The story of memory and executive functions in obsessive-compulsive disorder: a case-control study

Abstract Objective Neuropsychological findings in obsessive-compulsive disorder (OCD) are mainly clustered around the role of memory and executive functions. However, outcomes vary across different OCD populations. In addition, the extent to which each of these factors can distinguish patients with OCD (PwOCD) from healthy individuals remains uncertain and attracts great attention. The present study aims to investigate the above issues. Method This was a cross-sectional study of 182 individuals (90 PwOCD and 92 matched healthy controls). After screening for inclusion and exclusion criteria, the participants were administered neuropsychological tests including, the Wechsler Memory Scale-III (WMS-III), the Wisconsin Card Sorting Test (WCST), and the Stroop Color-Word Test (SCWT). Data were analyzed to test the study hypotheses using comparison of means and regression analysis methods. Results The results showed that PwOCD had poorer performance than the control group in Immediate Memory, General Memory, and Working Memory and also according to response inhibition indexes. The results also showed that General Memory and Reaction Time2 from the SCWT index could be predictive variables for discriminating between PwOCD and controls. Conclusion The findings of this study support the prior assumptions that PwOCD would have impaired memory dimensions and response inhibition, but did not support worse set-shifting performance. We also present an initial model for the predictive role of these neuropsychological variables in discriminating OCD from healthy individuals and increasing diagnostic accuracy.


Introduction
Obsessions are repetitive thoughts, urges, or images that are intrusive and unwanted, and in most people cause anxiety or distress. Compulsions are repetitive mental or overt acts that are experienced as being urge-driven either in response to an obsession or according to a rule that must be applied decisively and are aimed at preventing or reducing anxiety or distress or preventing a feared event happening. 1 The diagnostic and statistical manual of mental disorders, 5th edition (DSM-5), states that these symptoms are not specific to obsessive-compulsive disorder (OCD) and at least 14 other disorders have symptoms broadly meeting these criteria such as body dysmorphic, trichotillomania, skin picking, generalized anxiety, depressive, and addictive disorders. 2 Nonspecific symptoms are not exclusive to OCD, but are common to almost all psychiatric disorders. Due to these not very precise boundaries, the National Institute of Mental Health (NIMH) has launched its own Research Domain Criteria (RDoC), which suggest a framework to define clinical disorders based on cognitive, physiological, molecular, and genetic levels. 3 Based on this view, a promising approach to clarify the neurobiology of psychiatric disorders is to seek structures that can be placed somewhere between lowlevel causative agents and phenotypic manifestations of the disorder, called the endophenotype. 4,5 A trait must have several characteristics to be considered as an endophenotype; it must be associated with the disease, heritable, state-independent, and found in unaffected family members of a patient more frequent than in the general population. 6 Memory is one of the cognitive domains that have attracted the most research attention. Patients with OCD (PwOCD) showed shortcomings in short-term memory, working memory, and visual-spatial abilities. 7 However, findings are still remarkably inconsistent. A group of studies have found no significant impairment in memory dimensions, while the others, which constitute a larger proportion of the studies, emphasize frequent memory impairments in PwOCD. Among the second group, identification of the memory aspects with the most disturbances in OCD is still controversial. Although, non-verbal memory such as long-term and immediate visuospatial memory problems have been consistently demonstrated in OCD, the findings related to verbal memory performance are relatively inconsistent. 8,9 Some researchers reported intact working and declarative verbal memory and some others have reported difficulties in recalling verbal episodic material. 7,10,11 A group of studies have suggested that defects in memorizing and recalling are due to defects in the organization of information and emphasize the role of executive functions in memory impairments in OCD. 12,13 Disturbances in several features of executive functions have been proposed as endophenotypes of OCD. 14 Impaired performance in working memory, attentional set-shifting ability that underlies cognitive flexibility, and response inhibition have been repeatedly reported by many studies, even after controlling for age, gender, and education. 9,14-18 Set shifting and response inhibition are assumed to be responsible for perseveration and repetitive behaviors. Defects in these areas could be the basis of impulsive and compulsive acts common in patients with OCD. 19,20 A meta-analysis of studies of cognitive functioning in OCD reported large effect sizes for nonverbal and visuospatial memory. In contrast, set-shifting, cognitive inhibition, and verbal memory had medium or smallto-medium effect size. Working memory results are more inconsistent, varying from small to large effect sizes. 7,21,22 The abovementioned studies in this area were mostly performed on smaller samples using simpler tools. So, in this study with a relatively large sample and using more comprehensive tools we will assess substantial areas of memory and executive functions in a large group of patients with obsessive-compulsive disorder.
Moreover, by analyzing mediating neuropsychological features, we will estimate each variable's contribution to developing OCD and the extent to which they can discriminate PwOCD from healthy individuals.
Although exact determination of this pattern is very complicated because of the high rates of comorbidity and dimensional distribution of traits in psychiatric disorders, we attempt to discover a preliminary model among a group of obsessive-compulsive patients.

Methods and materials Participants and procedure
The study participants were 182 individuals, comprising 90 patients with obsessive-compulsive disorder and 92 healthy controls. The sampling process is described in Figure 1. Healthy control subjects (students and staff at USWR and TUMS) (n = 108) Not willing to participate in the study (n = 5)

Screening against inclusion and exclusion criteria (n=90)
Consented to participation in the study (n = 104) Not willing to participate in the study (n = 4) Screening against inclusion and exclusion criteria (n = 98)

Final control sample (n = 92)
Matched to the OCD group for demographic variables Consented to participation in the study (n = 107)

Final OCD sample (n = 90)
Excluded Excluded   These data are presented in detail in Table 1.

Between-group comparison analysis
The results of the mean comparisons, standard deviation, and t test or Mann-Whitney U test between the two groups are shown in Table 2.

Wechsler Memory Scale-III (WMS-III). As
shown in Table 2, the OCD group showed a significantly poorer performance compared to the control group in Audio Immediate (p < 0.001), Visual Immediate (p < 0.001), Audio Delayed (p < 0.001), and Visual Delayed (p < 0.001) and in the three main indexes

Logistic regression (LR)
Logistic regression was employed to develop a hypothesized model for the role of each variable in distinguishing the OCD group from the healthy group.
As the results of Table 3 show, among the six variables entered into the model, "Response time" from the SCWT and "General Memory" from the WMS-III can significantly predict discrimination. The Hosmer and Lemeshow test (chi-square = 11.2, p = 0.17) showed the model had appropriate fit.
The results of statistical analysis showed that the model has an accuracy rate of 67% (Classification accuracy = 67.6) and can account for 18% (Nagelkerke R Square = 0.18) of differentiation between the two groups.  Therefore, since the current study's working memory task was letter-number sequencing and spatial span, which is considered a complex task, it was expected that PwOCD would have greater difficulty with it.

Limitations and future directions
Since all of the PwOCD in this study have been on medication for years, cognitive disturbances could be a predictable side effect. Hence, the memory and

Disclosure
No conflicts of interest declared concerning the publication of this article.