Module 4: Cognitive bias modification Flashcards
1. Module background (1-10), 2. Wiers et al (2011) (11-27). 3. Schoenmakers (2010) (28-42), 4. Wiers et al. (2020) (43-53), 5. Lecture (54-104)
1
Q
According to the dual-process account of CBM, different cognitive biases in addiction are
A
driven by bottom-up mental associations. The effects of which are moderated by top-down executive control processes
2
Q
how are the clinical relevance of the approach bias and attention bias evident?
A
in the fact that existing treatments like CBT are not effective for all patients
- > high relapsing (half of those in treatment within 6 months, and 70% relapse within a year)
most importantly studies indicate that additional cognitive bias retraining may reduce the risk of relapse
3
Q
Task to measure:
Approach bias
A
- Tendency to approach appetitive stimuli
- investigated with AAT: approach avoidance task
1) pictures of alcoholic and non-alcoholic drinks on comp. screen - pictures tilted to left or right
2) participants pull a joystick towards them (approach response) or away from them (avoidance response) based on whether image is tilted left or right
–> a faster reaction time when drawing alcoholic drinks towards you than when pushing it away is interpreted as an approach bias
4
Q
Task to investigate:
Attention bias
A
- a form of selective attention that addicts have for drug-associated stimuli
- The dot probe test:
1) Participants presented with a drug-related stimuli and a neutral stimuli
2) one of the two stimuli is replaced with a specific stimulus to which subject has to respond
shorter response time to a stimulus that replaces a substance-related cue is interpreted as an attentional bias for substance-related peripheral stimuli
5
Q
Task to investigate
Memory bias
A
- automatically activated memory associations
- IAT: Implicit association task
1) participants categorize words or pictures ito 2 x 2 categories with a left and right button
2) e.g., the images categorized as alcoholic or nonalcoholic (left or right) and they are also categorized as active vs. non-active
If subjects respond more quickly when alcohol and ‘active’ share a button, than when the categories are divided over the buttons the opposite way, this would indicate that they associate alcohol with high arousal.
6
Q
Experimental proof-of-principle studies
A
conducted in a lab to reveal psychological mechanisms underlying human behavior in the lab
7
Q
clinical RCT studies
A
conducted in a clinical setting with patients to thest the efficacy of an intervention in a clinical sample
8
Q
Wiers, Boffo and Field
reviewed evidence for CBM interventions for AUD, they distinguished between
A
(1) proof-of-principle studies
(2a) online studies in which self-identified problem drinkers receive CBM as a stand-alone intervention, and
(2b) RCT’s in which CBM is added to treatment as usual of alcohol-dependent patients.
9
Q
Wiers, Boffo and Field
what did they conclude upon their review (2)
A
Proof-of-principle studies are important as they provide the basis for clinical trials but CBM has small, short lived effects on drinking in students
Clinical trials show that CBM does hold promise as an add-on intervention to treatment of alcohol-dependent patients
–> differences between these types of researchers can account for ifferences, but the table shows those that are most important (from background module 4)
10
Q
ABC training
A
patients are trained with personally relevant antecedent cues (A) to make goal-relevant behavioral choices (B) in light of their consequences (C).
–> a recent suggestion for improving traditional CBM (suggested by Wiers and colleagues)
11
Q
What is the article of Wiers et al (2011) about
A
the effect of a CBM intervention (in addition to a regular CBT treatment) on the approach bias in alcohol use disorders and treatment outcome
12
Q
The alcohol approach/avoidance task
A
1) pictures of alcoholic and non-alcoholic drinks on comp. screen
- pictures tilted to left or right
2) participants pull a joystick towards them (approach response) or away from them (avoidance response) based on what they are instructed
3) pulling the joystick towards them, makes the picture bigger (zooms it) and pulling it away makes the image smaller
–> the zooming effect generates a sensation of approach or avoidance
–> a faster reaction time when drawing alcoholic drinks towards you than when pushing it away is interpreted as an approach bias
13
Q
what was shown in the first study using alcohol-AAT
A
heavy drinkers show an approach bias: they pulled the joystick faster in response to alcohol pictures
–> not the case for general positive or negative pictures
14
Q
What was done in the first preclinical CBM study targeting an alcohol-approach bias
A
student’s action tendencies were experimentally modified: half were trained to avoid alcohol, and other to approach it
-> successful retraining was associated with congruent changes in alcohol consumption in a taste test
15
Q
what does the current study add on to these existing studies?
A
Use the new CBM method of retraining in a clinical sample
–> in the experimental condition, participants were trained to consistently make avoidance movements to alcohol pics and approahc movements to non-alcohol pics
16
Q
patients in the control condition
A
not trained to avoid alcohol at all / received sham training
17
Q
What were predictions of the study?
A
minimal CBM intervention would change approach bias for alcohol
18
Q
Experimental tasks at pretest and posttest
IAT, what is it and how was it used in assessment in this study?
A
strength of associations between concepts
- 2 target categories and 2 attribute categories
- in 1st block, alcohol and approach stimuli categorized together on one side
- in 2nd block alcohol and avoidance stimuli categorized together
strength of alcohol-approach associations were estimated by the extent to which participants responded more quickly to the first combined sorting condition
19
Q
Experimental tasks at pretest and posttest
Alcohol-AAT, how was it implemented as an assessment?
A
- 20 pictures of alcoholic and 20 non-alcoholic drinks were presented in landscape or portrait mode
- participants instructed to pull joystick towards (approach) to portrait formats and to push (avoidance) to landscape formats
–> so required response is unrelated to the content’s of picture
20
Q
2 experimental conditions of the study
A
1st experimental condition: (explicit instruction) patients respond to alcohol pics by making avoidance movement + responding to non-alcoholic pics by approach movement
2nd experimental condition: (implicit instruction) continue to respond but all clohol pics presented in the format that prompted an avoidance response. All non-alcoholic pics in the format prompting an approach response
21
Q
2 control conditions of the study
A
1st control condition: no training at all
2nd control condition: received 4 additional sessions of the assessment task (sham training). Which required an equal number of approach and avoidance movements to both alcoholic and non-alcoholic drinks
22
Q
what are the authors referring to when they say that CBM was successful in changing automatic approach tendencies with generalized effects
A
during training only half of the available alcohol and non-alc images were shown.
at posttest they were shown all available pics (including those that they didn’t train with)
–> results: the training effect generalized to those pics that weren’t trained
23
Q
Discussion: generalization effect
what is said about the generalization effect that was found?
A
brief CBM effective in changing automatic approach tendencies with generalized effects
- this generalizing effect was not found in attentional retraining
24
Q
Discussion: CBM procedure
What remains unanswered: Is the zooming effect/joystick pushing necessary for the CBM procedure. Why is it unanswered?
A
theories of embodied emotion and cognition: yes
- physical movements can influence your emotions and thoughts
other researchers: No, associations between object and concept of approach or avoidance are most crucial
This study found that those trained to avoid alcohol using joystick also scored lower on the IAT, a task that doesn’t inolve any joystick movement –> something in their deeper alcohol-related associations changed—not just their motor habits
25
what is one proposed mechanism that could explain the beneficial change in treatment outcome following the change in approach tendencies
patients probably profitted more from other treatment components
26
limitations of the study
* changes in cognitive processes did not mediate clinical outcomes
* did not assess action tendencies for alcohol later during treatment or at follow-up
additionally the practical question of how many training sessions are actually needed / optimal
27
what is suggested for future studies to explore (4)
* whether training outcomes are genetically moderated
* neurocognitive effects of training
* assessment measures decomposed and investivated separately
* extent to which modification of one bias also changes a related bias
28
What is the study of Schoenmakers et al. (2010) about
testing the effects of attentional bias modification training (ABM) in addition to traditional CBT in a clinical sample
29
attentional bias AB in the context of alcohol dependence
a selective attention for drug (alcohol-related) cues
--> hypothesized to have causal effects on substance abuse, addiction development and maintenance, craving, treatment outcome and relapse
30
According to incentive sensitization, AB results from
repeated pairing of drug (alcohol) cues with direct effects of alcohol -> sensitized reaction to alcohol cues (they are salient)
31
what evidence suggests that targeting AB may be beneficial for patients with alcohol use disorders (3)
1. AB has been theorized to share a reciprocal causal relationship with craving (not that strong + not under all circumnstances)
2. association between AB and severity of addiction
3. excessive drinkers with low compared to high AB are 3 times more successful in cutting down
32
Describe the main procedure of the attentional bias modification (ABM) training used in the Schoenmakers et al. (2010) study. What was the goal of this training?
The ABM training involved five sessions where patients in the ABM group were trained using a modified visual-probe task.
In this task, probes consistently replaced neutral pictures paired with alcohol pictures, aiming to train patients to disengage their attention from alcohol-related stimuli.
33
What were the two components of AB they assessed?
The researchers assessed the **speeded detection of alcohol stimuli (shorter presentation time)** and the **difficulty to disengage from alcohol stimuli (longer presentation time)** by measuring reaction times to probes.
34
Experimental group: ABM group
Participants in the experimental group undergo the AB modification training
-> aims to reduce attentional bias for alcohol cues by training participants to disengage from these cues
35
Control group
undergo a control training task
-> categorization task similar to the IAT (does not influence AB)
36
What were the key findings of the study regarding the effect of ABM on attentional bias towards alcohol-related stimuli? Did the effects generalize?
The study found that ABM was effective in reducing the difficulty that alcoholic patients had in disengaging their attention from alcohol-related stimuli.
This effect generalized to new alcohol-related stimuli that were not used during the training.
37
what implication does generalization have? what do authors suggest is behind this effect?
it suggests that patients might be able to disengage attention from a wider range of alcohol cues encountered in their daily lives, not just the ones used in the training
authors suggest that the use of a large number of different stimuli in the training likely facilitated this generalization by encouraging participants to develop a general strategy to disengage from all alcohol-related stimuli
38
Did the ABM intervention have a significant impact on self-reported craving for alcohol in the abstinent patients, according to the study's results?
the ABM intervention did not show significant effects on subjective craving for alcohol
The authors suggest this might be because ABM affects emotional reactivity rather than conscious craving, or due to the low levels of reported craving in the sample.
39
What were the two main exploratory findings of the study concerning relapse and overall treatment success in the ABM group compared to the control group?
1. although the percentage of patients who relapsed was similar in both groups, the time to relapse was longer in the ABM group.
2. among patients in the 3-month standard treatment program, those in the ABM group were discharged from the clinic earlier based on their therapists' positive judgment of treatment success.
40
What were the three factors identified by the researchers, based on previous literature, that appeared to increase the effectiveness of ABM interventions, and how were these incorporated into their study design?
* motivating participants
--through feedback and goal setting
* presenting a large number of different stimuli
-- used a larger set of alcohol-neutral picture pairs compared to previous studies
* performing multiple training sessions
-- conducted five training sessions.
41
What was the purpose of the control training condition in the study, and how did it differ from the ABM training?
the control is a categorization task designed to provide similar feedback, goals, and exposure to pictures as the ABM but without specifically training disengagement from alcohol cues.
Participants categorized stimuli as alcohol-related, neutral, numbers, or color names by pressing different buttons.
42
What are some limitations of the study that the authors acknowledged, and what future research directions did they suggest?
the sample size (N=43) was modest --> caution in interpreting these findings.
Suggest future studies with larger samples to replicate the findings, investigate the underlying mechanisms of ABM, and confirm its impact on improving addiction treatment effectiveness.
43
What is the central premise behind cognitive-bias modification (CBM) interventions for addiction?
that addiction is often maintained by automatic cognitive processes, or biases, and that modifying these biases can supplement treatments targeting more controlled cognitive-motivational processes.
44
what were the initial findings regarding the effectiveness of CBM as an add-on treatment for alcohol use disorders in clinical samples?
RCTs in clinical samples demonstrated that adding CBM to the treatment of alcohol use disorders reduced relapse rates compared to no training or sham training
45
What is the main distinction between the associative and inferential accounts of how CBM works?
The associative account: CBM works by replacing maladaptive associations (e.g., alcohol-approach) with new ones (e.g., alcohol-avoidance) through repeated pairings.
The inferential account: CBM changes propositional representations about causal relationships between stimuli, responses, and outcomes.
46
Why do authors suggest that simply training avoidance of addiction-related stimuli in the lab might not always translate to changes in addictive behavior in real life
--> this is one reason why traditional associative accounts of CBM can be flawed
The effects of avoidance training depend on beliefs about the implications of the learned relation
- such as the belief that avoiding alcohol will help one refrain from drinking
- the avoidance action like pushing away needs to *mean* something to the individual for it to translate to outside the lab
Additionally people need to understand the connection between what they do and why it matters in order for effects to work better
- this doesn't always come across in a lab --> so it also doesn't come across in real life
47
Describe the three key components of the authors' proposed ABC training and how they differ from traditional CBM.
(A) personally relevant antecedent cues (B) goal-relevant behavioral choices
(C) health-relevant consequences.
Unlike traditional CBM which often trains general associations (e.g., cue-avoidance),
ABC training personalizes all these elements based on an individual's goals and context --> this is important because CBM was experienced as meaningless for patients
48
Why do the authors believe that including goal-relevant alternative behaviors (the "B" in ABC training) is an improvement over previous CBM approaches, especially for non-alcohol related addictions?
For addictions without a universal alternative (like smoking), previous CBM used visually matched but not goal-relevant alternatives (e.g., a pen instead of a cigarette).
Including goal-relevant alternatives (like exercising to reduce stress for a smoker) **aims to train behaviors that actually fulfill the underlying needs or motivations associated with the addiction.**
49
How does the inclusion of personally relevant consequences (the "C" in ABC training) aim to enhance the effectiveness of CBM?
Including personally relevant consequences makes the learning of alternative behaviors more meaningful
- allows individuals to experience effectiveness of those behaviors in relation to their goals.
e.g., in the unhealthy eating study, approaching healthy foods had positive consequences for the avatar's health, leading to real-life reductions in unhealthy eating
50
What role do personally relevant antecedent contexts (the "A" in ABC training) play in improving the transfer of learned responses to real-life situations?
e.g., specific locations or emotional states, help to link the trained behavioral choices and their consequences to real-life high-risk situations.
This contextualization aims to improve the likelihood that the learned adaptive responses will be triggered in those specific environments or situations.
51
How does ABC training relate to existing therapies like cognitive behavior therapy and treatments analyzing antecedents and consequences of behavior? What is the key difference highlighted by the authors?
ABC training **combines** elements of CBT (focus on cognitive and behavioral change) and analysis of antecedents and consequences.
However, a key difference is its explicit focus on the automatization of adaptive inferences and behaviors through practice, aiming to reduce the need for effortful control in high-risk situations.
52
According to the conclusion, what is the primary goal of ABC training, and what kind of empirical support is still needed?
The primary goal of ABC training is to **foster automatic behavioral choices that are in line with patients' health goals in specific contexts.**
Authors state that while preliminary findings are promising, empirical support from large clinical trials is still needed to confirm its effectiveness.
53
Cognitive biases in addiction
* Attention bias
-relatively strong attention (engagement/disengagement) for substance-related stimuli
-where you look
* Action tendency: approach bias
-we constantly appraise the world and generate automatic action tendencies
-in SUD, you are unable to inhibit these generated tendencies
-what movement would you tend to do
* Memory bias
-relatively easy retrieval of substance-related associations (antecedents/effects)
-first thing you think of?
54
Why do we focus on these 3 biases specifically
studies show that these predict **unique** individual variance in those with addiction
55
Attentional bias
* generally measured with eye movements
* motivationally relevant stimuli attract and capture attention
* related to subjective craving (with elaboration in WM)
56
Attention bias: dot probe test
horrific reliability
1. alcoholic vs. non-alc picture
2. images replaced with pixels (1 or 2 pixels)
3. you have to indicate how many pixels you see by pressing a button
if you are faster to react to the pixels when they replace pictures of alcohol, then its an indication of an attentional bias towards alcohol-related stimuli
57
Approach bias: irrelevant feature paradigm
1. you see landscape formatted or portrait formatted pictures
2. instructed to pull a joystick or a portrait and push for a landscape
if you are faster to pull than push, it indicates that you approach (as opposed to avoid) easier
58
Approach bias: irrelevant features paradigm by Wiers et al. 2009 assessment results
different pictures presented
1. no difference between pulling and pushing to neutral stimuli
2. sig. difference between heavy and light drinkers when it came to alc pics
-heavy drinkers faster to approach alcohol
-especially those with risk allele OPRM1 (also associated with cue-induced craving)
59
why measure an irrelevant feature version?
no instructions, so we get to measure more implicit behavior
- in CBM participants automatically trained to push away alcohol pics because they are simply told to push as a response to a format and not the content of the picture (implicitly trained)
- on the other hand using relevant feature paradigms is more reliable
60
Studies find that both attentional & approach biases are
* Related to substance use
* There are some measurement issues, related to “irrelevant feature methods” Relevant fature makes it more reliable but less implicit
* Indirect; generally rather unreliable (but much better for new measure: dual probe (seeing 2 videos at the same time; alcohol vs. non-alc --> under testing rn))
* Indirect measures good for research into relatively automatic mechanisms (group-level), not for individual diagnosis.
61
Attentional - approach bias general remarks
> relative tencency for alcohol (not all show these biases)
-50% show biases
-30% no biases
-16,7% have negative biases
> found across addictions
> could be "remediated" with abstinence or targeted training ;) (CBM)
62
Memory: Reaction time (RT) tests
General idea
A) phase 1
1. one combinations of words (positive and arousal vs neutral) and drinks (alcohol vs non-alc)
2. press left when you see a positive, arousal-related word OR when you see alcohol
B) phase 2
1. another combination of same words and same pictures
2. but now neutral words paired with alcohol pictures
if you are a heavy drinker this phase 2 will be way more difficult
--> predicted alcohol use in adolescents 1 year later
63
Memory associations
* related to drinking, also after controlling for explicit expectancies
* not a pure measurement of associations (you get a small IAT effect are able to switch / EC artefact)
* Alcohol-arousal associations related to heavy drinking, equivalent of sensitization?
* More recent work highlights relevance "me"-drinker associations
-Identifying yourself as a drinker
-Has been shown to predict drinking patterns over a period of 5 years
64
Non RT tasks from meory research
e.g., asking participants to fill out what the first thing that comes to mind is when they see the words "having fun"
- these measures predict addictive behaviors over time
65
Cognitive control processes (WM) as a moderator of the effects of implicit cognitive processes: study hypothesis
if you score higher on WM your more rational pro vs cons reasoning is the better predictor of alcohol use
- in contrast if you score low, its the implicit automatic pilot that predicts usage
66
Working memory: SOPT test
| self ordered pointing task
1. Pick an object
2. Cards are shuffled
3. Now pick an object that is not the one you have previously picked
clear dorsolateral prefrontal cortex associations
67
WM as a moderator: what has been found across studies?
* Low WM: IAT measures/associations pedict alcohol use/problems
-automatic pilot determines whether they drink or not in a certain situation
* high WM: explicit expectations predict alcohol use problems
68
More evidence from other studies
* relatively automatic processes predict alcohol/drug use in individuals with relatively low executive control
* also evidence for similar pattern in other behaviors where impulsive and reflective processes may clash: aggression, eating, sex
-these predict your behavior when your WM is low
69
Conclusion on cognitive bias part
1. cognitive biases related to substance use
2. biases in attention, action tendencies, cognitions
3. cognitive control processes (e.g., WM) moderators: stronger influence of automatically activated cognitions in individuals with relatively weak control capacity
70
Cognitive training: training the general ability (WM,general inhibition, self-control taining)
* if we succeed in training general ability then no matter what domain you will improve in controlling your impulses
* WM training: positive results in children with ADHD and other problems have been claimed (Klingberg TICS 2010)
-Training kids with adhd who are at risk for developing addictions early on to prevent future problems
-Or already to reduce attention deficits in class
-Kids or adults get better at the specific task used to measure
* However: problems with generalization (Sonuga-Barke et al, 2013)
-Training on Task A and testing whether effects persist for task B --> if they improve on task B (closed generalization)
-If kids get better by concentrating more on task in class that is an example of a far generalization
-Typically you don't find anything which is disappointing
71
Does training general abilities affect drinking behavior?
* number of studies: no
* promising effects:
-reduced delay-discounting: (1 dollar now vs. 5 dollars in 1 month)
--> after training participants get more future oriented
--> related to the finding: this effect generalizes to episodic future thinking = ability to imagine how you're gonna tackle a problem in the future in detail
* VM-training might be helpful in recovery process when better integrated with it
--> because it has been found to increase self confidence
conclusion: too early to dismiss general ability training (due to lack of no direct effects) - instead combine it with more applied training like episodic future thinking and relate it to someone's own goals and life histories (it is still useful)
72
Training targeted biases: Cognitive bias modification
Targeting specific biases with addiction relevant cues to see whether biases have a causal relationship with behavior in addiction
- started from the question: if we temporarily modify this bias, what is the effect on the behavior?
first experimental tests (PoP-studies) successful, after that clinical applications
73
Clinical appplication: dot probe task, training procedure and difference between experimental groups
assessment: probe replaces problem category (alcohol, cannabis, anxiety) and control category equally often
training:
* experimental group: probe replaces control category most (or all) of the time
* control group: continued assessment described above)
-you can also do:
-no training
74
Clinical application: Dot probe test by Schoenmakers (2007)
* binge-drinking students recruited for experiment on alcohol and reaction times, including taste-test
* control group: half trained away from alcohol and half received no training (continued assessment)
75
Effects from Schoenmakers (2007) study
Trained pictures: negative bias
- quickly react to coca cola than to the beer
BUT
no generalization to untrained picture (there is a non-significant effect into the right direction: small closed generalization effect)
ALSO
sadly no effect on drinking
76
Conclusions single session attention retraining alcohol
* possible to train heavy drinking students toward alcohol
* possible to train heavy drinking students away from alcohol, but
-no generalization
-no effects on behavior
* maybe there need to be more sessions?
77
Schoenmakers et al: increasing the amount of training sessions, experimental groups (Attentional bias)
* AR group
-Attentional training (AR): probe never replaces alcohol picture
-VP test: probe can replace any picture (??)
* controls
-Irrelevant IAT-like categorization
-same stimuli & feedback
--> new alcohol pictures every time to make it a more general rule
78
effects: Schoenmakers + increased training sessions
Experimental group:
-After 5 sessions -> generalized effect to untrained pictures
-Promising clinical effects, you get a later relapse if you undergo attentional training
control group:
increase in attentional bias
- ab increases naturally when you don't do anything in a clinic (incubation)
79
Approach bias: 1st proof of principal study
* medium drinking students (which is why it is not a clinical study)
-half trained toward
-half trained away
Results:
-generalized effects on untrained pictures after a single session
-also found on IAT and behavioral taste test (heavy drinkers acted like light drinkers and vice versa on a taste test afterwards)
80
Approach bias: 1st clinical study
German clinic allows for this
experimental groups
* distinguish between relevant feature training (instruction to push alcohol away) vs. irrelevant feature training (no instruction: portrait-landscape)
control groups
* continued assessment control (50-50)
* no training
**no differences in training groups nor control groups so they were collapsed**
81
Results of 1st clinical study on approach bias modification
significant generalizations to untrained pictures AND to IAT (a verbal memory association task, this indicates a far generalization)
pretest: on average we see an approach bias and no sig. difference between groups
- approach bias interesting because people in clinic often have had their life ruined due to alcohol yet the natural response is to approach it
posttest:
* control group: no effect
* experimental group: after training they are faster with avoid words and alcohol (IAT test) indicating a generalization effect to another TASK (far generalization)
very exciting
82
1st clinical study of approach bias modification: What was found in the 1 year follow-up?
Adding CBM to CBT had resulted in 13% less relapse
4 sessions of training on top of 3 months of general treatment + assessment after a year
83
Replication of 1st clinical study on approach bias modification, experiment groups
experimental group: training
control: no training
-as inspired from first study in which subgroups were merged
84
they also looked at mediation vs. moderation in this experiment (1st clinical study on approach bias modification)
mediation of change in alcohol approach bias: if you're in the training condition, chances are lower that you relapse
--> great cause this shows we are targeting a relevant mechanism
moderated mediation: only if you have a strong approach bias, it makes sense to reduce this approach bias
- we cant tell this beforehand in a reliably way so we can only train all of them
85
People who had had the training said that they consciously steered away from the beer section in the store, what is interesting about this?
we initially thought this to be an automatic implicit, unconscious process but these accounts disprove that notion
86
Independent replication (in detox) of the 1st clinical study on approach bias modification, results?
Difference of 22% between those who had training vs. no training
-but only on a short term relapse
87
Neural effects of CBM?
neuroimaging people before and after training and seeing how they react to images of alcohol vs soft drinks in the scanner
results:
Amygdala and nACC = strong activation before training
- decreases after training
interestingly
- an increase in the salience of the alternative
88
Meta-analysis: Is CBM effective in addiction? What's strange about their findings and why did it happen?
there is an effect of CBM on the bias but there are no clinically relevant effects
--> strange because all clinical studies showed effects still
They had mistakenly put together two totally different types of studies
- experimental lab studies vs. clinical trials
- 3 groups of studies put tgt
experimental lab vs. online RCTs vs clinical RCTs
89
What is needed in order to observe an effect in clinical samples?
a motivation to change behavior
90
Bayesian meta-analysis done to target only those studies in which the motive for change (reduce drinking) was established
* 14 studies
* small effect on cognitive bias
* also one on relapse rates
* no significant effect on use
conclusion: more data is needed
91
3 clinical RCTs after the Bayesian meta-analysis, conditions + results
conditions:
* approach bias training
* attentional bias retraining
* both
* placebo
* no training
results:
* statistically no differences between conditions but training is better than no training (8,5% difference)
92
Discussion (3 clinical RCT's after the Bayesian meta-analysis)
- Replicated beneficial training effects
- But we expected best outcomes of combined training.--> biases can be distinguished in meaningful ways (some people show more attentional bias), as the biases are correlated it was predicted that a combination would do best
- Probably too short for most (3 + 3); meanwhile we learned that median number of sessions needed in AAT-training is 6 (with lots of variation, Eberl et al., 2014, ACER)
93
Australian study of CBM on a clinical sample results
12 % less relapse within 3 months
94
For whom does CBM work best?
Lots of comorbidity, a third also have an internalizing disorder
- Does it work better or worse if you have a comorbid disorder?
--> it works better!!! This is a good result to see because usually treating those with comorbid disorders is quite difficult
95
Conclusion on CBM part
* effective as add-on in the treatment of AUD when people are motivated to change
* not in student binge drinkers who are not motivated to change nor in online training in people who want to reduce
96
Comparison with medication
Medication primarily works for those who don't find therapy effective
- Works in some cases
CBM only works as an add on for treatment --> these two proposed to work differently
97
Next steps: A) improve cognitive training
* Gamification, make it more playful --> not that useful
-For young people less boring but compared to other games that are played --> still boring
-Other population didn't like that it was treated like a game "I have a serious problem here"
* More personalized alternative goals: ABC-training
-In other addictions, you don't have a non-alcoholic drink choice like you would in alcohol use disorder --> problem for how to design training because what do you put in to the non-alc pics
-E.g., now you had to come up with alternative behaviors that are personally motivating
* Training based on more reliable assessment
* Training after reactivation (disrupting memory consolidation)
-More successful in anxiety than addiction
* Add neurostimulation
-People learn faster but clinical relevance is 0 because everyone will learn anyway
98
Next steps: B) ABC-training most relevant, why include consequences
* series of PoP studies show CBM effects to occur only with conscious awareness
and some are only generated by instruction
--> important to include consequence in the training
brain is constantly predicting future scenarios and we want to influence what that future scenario actually looks like
99
ABC training process
it is personalized: participants choose personally relevant:
A: Antecedent contexts : e.g., coming home stressed makes you think of smoking a joint
B: Behavioral alternatives relevant in that context: e.g., go for a walk rather than smoking
C: consequences: e.g., better health, save money, etc.
100
ABC - phases
1. forced choice to learn consequences (learned in sham-training)
2. Open choice with consequences
3. speeded open choice with consequences to foster automatization
101
initial PoP studies
* 193 hazardous drinkers recruited through Prolific randomized over ABC-training vs. sham-control (& 282 in replication study)
* Replicated stronger increase in self-reported and automatic negative alcohol expectancies after ABC- training (primary outcome)
* Exploratory/results:
* Stronger increase self-efficacy after ABC (both
studies)
* Heavier drinkers reduced drinking more after
102
initial studies of ABC training on abstinence challenge
* 263 (2021) and 320 (2022) volunteers NoThanks! Randomized over ABC, CBM, sham-CBM
* Stronger increase successful abstinence in ABC- training group vs. CBM / sham-CBM
* No differential effect in drinking during two weeks after challenge = as soon as february rolled in which shows the role that goals of people play
No longer lasting effect + not a clinical group
103
Conclusion (ABC)
* ABC training is theory-based new variety of cognitive training aimed at targeting actions based on automatic inferences
* promising initial results in volunteers
* next step: RCTs in clinical samples
104
Overall conclusions
* cognitive biases play a role in addiction
* cognitive training to change biases can be of use in the treatment of addictions
-general ability training some promise but long; probably only inpatient recovery
-CBM effective as add-on to inpatient abstinence-oriented treatment for alcohol use disorders
* room for improvement: theory-based ABC-training
