Goal-directed learning and obsessive-compulsive disorder
- Type:#article
- Year read:
- Subject: (in brackets, can also bracket keywords in text)
- Bibtex: @gillan2014
- Bibliography: Gillan, C. M., & Robbins, T. W. (2014). Goal-directed learning and obsessive-compulsive disorder. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 369(1655). https://doi.org/10.1098/rstb.2013.0475
Why and when I was reading this
I was doing a deep dive into the role of Habits in OCD
Key takeaways
- Habit hypothesis of OCD, compulsions are not goal-directed avoidance behaviours but excessive habit formation
- OCD patients continue to respond to devalued stimuli, which the authors argue supports the habit formation model
- Compulsivity is “a hypothetical trait in which actions are persistently repeated despite adverse consequences”
This position holds that patients with OCD largely understand the relative value of the available outcomes and the cost of actions, and aim to promote expected values of outcomes and desist from compulsive behaviour, but cannot exert the necessary control over their actions to realize this goal.
- Double dissociation of subregions of dorsal striatum, found through lesions:
- Dorsolateral striatum (putamen): habitual responding
- Dorsomedial striatum (caudate): Goal-directed action control
- The pattern is less clear in frontal regions but the medial orbitofrontal cortex is thought to be involved in goal-directed action selection
Here they critique the classical CBT model of OCD presented by Paul Salkovskis and Stanley Rachman
Classic cognitive models of OCD posit that obsessions precede compulsions, which are considered active attempts to gain relief from obsessive thoughts. Indeed, it is impossible to ignore the tight coupling between the content of obsessions and compulsions in OCD, which leads to the intuitive inference: “I fear contamination and therefore I feel compelled to clean excessively”. Based on recent observations, we propose that the reverse - “I feel compelled to clean excessively and therefore I must be afraid of contamination” - may better capture the OCD phenomenon.
Page 1: Highlight annotation by Oskar Flygare on June 8th 2018, 4:37:35 pm:
Based on this research evidence, which suggests that rather than goal-directed avoidance behaviours, compul- sions in OCD may derive from manifestations of excessive habit formation, we present the details of a novel account of the functional relationship between these habits and the full symptom profile of the disorder. Borrowing from a cognitive dissonance framework, we propose that the irrational threat beliefs (obsessions) characteristic of OCD may be a consequence, rather than an insti- gator, of compulsive behaviour in these patients.
Page 1: Highlight annotation by Oskar Flygare on June 11th 2018, 9:33:08 am:
Although the list of overt behaviours that have been classified as compulsive is quite varied, there is consensus that compul- sivity is ‘a hypothetical trait in which actions are persistently repeated despite adverse consequences’
Page 4: Highlight annotation by Oskar Flygare on June 11th 2018, 9:48:28 am:
OCD patients have increased grey matter volumes in the putamen (extending to the caudate nuclei) relative to healthy controls and other anxiety disorder groups [60], whereas all anxiety groups (including OCD) showed common decreases in dorsolateral prefrontal and anterior cingulate cortex. Another meta-analysis found increased volume of the OFC, putamen and insula in OCD patients, which was a function of age, such that the normal loss of volume was not observed in patients as they aged.
Page 4: Highlight annotation by Oskar Flygare on June 11th 2018, 9:50:20 am:
The results indicated that OCD patients have a signifi- cant bias towards stimulus – response learning, at the expense of acquiring action – outcome associations.
Page 5: Highlight annotation by Oskar Flygare on June 14th 2018, 3:13:47 pm:
The first is that OCD patients may have a deficit in action – outcome associative learning, which causes them to rely excessively on stimulus – response links that were previously reinforced.
Page 5: Highlight annotation by Oskar Flygare on June 14th 2018, 3:14:33 pm:
Next, an alternative possi- bility is that excessive stimulus – response learning in OCD might cause patients to lose their sensitivity to action – outcome links, producing deficits in explicit action – outcome knowledge.
Page 5: Highlight annotation by Oskar Flygare on June 14th 2018, 3:11:48 pm:
OCD patients experienced regretful trials as being even more aversive than healthy comparison subjects. OCD patients and com- parison subjects did not differ in the extent to which their choices were based on the expected value of the available options, suggesting that basic decision processes were not affected in these patients.
Page 5: Highlight annotation by Oskar Flygare on June 14th 2018, 3:12:15 pm:
OCD is a disorder of compulsive avoidance rather than reward-seeking behaviour, and habits in avoidance had previously not been experimentally demonstrated in humans or animals.
Page 5: Highlight annotation by Oskar Flygare on June 14th 2018, 3:16:06 pm:
This indicates that habit biases in OCD are not necessarily driven by deficits in goal- directed contingency knowledge. It is plausible then that both habit-based and goal-directed learning may be affected in OCD, however, until a behavioural definition of habit learn- ing that does not rely on an absence of goal-directed control can be formalized, there is perhaps little use in making a distinction between processes that are somewhat reciprocal.
Page 5: Highlight annotation by Oskar Flygare on June 14th 2018, 2:46:07 pm:
Using this paradigm, there is recent evidence to suggest that model-based control over action is selectively diminished in OCD patients
Page 5: Highlight annotation by Oskar Flygare on June 11th 2018, 9:53:04 am:
A devaluation sensitivity test revealed that OCD patients were proficient in their goal-directed control over their responses prior to over-training and that their behaviour became excessively habit-based over the course of over-training.
Page 5: Highlight annotation by Oskar Flygare on June 14th 2018, 3:16:53 pm:
A third possibility, which we will touch on only briefly, is that rather than abnormalities in goal-directed control or habits, in OCD the problem could lie in the arbitration between these controllers.
Page 5: Highlight annotation by Oskar Flygare on June 11th 2018, 9:53:16 am:
However, it is possible that avoidance habits are associated with impaired instrumental extinction in a more general sense, a prop- osition, which to our knowledge has not yet been tested. It is however unclear if and how these might be parsed experimentally.
Page 5: Highlight annotation by Oskar Flygare on June 11th 2018, 9:53:29 am:
Using different methodologies, these three studies show that in OCD there is a consistent shift in balance away from goal-directed associative control over action towards stimulus – response habits.
Page 6: Highlight annotation by Oskar Flygare on June 18th 2018, 10:24:56 am:
ego-dystonic disorder; the thoughts experienced and actions performed by patients are discordant with their concept of self, either categorically or proportionally. In other words, patients have insight (although it can be diminished in some cases) into the irrationality of their compulsive actions; they want to stop but cannot exert control over the urge to act.
Page 6: Highlight annotation by Oskar Flygare on June 18th 2018, 10:24:15 am:
classic cognitive models of OCD posit that obsessions precede compulsions, which are considered active attempts to gain relief from obses- sive thoughts. Indeed, it is impossible to ignore the tight coupling between the content of obsessions and compulsions in OCD, which leads to the intuitive inference: ‘I fear contami- nation and therefore I feel compelled to clean excessively’. Based on recent observations, we propose that the reverse—‘I feel compelled to clean excessively and therefore I must be afraid of contamination’—may better capture the OCD phenomenon.
Page 6: Highlight annotation by Oskar Flygare on June 18th 2018, 10:25:27 am:
dinal feature of the disorder. By definition, habits are behaviours insensitive to contingency and outcome value; in other words, they are ego-dystonic, purposeless acts. We pro- pose that the excessive habit learning reliably observed in these patients captures the divergence between will and action that typifies OCD. We hypothesize that this behavioural disturbance is the critical component of the OCD diagnosis and has its neurobiological basis in the circuits running between the OFC and the caudate, whose ( putative) hyper-activation dis- rupt normal goal-directed behaviour, fostering reliance on habits (figure 3).
Page 6: Highlight annotation by Oskar Flygare on June 18th 2018, 10:24:39 am:
But in the three studies of habit formation described above, there is clear evidence that excessive compulsive-like, automatic behaviours develop in OCD patients in the absence of any prior obsessions relating to the experimental task procedures.
Page 7: Highlight annotation by Oskar Flygare on June 18th 2018, 10:26:36 am:
Recent data, however, have challenged this popular conception of habits as mere action-slips, finding that habits are associated with a hitherto unreported premoni- tory ‘urge to respond’
Page 7: Highlight annotation by Oskar Flygare on June 18th 2018, 10:29:44 am:
It may be surprising to some how irrational threat beliefs such as these could survive without being promptly disconfir- med through experience. However, the fact that compulsions in OCD are avoidant, rather than appetitive, can readily account for this. It is a feature of avoidance that performance of this response naturally precludes the extinction of irrational beliefs about contingency (e.g. fear), because when avoidance responses are continually, but unnecessarily, performed, the only demonstrable contingency the individual is exposed to is one where a state of safety follows the performance of an avoidance response. This prevents exposure to the crucial dis- confirming case or extinction, i.e. when a state of safety is also followed by no response
Page 7: Highlight annotation by Oskar Flygare on June 18th 2018, 10:27:18 am:
in situations of cognitive dissonance, when behaviour contradicts belief, humans alter their beliefs to match their behaviour. As has been elegantly put by others, ‘actions create—not just reveal—preferences’
Page 7: Highlight annotation by Oskar Flygare on June 18th 2018, 10:28:03 am:
In the case of OCD, the argument is analogous, wherein the irrational thoughts often considered to induce compulsive responding, may in fact be the product of the mind’s attempt to resolve the dis- crepancy between patients’ cognitions and their otherwise inexplicable urge to perform compulsive behaviours. Specifi- cally, the experience of the irresistible urge to perform, or the very performance of, compulsive avoidance behaviours may engender cognitive dissonance that is reconciled by the development of a new irrational belief about threat in the environment. This new ‘fear’ makes sense of the need to compulsively perform avoidance responses and may of course contribute to the motivation of subsequent avoidance responding, forming a vicious cycle of sorts.
Page 7: Highlight annotation by Oskar Flygare on June 18th 2018, 10:32:04 am:
Indeed, the available empirical data relevant to this issue suggest that there is no direct association between trait anxiety or physiological measures of conditioned fear learning and extinction and habit formation biases in OCD patients
Page 8: Highlight annotation by Oskar Flygare on June 18th 2018, 10:34:11 am:
Whether there is a specific causal role for obsessions in this formulation is unclear. It is possible that obsessions in OCD reflect dysfunction in an entirely independent process that interacts with compulsivity in a bidirectional fashion, much like what we have proposed for anxiety.
Page 8: Highlight annotation by Oskar Flygare on June 18th 2018, 10:34:31 am:
Another interesting possibility is that the histori- cal distinction between obsessions and compulsions in OCD may be superficial and that obsessions are a form of compul- sive, automatic thought. In this way, both obsessions and compulsions could be considered products of a disrupted goal-directed system, leading to over-active automatic thoughts (obsessions) and actions (compulsions), rather than being discrete traits that interact with one another.
Page 8: Highlight annotation by Oskar Flygare on June 18th 2018, 10:33:09 am:
At a minimum, it is likely that trait anxiety may play a role in targeting OCD patients’ general tendency towards excessive habit learning specifically to the avoidance domain, rather than towards ‘appetitive compulsivity’, e.g. stimulant drug addiction
Page 8: Highlight annotation by Oskar Flygare on June 18th 2018, 10:33:29 am:
With this in mind, the considerable neurobiological, pharmacological and genetic heterogeneity of OCD might be explained by understanding how trait anxiety, as an indepen- dent contributor to compulsive avoidance habit learning, fits into a trans-diagnostic model of the disorders along the respective compulsive and anxiety spectrums. In other words, it may be the case that there are many routes to the OCD phenotype, and that dysfunction in habit learning and trait anxiety are independent, yet interacting diatheses.
Page 8: Highlight annotation by Oskar Flygare on June 8th 2018, 4:40:35 pm:
The hypothesis that a shift from goal-directed to habitual con- trol over action mediates compulsivity in OCD ties well with the neurobiological and pharmacological basis of habit learn- ing in rodents and humans. This hypothesis also accords well with the neurocognitive profile of motor inhibition failures in OCD that are observed following repetition of action [95]. The ‘COD’ model of OCD proposed here has implications outside of the specific domain of this disorder. It is a plausible interpretation that the experience of premonitory ‘want’, or ‘urge’, reported in not only OCD but also substance-dependent individuals and tic disorders, may be a consequence of exces- sive stimulus – response associations.