likelihood/priors reliance) study in the slides
detailed answers in the slides - to complicated for flashcards
122
The 'Bayesian Framework'
The Bayesian framework proposes that percepts are formed from a combination of sensory and prior information.
