Ninewells Hospital & Medical School, Dundee, DD1 9SY

Compulsions, rituals, and repetitive behaviours

People often assume that if you have compulsions (or other ritualised behaviour) then you must have OCD. However, this isn't automatically the case since rituals and compulsions occur in other conditions, usually those associated with anxiety. This blog explores what we know about compulsions in Generalised Anxiety Disorder (GAD) and Autism Spectrum Disorder (ASD) versus those that occur in OCD. A similar post, exploring obsessions vs worries vs ruminations is also available.


Repetitive behaviours are very common, and most people will have some aspect of their behaviour that is ritualised. For example, we may clean the bathroom in a particular way, or have a strict/ preferred routine for showering. Some people might have small rituals that are associated with sports, and involve the order of putting on kit. Most children will have periods of ritualised behaviour during normal development and 'magical thinking' (if I don't do x, then y might happen) is common. Our familiarity with 'normal' rituals can make it difficult to confidently say if someone's ritualised behaviour represents 'normal' patterns of behaviour or is indicative of a disorder.

The continuum hypothesis

Some researchers will argue for what is called the 'continuum hypothesis'. Essentially, this means that there are no fundamental differences between 'normal' and 'abnormal' beliefs/ behaviours and the main difference is simply one of quantity. Others take an opposing view, and suggest that there are differences between normal and abnormal mental states. There has never been a conclusive study to say which view is correct, but there is (over time) some evidence that 'abnormal' mental phenomena differ in notable ways from the kinds of experiences that people without significant mental disorder have. More about this is in the 'Obsessions vs Worries' post.

Evidence for the continuum hypothesis for rituals/ compulsions

In a systematic review, De Caluwe et al, 2020 found many similarities between rituals/ routines and OCD compulsions, concluding that a continuity model was supported. They also found that age was a moderating factor: in younger children, continuity was likely but as people got older the behaviours were more likely to be maladaptive and pathological.

Ultimately, these kinds of symptoms are likely to be quite 'fluid' and people who have higher levels of rituals in early life are more likely to develop pathological rituals (and possibly OCD) in later life. In this way, the thresholds between 'trait' and 'state' are probably not fixed or reliable.


In order for this discussion to make sense we need to have some clear definitions regarding the terms that we're going to be talking about. For example, what's the difference between a 'compulsion' and a 'ritual'? We will also define other action/ movement-related behaviour such as tics, so we can understand the similarities (and differences) between the kinds of movements that can occur in psychiatric illness.


The definition given in DSM-III-R (American Psychiatric Association, 1987) is probably as good as any. This gives the following characteristics as being core to compulsions:

  1. "Compulsions are repetitive, purposeful, and intentional behaviours that are performed in response to an obsession, according to certain rules, or in a stereotyped fashion."
  2. They are designed to neutralise or prevent discomfort, or prevent some dreaded event or situation. For example, harm coming to others due to inattention, or illness arising from feared contamination.
  3. Compulsions are goal-directed. This means that they are directed at a specific goal or outcome, and this separates compulsions from tics and stereotypies.
  4. The behaviour is not connected in a realistic way with what it is intended to neutralise or prevent. Or, it is clearly excessive.
  5. The compulsions is usually (in the early stages, at least) associated with a desire to resist the behaviour. It is important to note that in most cases, the resistance is to the repetition of the act, and not the act itself. So someone might think it reasonable to check the door is locked (an act which is not resisted), they then resist having to check it over and over.
  6. The person recognises that the behaviour is excessive and/ or unreasonable.
  7. Although the behaviour usually reduces anxiety, it is not inherently pleasurable.

The most common types of compulsions in OCD involve hand-washing, counting, checking, and touching. Mental rituals can also be common.

One of the key features (and this is relevant when we think about rituals in conditions such as Autism Spectrum Disorder, or ASD) is that compulsions in OCD are usually preceded by a clear obsession that generates anxiety. The compulsion is intended to alleviate the anxiety associated with the obsession and/ or prevent a feared outcome. When we are assessing people with OCD, this is sometimes described as the 'obsessional sequence'. Over long periods of time people may lose the conscious connection between obsession and compulsion, but it is usually possible to identify a time when 'behaviour B' follows on from 'thought A'.

Sometimes people have 'mental compulsions' (such as counting, reciting phrases) which are not so evident but the key features described above are still present.


Similar to compulsions, rituals are behaviours (or mental acts) that are repeated excessively. They may involve behaviours such as checking, counting, or touching. However, there are some subtle differences which are usually present:

  1. In rituals, there is often less internal resistance to the performance of the rituals. They are, therefore, less 'ego-dystonic' than OCD-based behaviours. Cleaning rituals might be accepted as the best way to do things properly, or to get things fully clean. A 'fully clean' state is often desirable and/ or is the goal of the behaviour.
  2. There is a very specific sequence for the behaviour, which usually has to be followed. You may shower excessively, but if there isn't a particular order it's probably not ritualised. This means that compulsions can be rituals, but not all rituals are compulsions.
  3. There is often a 'realistic' connection between the ritual and the outcome it is meant to prevent. People who have rituals of checking plugs and switches before they go to bed are usually doing it because it makes sense to them, and the amount of time expended is usually considered acceptable. Someone who has rituals relating to showering (for example) will often say that their rituals have a clear purpose: to get fully clean.
  4. In most cases, non-OCD rituals take less time and have less impact upon someone's life. They may take up more time than the person wants, but in OCD people might spent upwards of 6-9 hours engaged in rituals. This is rarely the case in non-OCD rituals.

Whilst many compulsions can be rituals, not all rituals are compulsions.

Safety behaviours

Safety behaviours are not clearly defined in current classification systems, although it is a term that is commonly used by therapists treating anxiety disorders, including GAD. The concept is fairly firmly rooted in cognitive-behavioural conceptualisations of anxiety disorders.

In brief, safety behaviours are more closely related to rituals, and they are commonly performed in order to reduce a sense of worry, distress, or anxiety. In this sense, the purpose of the behaviour is to reduce anxiety and it is the anxiety (rather than a specific feared outcome) that is the primary driver.

They are performed to reduce worry and to make us feel more comfortable in situations where we feel anxiety. Within CBT-based formulations, safety behaviours have problematic consequences such as: preventing us from directly facing or dealing with our fears; they lead to significant avoidance; and they increase the amount of time that we spend focusing on our 'internal' world, rather than addressing 'real' external factors.


Tics have the following features:

  • They are sudden, rapid, non-rhythmic, and recurrent movements or vocalisations;
  • They are involuntary. Although most people can usually suppress tics for a short period, the experience of tics is that they are involuntary and unpleasant;
  • They are not 'goal-directed'. That is, they are not performed with a specific goal/ purpose/ or outcome. This differentiates tics from stereotypies.

Tics can co-occur with compulsions and obsessions, and many people with Tourette syndrome will also have compulsions.


A stereotypy is a regular, repetitive, and non-goal directed movement. Stereotypies are similar to tics in that they are non-goal directed, but rather than being unpredictable and sudden they are repetitive and/ or rhythmic. Examples would include foot tapping or body rocking. Stereotypies can be seen in conditions such as chronic schizophrenia, and are quite common in autism.

Types of symptom that are less likely to be due to OCD

If we accept the above definitions, then certain things are much less likely to fit into the OCD-related 'compulsion' domain. Examples include the following.

'Compulsive' sexual behaviour

Although the repetitive (and sometime harmful) nature of compulsive sexual behaviour can cause problems, the motivation for such behaviour is usually pleasure (but not exclusively so), and there is much less overt (or conscious) resistance to the behaviour. People might regret their actions (and experience harmful consequences) but negative emotions follow on from the behaviour, rather than a thought causing negative emotion which is 'undone' by the behaviour..

However, at the time resistance is hard to reliably identify. Also, there is no obvious link between a feared consequence and the behaviour. ICD-11 classifies 'compulsive sexual behaviour disorder' within 'impulse control' disorders rather than within obsessive-compulsive disorder and related disorders.

Excessive cleaning in obsessive-compulsive (anankastic) personality disorder (OCPD)

Although people may spend significant time ordering, tidying, and cleaning they engage in this behaviour because the outcome (a clean and tidy environment) is a desired one. They are not doing the behaviour to avert a feared consequence, and there is much less resistance. People might resent the amount of time engaged in the behaviour but the actual behaviour itself is not seen as unreasonable. This is commonly referred to as being 'ego-syntonic'.

Checking in OCPD

Checking is fairly common in OCPD. Features that differentiate checking in OCPD from checking in OCD include:

  1. The checking is typically in response to feelings of doing things properly, or in the correct way. For example, making sure that doors are locked, or switches are off.
  2. This feeling of 'right' and 'proper' is resisted less than typical compulsions in OCD, although the individual may resent the amount of time it takes. Despite this, it is done because it is felt as though there is a 'correct way' of doing things.
  3. It is much less likely that clear obsessions are driving the compulsive behaviour. Instead, a sense of moral 'correctness' is more common, and a clear 'obsessional sequence' is not always present.

Overlaps with Autism Spectrum Disorder (ASD)

OCD and ASD often overlap and can be tricky to differentiate. However, the key difference is usually in the relationship between the behaviour and anxiety.

In ASD, whilst rituals may be time-consuming the goal is completion of the ritual and anxiety/ distress occurs due to an inability to complete the rituals; rather than the rituals arising from a preceding obsession. This means that a clear (and triggering) obsession is often not present; as it usually is in OCD.

Although some people will have premonitory worry about whether they will be able to complete their compulsions in particular circumstances, this worry in itself is unlikely to be an obsession since the sequence is one in which the compulsion is the primary focus, and the anxiety arises from a prediction about the compulsion. In OCD the obsession (which generates anxiety) is the driver for the behaviour, not the other way around.

Can you have OCD with just compulsions?

Some of these fuzzy boundaries are highlighted by the clinical description and diagnostic guidelines for OCD in ICD-11. ICD-11 states that OCD is characterised by:

  • "Presence of persistent obsessions and/or compulsions..."; and,
  • "Compulsions are repetitive behaviours or rituals, including repetitive mental acts, that the individual feels driven to perform in response to an obsession..."

Therefore, if compulsions are performed in response to an obsession (as indicated above), a behaviour cannot be a compulsion without an associated obsession. So, you can't have OCD with just compulsions because the definition of a compulsion requires an associated obsession.

Clearly, this is a very concrete approach but the interdependencies between different types of symptom and the boundaries between diagnosis require comprehensive assessment and careful diagnostic reasoning.

Compulsions in Generalised Anxiety Disorder (GAD)

Prevalence of obsessions and compulsions

A study of obsessions and compulsions in GAD found that such symptoms were quite common (Schut et al, 2001). The prevalence of obsessions in GAD was 11% - 14%, and the prevalence of compulsions was 22% - 27%. The most common type of compulsion was checking (50% - 60%), followed by washing (8% - 13%).

Mahoney et al (2001) reported that most patients with GAD (90%) reported at least one maladaptive behaviour that was used to manage their worry. These maladaptive behaviours included: hypervigilance; checking behaviours; avoidance of saying or doing worrisome things.

Rituals in generalised anxiety disorder (GAD)

When GAD is severe, it is common for people to have safety behaviours. These can be ritualised. For example, some people might have certain repetitive behaviours that are intended to reduce anxiety. In OCD, the goal of the behaviour is to avert or prevent feared consequences although anxiety is usually present, the behaviour is clearly related (in type) to the obsession.

In GAD, the goal of the behaviour is to reduce anxiety and it is harder to identify very clear causal and intentional links between a clear obsession and compulsion. Difficulties can arise when someone worries excessively about bad things happening and they may have to check doors, locks and switches. This can make it hard to clearly separate out GAD-type rituals from OCD-type compulsions. Detailed exploration of the timing, sequence, and purpose of the behaviours is required.

GAD vs OCD: types of behaviour

In a small, opportunistic sample of patients with GAD (N=27), OCD (N=26), and controls (N=25), Townsend et al (1999) described compulsions in both GAD and OCD. This was not a highly-symptomatic group of participants. The GAD participants came from a clinical trial, the OCD participants came from an OCD clinic, and control participants can from signs at local University. Of note is that symptom scores (Y-BOCS) were low, with OCD participants only having baseline scores of 8.5 ± 4.1.

Findings included the following:

“GAD patients reported fewer types of behaviors than did OCD patients, with 74% of GAD patients reporting only one type of compulsive behavior compared with 15% of OCD patients”

“...GAD patients were more likely to demonstrate checking behaviors and to do so in the absence of other types of repetitive activity. By contrast the types of compulsions reported by OCD patients were more varied and relatively evenly distributed across categories.”

“In general, GAD patients performed the behavior in response to a worry in order to find reassurance or to provide a distraction from their worries. OCD patients usually describe performing compulsions in a ritualistic manner in direct response to an obsession.”

“The GAD patients did not tend to have rigid rules about how to perform these behaviors.”

GAD vs OCD: mechanisms of checking

In another comparison, Coleman et al (2011) described differences between how symptoms might be generated. Specifically:

“...checking behaviors associated with OCD symptoms arise from the anxiety and distress associated with maladaptive interpretations of intrusions. Specifically, cognitive models of OCD propose that particular beliefs and ways of interpreting intrusive thoughts and images are likely to increase anxiety and urges to check.”

“However, the specific content of the checking differed, with OCD symptoms associated with interpersonal checking but associated most strongly with object checking, and GAD symptoms most strongly associated with interpersonal checking.”

Implications for treatment

Understanding the nature and type of symptom is important for reliable diagnosis and the subsequent targeting of treatment. It has been suggested, for example, that emotional regulation strategies (commonly used for treating patients with GAD) may be helpful for patients with OCD who engage in interpersonal checking. Similarly, exposure-and-response prevention (ERP) approaches for OCD might be helpfully adapted to treat interpersonal checking in GAD.

This doesn't mean that the two conditions are so similar that treatments for one can be used for the other, but some treatment approaches might be helpful for particular types of symptom. This means that a 'symptom-focused' approach might be more helpful than a 'diagnosis-based' approach in some circumstances. However, the evidence to support this proposal is not available and having a robust diagnostic formulation is likely to be a key part of any treatment approach.

Comparison table

The table below summaries some of the above discussions. No single characteristic is determinant of one type of repetitive behaviour versus another, but looking at the overall pattern of behaviour should help in differentiating them. Please note that whilst compulsions are frequently ritualistic, not all rituals are compulsions.

Key: ++ Strongly characteristic; + Moderately characteristic; - Not strongly characteristic.

Feature OCD-related compulsions Non-OCD rituals (e.g. in ASD) GAD-related safety behaviours
Presence of clear obsession(s) that cause anxiety, and which are neutralised by the behaviour ++ - +
The anxious thought/ worry/ doubt precedes the behaviour (Thought -> Anxiety -> Behaviour) ++ - +
Distress arises subsequent to the inability to complete the compulsion (Behaviour -> Anxiety) + ++ +
The worry (and associated anxiety) is focused on the outside world, rather than the inner world ++ - -
Real-life worries precede the behaviour, and the behaviour is intended to reduce anxiety rather than neutralise a feared consequence - - ++
Significant time spent in repetitive behaviour, or very significant impact upon functioning ++ ++ +
Resistance to the repetitive behaviour ++ - +
Checking related to external world (e.g. harm) ++ - -
Checking related to interpersonal issues - - ++

Assessing obsessions and compulsions in OCD and OCPD

Key questions

Given the above, there are some useful questions that can help differentiate between the different presentations of intrusive thoughts and repetitive behaviours. The wording is chosen to avoid assuming that all intrusive thoughts are obsessions and all repetitive behaviours are compulsions.

  1. What is the underlying intrusive thought?
  2. Is the intrusive thought consistent with themes commonly seen in OCD (for example, harm to others, contamination)?
  3. What is the feared outcome?
  4. Is the intrusive thought clearly linked (both in theme and sequence) to the feared outcome?
  5. Is the repetitive behaviour intended to: a) prevent the feared outcome; b) undo the feared outcome; or, c) provide reassurance that the feared outcome has not happened?

To illustrate this, we will use two examples. First, harm obsessions consistent with OCD and then, second, perfectionism associated with OCPD.

Typical pattern in OCD

In OCD, the person has an intrusive thought about stabbing a family member. It is a common type of obsession for OCD. The feared outcome is that a loved one will be harmed by the person, and that the person will be responsible for a terrible outcome. The thought (stabbing a family member) is linked closely to the feared outcome (harm to others); particularly in terms of the sequence. The repetitive behaviour is checking/ seeking reassurance that they haven't actually harmed the family member, and they might keep detailed notes of their activities to demonstrate to themselves that they couldn't have caused harm.

Typical pattern in OCPD

In OCPD, the person has an overwhelming need for things to be perfect, untarnished, and untouched. If things are damaged, they are considered to be 'dirty' or imperfect. Even though they talk about 'contamination', they are not referring to bodily fluids or dirt or germs: the concept of 'dirty' is an abstract one and it is related more to objects losing their original quality than actually being covered in germs. The feared outcome is hard to establish and doesn't involve obvious consequences for other people. Instead, the avoided outcome is a personal sense of unpleasantness and distress that might occur if things are perceived to be 'dirty'. Importantly, the thought ('things will be dirty') is different from the 'things are dirty and will cause harm' that is more common in OCD. Finally, the repetitive behaviour does not follow on from the intrusive thought; rather, the behaviour is performed in order to prevent the thought (and the associated distress) from occurring. What's trying to be prevented is not a clearly-defined negative outcome, but is a future-based (rather than a current) negative emotional state.


DE CALUWÉ E, VERGAUWE J, DECUYPER M, BOGAERTS S, RETTEW DC, DE CLERCQ B. The relation between normative rituals/routines and obsessive-compulsive symptoms at a young age: A systematic review. Developmental Review. 2020; 56: 100913.

AMERICAN PSYCHIATRIC ASSOCIATION. Diagnostic and Statistical Manual of Mental Disorders (Third Edition - Revised). Washington, DC: American Psychiatric Press; 1987.

SCHUT AJ, CASTONGUAY LG, BORKOVEC TD. Compulsive checking behaviors in generalized anxiety disorder. Journal of Clinical Psychology. 2001; 57(6): 705-715.

MAHONEY AEJ, HOBBS MJ, NEWBY JM, WILLIAMS AD, ANDREWS G. Maladaptive Behaviours Associated with Generalized Anxiety Disorder: An Item Response Theory Analysis. Behavioural and Cognitive Psychotherapy. 2018; 46(4): 479-496.

TOWNSEND MH, WEISSBECKER KA, BARBEE JG, PEÑA JM, SNIDER LM, TYNES LL, TYNES SF, BOUDOIN C, GREEN-LEIBOVITZ MI, WINSTEAD DK. Compulsive Behavior in Generalized Anxiety Disorder and Obsessive-Compulsive Disorder. Journal of Nervous and Mental Disease. 1999; 187(11): 697-699.

COLEMAN SL, PIETREFESA AS, HOLAWAY RM, COLES ME, HEIMBERG RG. Content and correlates of checking related to symptoms of obsessive compulsive disorder and generalized anxiety disorder. Journal of Anxiety Disorders. 2011; 25(2): 293-301.


Last Updated on 25 November 2023 by David Christmas
Skip to content