Table 1.

Information on the five main studies described.

References	Study population	Design and measures	Selected outcomes reported
Julien et al. (12),	N = 33 individuals with OCD, N = 90 non-clinical participants	Measure: Intrusion Inventory II: self-report and expert ratings. Part A: Occurrence of intrusions. Report frequency of each of 68 intrusions. Part B: Context of occurrence. Selected three most disturbing intrusions experienced in life, indicate context link of these (direct, indirect, no link). Context link compared between groups.	↑ intrusions and distress in individuals with OCD compared to non-clinical participants (HC) Context: Proportion of direct and indirect links reversed in the two groups:  • Direct links: OCD < HC (1/3 vs. > 1/2)  • Indirect links: OCD > HC (1/2 vs. > 1/3  • No link: OCD > HC (16% vs. 8%) Context and content: Context link differed significantly depending on the content of intrusions.  • More direct links to context: symmetry, hoarding  • More indirect links to context: rumination, washing
Audet et al. (19),	N = 248 non-clinical participants	Measures: Vancouver Obsessional Compulsive Inventory (VOCI; 21), Fear of Self Questionnaire (FSQ; 22), Ego Dystonicity Questionnaire (EDQ; 23), Inferential Confusion Questionnaire Expanded Version (ICQ-EV; 80); Obsessive Beliefs Questionnaire Threat, Responsibility, Importance and control of thoughts, Perfectionism/uncertainty (OBQ-TRIP; 24) Report an intrusive thought plus two items from Personal Significance Scale (25): 1. “What caused these thoughts to occur when they started?”, 2. “Why do you think these thoughts keep coming back?” (assessment of context) Clinicians rating on 1) content (i.e. checking/contamination), 2) ego-dystonicity (yes/no), 3) evidence (with direct evidence/no direct evidence), 4) OCD-Relevancy (yes/no)	Frequencies:  • Intrusive thoughts contents: unacceptable thoughts and impulses (46.4%), contamination (6.5%), checking (15.4%), other (28.2%)  • Context links: without evidence (68.9%), with direct evidence (26.6%)  • Ego-syntonic (61.9%), ego-dyston (32%),  • OCD-relevancy: OCD relevant (32.4%), not OCD relevant (66.4%) Association context links, ego-dystonicity and symptoms:  • 95.9% of ego-dystonic intrusions occurred without evidence  • ↑ OCD symptoms, inferential confusion, OCD related beliefs and ego-dystonicity in participants having intrusions without evidence  • Ego-dystonicity not related to OCD symptoms  • Thoughts without evidence had a similar likelihood to be OCD-relevant (47.2%) or non-OCD-relevant (52.8%)  • Thoughts with evidence were almost always non-OCD relevant (98.5%) → Much higher likelihood of thoughts without evidence being OCD-relevant than thoughts with evidence
Fradkin and Huppert (13),	Study 1: N = 170 non-clinical participants Study 2: N = 180 (OCTQ + other questionnaires) and N = 211 (OCTQ only) non-clinical participants Study 3: N = 126 non-clinical participants	Study 1: Out-of-context thoughts questionnaire including the subscales Cognitive Context, Affective Context, Self Context and Resisted Thoughts (OCTQ, 13), Obsessive-Compulsive Inventory-Revised (OCI-R; 26): total score, obsessing subscale, and checking, ordering and washing subscales utilized Study 2: OCTQ (final 12-item scale), OCI, Depression anxiety and stress scale (DASS-21; 27), Big Five Inventory 10 (BFI-10; 28), Mind Wandering Deliberate (MW-D) and Spontaneous (MW-S) scales (29) Study 3: modified version of OCTQ (provide two examples of either negative or non-negative OOCTs and answer questionnaire on the basis of thoughts of particular valence, OCI-R (26)	Study 1: Frequencies: 49.41% experience mostly negative out-of-context thoughts (OCT), 39.41% mostly neutral OCT, 7.05% mostly positive OCT.  • Positive association between OCD symptoms (OCI-R total) and OCTQ total and subscales  • OCI-R obsessions subscale sign. associated with OCTQ  • Non-significant association between OCI-R subscales associated with reactive obsessions and OCTQ  • Negative OOCT → more negative emotional responses Study 2:  • Sign. positive association OCTQ and OCI-R  • ↓ correlations of reactive obsessions with OCTQ than obsessing subscale (as in study 1)  • Spontaneous mind wandering (MW-S) and OCTQ total and subscores: Sign. positive association  • ↓ correlations between OCTQ and BFI (divergent validity)  • Correlations OCTQ and OCI-R not moderated by valence. Study 3:  • ↑ OCTQ scores (and higher OCI-Rs) predicted ↑ negative emotional responses, stronger tendencies to resist and more appraisals regarding ones personality for negative OOCTs → reduced predictability of thoughts associated with more aversiveness  • Strong association between OCTQ scores for negative and non-negative OOCTs → Experience of OOCTs is independent of thoughts valence
Audet et al. (20),	N = 557 non-clinical participants	Task: Contextual vignettes task (CVT): 10 vignettes with different themes and two levels of evidence for their truth. Themes: aggressive, homosexual, religious/social convention, contamination, and checking). Each theme once combined with direct evidence and without direct evidence. Indicate which intrusions (from a list) would be experienced after each scenario and likelihood of having these, related distress and unwantedness from 0 to 100. Measures: Depression Anxiety Stress Scale 21 Item Version (DASS-21; 30); only depression subscale used); Vancouver Obsessional Compulsive Inventory (VOCI, 21), Feared Self Questionnaire (22) Inferential Confusion Questionnaire Expanded Version (ICQ-EV; 31) Obsessive Beliefs Questionnaire (32)	• All dependent variables significantly different from 0 → CVT simulates situations in which intrusions occur  • Scenarios with direct evidence associated with sign. ↑ intrusions, and ↑ ratings of likelihood, distress, and unwantedness  • Scenarios without direct evidence predicted OCD symptoms (predictors: number of intrusions, probability of these, distress, unwantedness)  • Inferential confusion predicted likelihood, distress and unwantedness of intrusions without evidence
Llorens-Aguilar et al. (14),	N = 68 individuals with OCD	Measures: Obsessional Intrusive Thoughts Inventory (INPIOS; 33) frequency unwanted obsessional intrusive thoughts, images, impulses and appraisals and control strategies of these. Divided in autogenous (e.g. aggressive, sexual) and reactive (e.g. contamination). Choose most upsetting one and rate it based on emotional reactions, difficulty in controlling it and dysfunctional appraisals. Report frequency of control strategies used. Interview: Semi-structured interview of obsessions and intrusive thoughts in OCD and INPIOS-2nd part (33): Describe intrusion constituting obsessions and intrusion, which never became obsession and other features (e.g. theme) of it. Elaborate functional consequences of these.	Frequency thoughts:  • Obsessions > intrusions Thought content and reactions:  • 1/2 autogenous, 1/2 reactive  • No content differences between obsessions and intrusions  • Obsessions more inconvenient than intrusions, associated with ↑ anxiety and sadness, ↑ difficult to control, associated with ↑ dysfunctional appraisals Context:  • First occurrence obsessions: negative mood, higher stress, relevant previous stressful life event (same pattern for 1/2 the intrusions)  • Most obsessions and intrusions occurred with indirect link (obs: 49.3-67.9%, in: 66.7-71.1%), some with no link (obs: 25-47%, in: 21.8-26.7%), few with direct link (obs: 3-7.1%; in: 6.7-7.9%)  • Context of occurrence of obsessions significantly changed from onset to last time experienced: onset most often indirect, last occurrence indirect or no link (no context differences between first and last occurrence for intrusions)  • Content and context: Onset: Autogenous and reactive obsessions and intrusions indirectly linked. Last time experienced: Autogenous obsessions and intrusions more frequently occur with an indirect link, while reactive obsessions are experienced with no link (obsessions) or indirect/no link (intrusions)

↑ = increased; ↓ = decreased; → = either "evoked/leads to" or it indicates a conclusion.