INTRODUCTION
Individuals with obsessive-compulsive disorder (OCD) consistently experience unwanted intrusions and intense fear and anxiety. OCD is characterized by recurrent and persistent thoughts and repetitive behaviors that cause significant distress (American Psychological Association, 2013; Williams & Steever, 2015). Given the difficulty of controlling anxiety, individuals with OCD find immediate relief in the repetitive performance of compulsions or rituals that end up increasing distress and reinforcing fear and avoidance (Van Noppen et al., 2006). This disorder is known to affect approximately 1-3% of the population worldwide (Abramovitch et al., 2015; American Psychological Association, 2013; Pallanti et al., 2011). The mean age of onset of OCD is 19.5 years, and it affects males and females, although males have an early start (Ruscio et al., 2010). OCD can impair academic, occupational, and social functioning, with this last domain being significantly reduced according to a study by Ruscio et al. (2010). Some of the risk factors for developing OCD include: being an older adolescent, trauma, substance abuse, depression, phobic disorders, ADHD, tic disorders, prior history of separation anxiety disorder, and family history for OCD (Fontenelle & Hasler, 2008).
Our limited understanding of the neurobiological and genetic basis of OCD remains a significant barrier that hinders elucidating the underlying mechanisms of this complex illness. OCD has been linked to hyperactivity and dysfunctional interactions, particularly in the orbitofrontal cortex (OFC), which may lead to obsessions and compulsions. However, studies have been inconsistent regarding the hyperac-tivity of the OFC in OCD (Wood & Ahmari, 2015). Disruption in the interactions of other brain areas, such as the ventral striatum (VS), the anterior cingulate cortex (ACC), and the ventromedial prefrontal cortex (vmPFC) has also been found to produce and maintain OCD symptoms (Haber & Heilbronner, 2013; Wood & Ahmari, 2015). Although the genetic basis for OCD is unknown and still being studied, twin studies have demonstrated that there is a heritability estimate that ranges between 27-65% (Pauls et al., 2010; van Grootheest et al., 2005). A fundamental question remains on whether genetic variations are associated with OCD, having a further impact on treatment outcome (Qin et al., 2016).
In Puerto Rico, approximately 2 in 10 adults are diagnosed with a psychiatric disorder, and the 12-month prevalence rate is 18.7% (ASSMCA, 2016). A recent study reported a 12-month prevalence rate of 22.5% of having any psychiatric disorder, including anxiety disorders, among Puerto Ricans living in Puerto Rico (Canino et al., 2019), suggesting a high incidence of psychopathology. In terms of OCD, there is a lifetime prevalence rate of 2.3% of adults with OCD in the United States (Ruscio et al., 2010) and 3.2 % in Puerto Rico (Canino, 1987). According to Weissman et al. (1994), the lifetime prevalence rates of OCD were consistent in the United States (2.3%), Edmonton (2.3%), Puerto Rico (2.5%), Munich (2.1%), Taiwan (0.7%), Korea (1.9%), and New Zealand (2.2%), with Puerto Rico’s rates being higher than the other countries. Obsessions and compulsions in OCD may not be limited to a specific culture. Still, given the variability of symptoms, cultural factors may play a role in the development of OCD, since obsessions and compulsions can be expressed and perceived differ-rently. Individuals from a particular culture can have concerns that may be reflected in their OCD symptoms (De Silva, 2006; Wheaton et al., 2013). In a comparative study of OCD subjects from Costa Rica and the United States, Chavira et al. (2008) found that the content of the obsessions and compulsions of North Americans and Costa Ricans was similar leads to the possibility that OCD may have a biological basis.
Moreover, Himle et al. (2008) found similar lifetime and 12-month rates of OCD (1.49%), OCD severity, and high standards of co-morbid psychiatric disorders between African Americans and Caribbean blacks. They also observed that mental health service use over a year for African Americans and Caribbean blacks with OCD was extremely limited in the United States (only 20% of patients with OCD received treatment). These studies highlight a pressing need to perform research and implement evidence-based treatment targeting minority populations that suffer from OCD.
Cognitive-behavioral therapy (CBT) with exposure and response prevention (ERP) have shown strong empirical support for the effective treatment of OCD (Geller et al., 2019; Leeuwerik et al., 2019). Although it is a challenging therapy, CBT is a structured and goal-oriented intervention that helps patients to re-evaluate dysfunctional intrusive thoughts that trigger intense anxiety. ERP is the main component of CBT, and it is based on the principles of fear extinction (Fullana et al., 2013). During fear extinction, an individual is exposed to repeated pairings of a neutral stimulus with an aversive stimulus. Then, it is exposed to the neutral stimulus in the absence of the aversive stimulus until the fear-conditioned response diminishes or disappears. In general terms, fear extinction generates an inhibitory memory that may suppress the manifestation of fear associations (Quirk et al., 2010). Research lacks on fear extinction in OCD, and associative learning processes seem to be significant in the maintenance of this complex illness, which warrants its importance.
Furthermore, recent neuropsychological studies of OCD have shown correlations between the activation of specific brain regions and cognitive function and behavioral changes (Nakao et al., 2014). In a systematic review, Kuelz et al. (2004) evaluated several studies and found that diverse neuropsychological deficits are present in OCD, including attention, executive functions, visuospatial abilities, and verbal and non-verbal memories, although no study has found a specific neuropsychological profile for OCD and researches on different cognitive functions, such as set-shifting, fluency, planning, and problem-solving abilities are contradictory. Moreover, a meta-analysis showed poor neuropsychological performance in individuals with OCD compared to healthy individuals (Abramovitch et al., 2013). However, Simpson et al. (2006) found that subjects with OCD performed similarly to healthy controls on most neuropsychological tasks that measured set-shifting, spatial working memory, and motor initiation and execution. Their results also showed that the severity of OCD had little impact on neuropsychological performance (Simpson et al., 2006). The results of these studies emphasize the importance of clarifying the neuropsychological effect of OCD to comprehend its pathophysiology better.
OCD in Latino/Hispanics is understudied and should be addressed given the diverse manifestations of the disorder and the psychosocial stressors (e.g., limited access to mental health services) imposed on this population. Another factor that warrants attention, especially in Puerto Rico, is the lack of knowledge of OCD by healthcare providers (Rodríguez-Acevedo et al., 2009), which may often lead to misdiagnosis and influence treatment outcomes (Cordioli & Vivan, 2012, Wetterneck et al., 2012). In this study, we aim to observe the physiological responses of a Puerto Rican sample with OCD during fear-extinct learning and their performance on neuropsychological and psychological assessments. We hypothesized that individuals with OCD would have impaired fear recall and extinction, as measured by a 2-day fear conditioning and extinction paradigm. We also hypothesized that OCD subjects would show deficits in neuropsychological functioning, such as attention, concentration, memory, and processing speed, as measured by several neuropsychological assessments. Besides, we hypothesized that OCD subjects would reveal psychological markers, such as anxiety and depression symptoms, compared to healthy individuals. The findings of this study may increase awareness and understanding of this complex disorder and its mechanisms. They can also enable the exploration and development of culturally-adapted assessments and interventions for the Puerto Rican population.
Fear Conditioning and Extinction in Humans
OCD is commonly treated with CBT and ERP, which is based on the theory of fear extinction (McLaughlin et al., 2015) and is therefore focused on extinguishing compulsive behaviors. Fear conditioning and extinction are associative models of learning that are used to predict fear. While fear conditioning is training the organism (rodent or human) to pair an aversive stimulus (unconditioned stimulus, US) with a sensory stimulus (conditioned stimulus, CS), extinction is training the organism to learn not to fear; to split that association by creating an inhibitory memory that temporarily suppresses fear (Quirk et al., 2010). During classical conditioning, the US follows the CS during training, which results in a conditioned response (CR) that occurs after the presentation of the CS. For example, a tone (US) is paired with the presentation of food (CS), which produces salivation (CR). By associating the tone with the presentation of the food, a person can start salivating at the sound of a tone.
In an experiment conducted by Milad et al. (2013), they used a classical fear conditioning and extinction paradigm, in which participants with OCD and healthy controls paired pictures of colored lamps (red and blue; CS) with an aversive electrical stimulation (US) and measured fear expression through skin conductance responses (SCR). They found that although individuals with OCD had intact fear conditioning and extinction, they had deficits in recalling the extinction memory (Milad et al., 2013). By using functional magnetic resonance imaging (fMRI), they observed a decreased activation of the ventromedial prefrontal cortex (vmPFC) in OCD subjects during every phase of the experiment, compared to healthy controls. They found a positive correlation between OCD symptom severity, which was measured by the Yale-Brown Obsessive-Compulsive Scale (YBOCS), and extinction recall. In other words, the more severe the symptoms, the higher the ability to recall extinction memory. Milad et al. (2013) believe that this unexpected finding may be associated with the coping mechanisms being used by severely affected OCD patients to suppress fear. In another study, McLaughlin et al. (2015) showed that participants with a lifetime diagnosis of OCD had poor extinction recall, compared to healthy participants. In contrast to the study conducted by Milad et al. (2013), they did not find correlations between symptom severity and fear extinction recall. Together, these studies provide suggestive evidence in that deficits in fear extinction recall may be an OCD trait (Milad et al., 2013; McLaughlin et al., 2015), although conclusive findings are warranted in future studies.
Some research on extinction retention has included Latinos, and ethnicity may impact the physiological responses of fear (Martínez et al., 2014). For this reason, in our study, we seek to evaluate fear extinction in Puerto Ricans with OCD. Based on previous studies on their difficulty suppressing fear, we hypothesized that a sample of Puerto Ricans with OCD would show impaired extinction recall. We also assessed fear renewal, to evaluate the memory of fear conditioning.
Given that the most common and effective treatment for OCD is cognitive behavioral therapy and ERP, evaluating extinction recall and extinction through physiological responses in a group of Puerto Ricans with OCD could lead to necessary implications in their treatment outcome. If participants with OCD can respond differently to the conditioned stimuli during the extinction learning phase of the experimental task, this may be a useful treatment response. For this reason, the findings of this study can be helpful and practical in explaining OCD symptom development, maintenance, and treatment guidance.
Neuropsychology of OCD
Research suggests deficits in attention, executive functioning, and processing speed as characteristics of patients with OCD, which may play a mediating role in brain dysfunction and the prevalence, maintenance, and severity of clinical symptoms (Saremi et al., 2017). Identifying neuropsychological deficits' nature can lead to understanding OCD's psychopathology of OCD, which can be useful in the development of effective prevention and treatment programs.
Neuroimaging studies in OCD have identified abnormalities in areas, such as the OFC, anterior cingulate cortex (ACC), and caudate nucleus (Nakao et al., 2014). Impairments in neural structures and circuits can influence cognitive and behavioral changes. Although several studies suggest underperformance of individuals with OCD in neuropsychological assessments (Abramovitch & Cooperman, 2015; Nakao et al., 2014), especially before treatment (Voderholzer et al., 2013), evidence on the neuropsychology of OCD has been divergent. Millet et al. (2013) used a neuropsychological battery to assess the executive functions of OCD outpatients with severe symptomatology. OCD outpatients showed lower verbal fluency scores, fewer correct responses on a test that measures selective attention and cognitive flexibility (e.g., Stroop). They took more time to complete a task that measures executive functions, such as planning (e.g., Tower of London) than individuals without OCD (Millet et al., 2013). In another study, Da Rocha et al. (2011) evaluated decision-making in patients with OCD by using the Iowa Gambling Task (IGT). They found that patients with OCD performed significantly worse on the IGT and showed deficits in decision-making, compared to healthy subjects (Da Rocha et al., 2011). Other studies have found deficiencies in tasks that require set-shifting (Tukel et al., 2012) and response inhibition in individuals with OCD (Pena-Garijo et al., 2010). However, Kohli et al. (2015) assessed verbal intelligence, memory, perceptual and motor functions in OCD outpatients using tests, such as the Wisconsin Card Sorting Test (WCST) and the Bhatia Battery of Performance tests of Intelligence (BSS) and participants did not show impairments in any of the cognitive functions, compared to healthy controls. The controversial findings of neuropsychological functions in individuals with OCD enhance the need to assess these functions, especially in Puerto Ricans, given the lack of research in Latinos with OCD (Williams et al., 2010). Assessing the neuropsychological functioning of Puerto Ricans with OCD can allow the identification of potential neurocognitive deficits for further rehabilitation to improve daily living and academic/occupational skills, as well as an understanding of the neurocognitive manifestations of OCD.
Psychological Assessment of OCD
OCD is strongly associated with elevated anxiety, which affected individuals to engage in maladaptive strategies to reduce it. The repetitive behaviors and rituals may have an impact on how individuals with OCD regulate their emotions. In a study, event-related brain potentials were recorded to observe how healthy controls and individuals with OCD responded to emotional stimuli by using distraction and cognitive reappraisal (Paul et al., 2016). It was found that when participants with OCD used cognitive reappraisal to reduce emotional responses elicited by OCD-related and general aversive pictures, they failed to show a reduction in the amplitude of the electrophysiological responses, as measured by late positive potentials, compared to healthy controls (Paul et al., 2016). Interestingly, at a self-report level, they reported reduced arousal. These results demonstrate that participants with OCD showed they marked ongoing processing at an electrophysiological level, which may be associated with their difficulty in disengaging their attention from emotionally relevant stimuli. In another study, using an undergraduate sample, Stern et al. (2014) found that OC symptoms were correlated with a poor understanding of emotions (positive or negative) and greater fear of experiencing them. This fear of experiencing emotions experienced by individuals with OC symptoms may be related to the fear of losing control (Stern et al., 2014). In another study, Macatee et al. (2013) evaluated distress tolerance or the capacity to experience aversive emotional states in an OCD sample. They found that although an OCD diagnosis was not a predictor of distress tolerance to obsessions, having higher obsessions, unlike other OCD symptoms, were associated with reduced distress tolerance (Macatee et al., 2013), which can contribute to a lack of emotional regulation. These studies are consistent with previous findings on emotion regulation in individuals with OCD, regarding the difficulty of maintaining control of behavior when they experience an emotion, mostly negative (Allen & Barlow, 2009; Fergus & Bardeen, 2014).
Furthermore, since the fear of anxiety-related sensations or anxiety sensitivity seems to be elevated in OCD, Blakey et al. (2017) investigated if this construct serves as a predictor of treatment outcome after cognitive-behavioral therapy (CBT). The results showed that anxiety sensitivity was positively correlated with baseline and posttreatment OCD severity, suggesting that anxiety sensitivity predicts treatment outcome (Blakey et al. 2017). These studies indicate that psychological constructs can provide valuable information regarding how individuals perceive their experiences with such a complex disorder, which can thus be compared with other measures, such as physiological and neuropsychological assessments, to obtain a better understanding of the manifestations of OCD.
METHOD
Participants
A total of 17 healthy controls (mean age 35.76±13.39 years old) and 11 individuals with OCD (mean age 32±11.84 years old) from Puerto Rico, aged 21 to 60, participated in the study. The participants were recruited through local advertising (flyers, posters, conferences). Interested individuals were screened via telephone for OCD symptoms before taking part in the study and potential participants were interviewed to assess OCD diagnosis. A written informed consent was obtained from participants before the study. The inclusion criteria were: (a) age 21-60 years, (b) Puerto Rican descent, (c) Spanish speaking, (d) normal or corrected-to-normal color vision, and (e) a diagnosis of OCD confirmed by SCID-IV interview for the experimental sample. Exclusion criteria included (1) medical conditions that could be affected by the fear conditioning paradigm (e.g., arrhythmias, heart failure), (2) history or active drug, alcohol or substance dependence, and (3) history of any neurological disorder or head trauma, as they could interfere with neuropsychological test performance. The study was performed at the Center for the Study and Treatment of Fear and Anxiety (CETMA for its Spanish acronym) with Institutional Review Board (IRB) approval (A5280109) from the University of Puerto Rico, Medical Sciences Campus.
Measures and Procedure
The study had an observational and cross-sectional design, which allowed the comparison of two groups at a single time. It was divided into two days due to the extended duration of the assessments. During Day 1, after participants were recruited and provided informed consent, demographic information was obtained, the clinical interview using the SCID-IV was completed, and the psychological questionnaires were administered. During Day 2, the neuropsychological assessment was performed. Fear conditioning and extinction were performed on Day 1, and extinction retention was tested on Day 2.
Physiological Assessment
We measured fear responses through skin conductance responses (SCR). These responses were recorded using a Coulbourn Isolated Skin Conductance Coupler (S71-23) (Coulbourn Inst., Allentown, PA). A current of 0.5 V was passed through 8mm diameter Ag/AgCl radio translucent electrodes (BioPac Systems Inc., Goleta, CA) filled with isotonic paste. The electrodes were placed on the palm of the participant’s non-dominant hand, and each was separated from another by 14mm. A Coulbourn Lablinc Analog-to-digital converter (V19-16) digitalized the analog signal stored in the computer. Changes in skin conductance levels were calculated for each conditioned stimulus (CS) trial by subtracting the mean skin conductance level during 2 seconds immediately before CS onset from the highest skin conductance level recorded during the 6 second CS duration. This procedure has been validated (Milad et al., 2005), and it facilitates the detection of the maximal increases in skin conductance levels during the 6-second presentation. This procedure has also been carried out effectively in the past with Puerto Rican healthy and OCD samples (Martinez et al., 2014). The average of SCR was quantified to determine the conditioned fear responses. The outcome measures that were calculated were: peak SCR during conditioning, average SCR during the first two trials of extinction, and percent fear during recall and renewal phases (average SCR during the first two trials of the period divided by the peak SCR during conditioning).
Neuropsychological Assessment
Wisconsin Card Sorting Test (WCST). The WCST is a neuropsychological test created by Grant and Berg (1948). It is used to measured cognitive functions, such as set-shifting and impulsive responses. In the WCST, participants sort cards according to characteristics: color (red, green, blue, or yellow), form (circles, stars, squares, or crosses), and number (1, 2, 3, or 4). They will try out ways to find the correct method for sorting the cards and are told if they correctly or incorrectly guess the rule. The computer calculates the scores of the participant, which are the number of runs of 10 correct scores and the perseverative errors or the number of errors where the participant chose the same rule as the previous one. Total errors (all incorrect responses), perseverative respon-ses (number of wrong answers that would have been correct for the preceding category), perseverative errors (number of errors in which a participant continuously responds incorrectly using the same pattern), non-perseverative errors (all the remaining incorrect responses), and conceptual level responses (all consecutive correct reactions that occur in runs of three or more) are scored.
Multi-Source Interference Task (MSIT). The MSIT was created by Bush et al. (2003). It measures attention and cognitive processing. In the MSIT, participants are presented with sets of three numbers (0, 1, 2, or 3), in which one number will always be different from the other two numbers. Participants identify the number that is different from the other two by pressing a button. During control trials (congruent), the different numbers will always match its position on the button press (e.g., 100, 020, 003). During interference trials (incongruent), the different number never matches its position on the button press. Participants are required to quickly and accurately select a different number, regardless of their location. The computer calculates the reaction times, and the averages of reaction times during control and interference trials are computed.
Emotional Stroop Task (EST). The EST is used to assess selective attention and processing speed under emotional content exposure. It is similar to the standard Stroop task, but it includes negative emotional words (e.g., threat-related) and neutral words. In the EST, participants are asked to identify the color of neutral (e.g., chair, table, etc.) or threat-related (e.g., death, kill, etc.) words. The words are presented in a computer monitor for 1.5 seconds with a 0.5-second interval between words. Participants have 45 seconds to identify the color of the word that is presented by pressing a button. The average reaction times to neutral and threat-related words is computed.
Psychological Assessment
Structured Clinical Interview for DSM-IV Disorders (SCID-IV). The SCID-IV was designed by the American Psychiatric Association (APA), and the Spanish version was used for this study. The SCID-IV is divided into Axis I and Axis II disorders. We used the Axis I assessment to evaluate symptoms of anxiety disorders, medications, and history of substance use.
Beck Anxiety Inventory – Spanish Version (BAI). The BAI was created by Beck et al. (1988) and is commonly used to measure the severity of anxiety. It includes 21 items, rated from 0 (not at all) to 3 (severely, bothered me a lot), with scores indicating very low, mild, moderate, and severe anxiety symptoms. The English version has an internal consistency of α = .92 (Beck et al., 1988). A study that included a Spanish-speaking sample obtained an alpha coefficient of .93 (Magán et al., 2008).
Beck Depression Inventory-II – Spanish Version (BDI-II). The BDI-II was created by Beck (1996) to evaluate the severity of depression symptoms. The BDI-II has 21 items, ranging from 0 to 3, indicating very low to severe symptoms. It has an alpha coefficient of α = .91 for psychiatric outpatients (Beck et al., 1996). An adapted version of the BDI (BDI-S) obtained high internal consistency α = .88 in a study that included Puerto Rican undergraduate students (Bonilla et al., 2004).
State and Trait Anxiety Inventory – Spanish Version (STAI). The STAI was created by Spielberger et al. (1983) to assess state and trait anxiety. The STAI includes 20 items to measure state anxiety and 20 items to measure trait anxiety. Scores range from 1 (very slightly) to 4 (extremely). The internal consistency of the STAI ranges from .86 to .95 (Spielberger et al., 1983). The Spanish version of the STAI has been used in several studies (González-Barrios et al., 2016) and a study that included the assessment of Puerto Ricans yielded a high internal consistency in State α = .83 to .92 and Trait α = .86 to .92 (Spielberger & Díaz-Guerrero, 1975).
Positive and Negative Affect Schedule (PANAS). The PANAS was created by Watson and Clark (1988), and it evaluates mood state, positive and negative affect. It includes 10 items that describe negative emotional states and 10 items that describe positive emotional states. Scores are rated on a 5-point Likert scale, ranging from 1 (very little or have not felt it) to 5 (extremely). It has an alpha coefficient that ranges from .86 to .90 for the Positive Affect scale and .84 to .87 for the Negative Affect scale (Watson et al., 1988). A study that used a Spanish adaptation of the PANAS showed alpha coefficients of .89 (Positive Affective State) and .91 (Negative Affective State) for males and alpha coefficients of .87 (Positive Affective State) and .89 (Negative Affective State) for females (Sandín et al., 1999). This scale has not been validated in Puerto Rico (González-Barrios et al., 2016).
Emotional Dysregulation Scale (EDS). The Emotional Dysregulation Scale is composed of 40 items, and it evaluates emotional changes and perceptions of an individual’s capacity to manage emotions. Although it has been previously used, no studies have assessed the psychometric properties, and this scale has not been validated for the Puerto Rican population (González-Barrios et al., 2016).
Statistical Analyses
Descriptive statistics were used to analyze demographic data. Kruskal-Wallis tests were used to compare physiological data between OCD and healthy groups during fear conditioning and extinction. Non-parametric Friedman tests were performed to analyze the SCR of all participants between trials during phases of the fear conditioning and extinction paradigm. Post-hoc Wilcoxon signed-rank tests were used to assess differences in SCR across trials of each phase by group. Mann-Whitney U tests were used for group comparison in neuropsychological and psychological measures. Statistical significance was set at p < 0.05. All statistical analyses were performed using the SPSS software package, version 22.
RESULTS
Healthy controls and participants with OCD did not differ in age, sex, and years of education (Table 1). We found no differences in the SCR to the conditioned stimulus during the presence of the shock (CS+) between the OCD and healthy groups (Figure 1). During CS+, no significant differences between the OCD and healthy groups were observed on trials of the Habituation, Conditioning, Early Extinction, Late Extinction, Recall, and Renewal phases (p > 0.05). Likewise, we found no differences in the SCR to the conditioned stimulus during the absence of the shock (CS-) between OCD and healthy groups on trials of the Habituation, Conditioning, Early Extinction, Late Extinction, Recall, and Renewal, as shown by the Kruskal-Wallis test (p > 0.05).
Non-parametric Friedman tests were conducted to evaluate differences in medians among the SCR during the trials of each phase. The Friedman test showed significant differences in SCR between trials in the Conditioning phase during CS+ (χ2(4) = 11.23, p = 0.024) and during CS- (χ2(4) = 16.75, p = 0.002) when both groups were assessed. Post-hoc using a Wilcoxon signed-rank test showed that SCR of healthy subjects in the Conditioning phase during CS+ was significantly different (z = -2.58, p = 0.010) across the early and late trials, but this difference was not observed in the OCD group (z = -0.53, p = 0.594). However, SCR in the Conditioning phase during CS- were significantly different in the healthy group (z = -2.44, p = 0.015) and OCD group (z = -2.22, p = 0.026) during the first and last trials. In other words, both OCD and healthy groups learned to fear the shock during the Conditioning phase, as fear increased from early to late trials, although the increase in fear throughout the trials was significant only in the healthy group. SCR in both groups decreased significantly during the course of the trials of the Conditioning phase during CS-.
The Friedman test also showed significant differences in SCR between trials in the Recall phase during CS- (χ2(4) = 18.85, p = 0.001) when both groups were assessed. Follow-up Wilcoxon signed-rank test revealed a significant difference in SCR during CS- early and late trials of the Recall phase in the healthy group (z = -2.20, p = 0.028), but not in the OCD group (z = -1.16, p = 0.248). This suggests that both OCD and healthy groups, learned to extinguish fear during CS- in the Recall phase, when the memory for fear extinction is assessed, but SCR only decreased significantly across early and late trials in the healthy group.
Furthermore, significant differences in the SCR of both groups were observed during CS+ (χ2(4)19.66, p = 0.001) and during CS- (χ2(4)12.74, p = 0.013) in the Renewal phase, when the memory for fear conditioning is assessed. Follow-up Wilcoxon signed-rank test revealed significant differences in SCR across early and late trials during CS+ in the Renewal phase in the OCD group (z = -2.22, p = 0.026) and healthy group (z = -2.86, p = 0.004). Follow-up tests also revealed significant differences in SCR across early and late trials during CS- in the Renewal phase in the OCD group (z = -2.22, p = 0.026) and healthy group (z = -2.39, p = 0.017). These results suggest that both groups experienced a renewal of the fear response when re-exposed to the shock in a novel context. No other significant differences in SCR in OCD and healthy groups were observed throughout trials of other phases (p>0.05).
Mann-Whitney U tests showed no significant differences in percent fear extinction recall between groups when the conditioning stimulus was paired with a shock (CS+), z = - .21, p = 0.83, and when it was not paired with a shock (CS-), z = - 0.26, p = 0.80. Likewise, there were no significant differences between groups in percent fear renewal between groups during CS+, z = -0.40, p = 0.69 and during CS-, z = -0.97, p = 0.34.
However, subjects with OCD showed increased symptoms of general anxiety, compared to healthy controls, as measured by the administered psychological assessments measuring general anxiety, z = - 3.57, p = 0.00 (BAI), trait anxiety, z = - 3.63, p = 0.00 (STAI), depression, z = -2.92, p = 0.00 (BDI), negative affect, z = - 3.30, p = 0.00 (PANAS), and emotional dysregulation, z = - 2.80, p = 0.01 (EDS) (Table 2). However, no significant differences were observed between healthy and OCD groups in state anxiety, z = - 1.79, p = 0.73 (STAI) and positive affect, z = - 0.90, p = 0.37 (PANAS).
Mann-Whitney U tests were conducted to evaluate the hypothesis that subjects with OCD will have lower performance on neuropsychological tests compared to healthy controls (Table 3). The results on the WCST, which measures executive functions and set-shifting, were not significant, p>0.05, as shown in the assessment of total errors (z = - 0.33, p = 0.74), perseverative responses (z = - 0.15, p = 0.90), perseverative errors (z = - 0.13, p = 0.90), non-perseverative responses (z = - .43, p = 0.67), and conceptual level responses (z = - 0.08, p = 0.94). Furthermore, results on accuracy on congruent (z = -0.80, p = 0.42) and incongruent trials (z = -0.73, p = 0.46) and reaction times of congruent (z = -1.06, p = 0.29) and incongruent trials (z = -0.31, p = 0.76) on the MSIT, which evaluates attention and cognitive processing, showed no significant differences between groups, p>0.05. Additionally, differences were not significant between groups inaccuracy to neutral (z = -0.72, p = 0.47) and threat-related words (z = -1.32, p = 0.18) nor in reaction time to neutral (z = 0.00, p = 1.00) and threat-related words (z = -0.60, p = 0.54), as assessed by the performance on the EST; a test that evaluated the effect of emotional content on participants, as shown by Mann-Whitney U tests. Nevertheless, significant differences were observed between groups in differential reaction time (the difference between the reaction time to threat-related and neutral words; reaction time to threat-related words minus reaction time to neutral words). In other words, there was a significant difference in the effects of emotional interference between groups, z = - 2.16, p = 0.031 (Figure 2 and Figure 3). The effect size was moderate 0.41 (Cohen, 1988), suggesting that the OCD group showed higher reaction times to neutral words than to threat-related words, compared to healthy subjects who had similar reaction times to both threat-related and neutral words.
DISCUSSION
In the present study, we evaluated the manifestations of OCD in physiological, neuropsychological, and psychological measures using a Puerto Rican sample, compared to healthy controls. Interestingly, in our study, participants with OCD did not show impaired extinction recall, contrary to what we hypothesized and reported with a non-Latino White sample. In other words, there were no differences between OCD and healthy groups when the memory of fear extinction is tested. However, differences in SCR across trials during Conditioning and Renewal were observed in both groups, suggesting adequate acquisition and extinction of fear. The lack of differentiation in fear conditioning and extinction between OCD subjects and healthy controls is inconsistent with the results of a previous study by Milad et al. (2013), in which the same fear conditioning and extinction paradigm were used, and they found deficits on the recall of extinction in an OCD Caucasian sample. This result points to the possibility that Puerto Ricans with OCD can learn how to fear a specific stimulus and may be able to extinguish this fear as well as healthy individuals.
Nevertheless, this outcome would likely be different if the fear conditioning and extinction paradigm consisted of common themes relevant to OCD. For example, Armstrong and Olatunji (2017) used an associative learning task that showed a neutral face that was followed by a disgusting image (unconditioned stimulus) and a neutral face that was not paired with a disgusting image (CS-) and found that individuals with high and low contamination concerns did not differ in discriminant responding. However, those with high contamination concerns, who exhibited symptoms consistent with individuals with OCD, reported more considerable disgust to the CS+, and their unconditioned stimulus expectancy during the CS+ was resistant to extinction. It is also possible that OCD subjects in our study were receiving treatment, which was not evaluated, and were consequently efficient in acquiring and extinguishing fear memories. Additionally, it may be possible that the two samples differed in other characteristics not assessed, such as the severity of OCD symptoms or years living with the condition.
Although no differences were observed in physiological responses, we found marked differences in trait anxiety, depressive symptoms, negative affect, and emotional dysregulation in OCD participants, compared to the healthy group. These findings confirm that in addition to the SCID confirmed the presence of OCD, and the clinical sample showed marked pathological symptoms as compared to the healthy controls. These results are consistent with previous studies that highlight the difficulty of individuals with OCD in regulating their emotions, which may be related to their fear of expressing feelings, such as anxiety, and losing control (Allen & Barlow, 2009; Macatee et al., 2013). Significant differences were found on the EST in terms of the effects of emotional interference (differential reaction time) in the OCD group, compared to the healthy group. OCD subjects showed quicker responses towards threat-related words than the whole group, which had similar reaction times towards both threat-related and neutral words. This finding suggests that OCD subjects may perform better under exposure to threatening content, specifically to written information.
Previous research on attentional bias in OCD has generated conflicting findings. Foa and McNally (1986) conducted one of the first studies to assess attentional bias in OCD by using a dichotic listening task before and after exposure and response prevention treatment. Participants were presented with different fear-relevant and neutral prose passages to each ear. They were asked to shadow the passage given in the dominant ear as they were spoken on the tape while pressing a button when they heard the designated target word or phrase that was shown on a printed card, placed in front of the subject. Foa and McNally (1986) found that fear-relevant words were detected more than neutral words before but not after treatment, according to behavioral (button press) and physiological (skin conductance responses) measu-res, thus skin conductance responses to fear-relevant words were more significant than to neutral words before treatment. After exposure and response prevention treatment, OCD subjects did not detect fear-relevant and neutral words differentially.
Similarly, in our study, the EST included a verbal paradigm with threat-related words that were not associated with OCD-related concerns (e.g., washing, checking), and OCD subjects responded quicker to threat-related than neutral words. This leads to questioning if the attentional bias found in OCD subjects on previous research may be towards personally relevant threatening content (Amir et al., 2010). In a study by Rao et al. (2010), clinically symptomatic and remitted OCD patients were examined using the EST, including OCD relevant words, positively valenced (e.g., neat) and negatively valenced (e.g., dirty). They found that symptomatic OCD patients had significantly higher emotional interference compared to healthy subjects; thus, OCD subjects selectively attended to threatening stimuli that were associated with their OC concerns (i.e., took longer to process negatively valenced OCD relevant words). Their results also showed that the emotional interference was only present in symptomatic, but not in remitted OCD subjects, which suggests it may be a state rather than a trait marker. However, another study showed different results (Moritz et al., 2008). Moritz et al. (2008) used the EST with words associated with washing and checking OCD subtypes to assess if OCD patients share an attentional bias towards stimuli related to their disorder. Results in their study showed that OCD participants had no interference nor bias towards OCD subtype-congruent words. It may be possible that the worries of individuals with OCD may be mostly triggered by visual cues or images (e.g., dirt on the table). Thus Moritz et al. (2008) suggest that using a visual paradigm can be more suitable in assessing attentional bias in OCD.
The fact that there were no significant differences between groups on other neuropsychological tests suggests that the neuropsychological functioning of OCD subjects can be altered upon the presence of emotional stimuli. These results add relevant information to the controversial findings on the neuropsychological performance of individuals with OCD (Abramovitch & Cooperman, 2015; Kohli et al., 2015). Lack of differences between groups in fear recall and fear renewal suggests that OCD in Puerto Ricans may not be characterized by difficulty in maintaining fear acquisition and extinction memories. Significant differences in trait anxiety, negative affect, and emotional dysregulation between groups suggest that Puerto Ricans with OCD may have neuropsychological and physiological characteristics that are similar to healthy subjects but have a distinct psychological profile, which can be used to tailor cultural adaptations of evidence-based treatments for OCD. Considering cultural adaptations or changes in the process and content of psychotherapy according to cultural differences is essential to protect the scientific integrity of evidence-based research, promote valid studies, and reduce health disparities (Bernal & Domenech-Rodríguez, 2012) and not considering them may limit enhanced treatment response in minority populations, including Puerto Rico.
For this reason, according to the results of this study, treatment interventions in Puerto Ricans with OCD should be focused on the regulation of emotions, given the intense anxiety they experience. Although OCD is experienced globally, significant differences in the expression and severity of symptoms, primarily associated with psychological stressors, may exist and should be deliberated for effective treatment outcomes. This view is consistent with findings in a study by Chavira et al. (2008), showing similarities in the phenotypic expression of OCD of Costa Ricans and North Americans, such as the content of obsessions and compulsions, and symptom frequency and common subtypes, which may have a biological component. However, differences were observed in severity, suggesting the influence of culture. Furthermore, Chavira et al. (2008) found differences in trait anxiety between Costa Ricans and North Americans; North Americans had higher levels of trait anxiety than Costa Ricans. These results highlight possible disparities in the psychological functioning of different ethnic groups. According to Hwang (2006) (as cited in Zane et al., 2016), adapted interventions should focus on several areas that are relevant to culture, such as contemplating variation of the expression and management of emotional distress according to culture and addressing cultural issues and stressors that may be specific to ethnic minorities.
The most common and effective evidence-based treatment for OCD includes selective serotonin reuptake inhibitors (SSRIs) and cognitive behavioral therapy with exposure and response prevention (Abramowitz & Arch, 2014; Franz et al., 2013; Pauls et al., 2014). Limitations in the development and promotion of treatment for the minority population and the lack of professional cultural competence emphasizes the need for research that includes ethnic considerations (Williams et al., 2010; Zane et al., 2016). Our study's findings draw attention to the possibility that Puerto Ricans with OCD differ from healthy subjects, predominantly on subjective psychological measures. Although this study did not evaluate neurobiological structure or functioning in individuals with OCD, this should be considered in future studies regarding the manifestations of OCD in Puerto Ricans.
Limitations and Directions for Future Research
This study has several limitations. A small sample size may have limited the detection of OCD and healthy group differences in physiological responses and neuropsychological functioning and, thus, the generalization of results. However, differences were observed in psychological reports and emotionally-linked neuropsychological measures, despite a small sample size. Additionally, illness duration and if treatment was received was not assessed for participants with OCD. Changes in the severity of symptoms and illness duration complicate the adequate evaluation of neuropsychological performance. Also, if participants received or were currently receiving treatment could have an impact on the process of fear acquisition and extinction in OCD subjects. Future studies should consider these limitations to reach accurate conclusions. Future studies can include a larger sample and consider illness duration and if participants have received treatment in the past, as well as consider OCD subtypes and symptom severity, which may influence neuropsychological and psychological functioning.
CONCLUSION
In conclusion, Puerto Ricans with OCD reacts similarly to healthy individuals during the process of maintaining fear acquisition and extinction memories. Thus, they may be efficient in learning how to fear and extinguishing the fear memory when concerns are not OCD-related. Furthermore, emotional stimuli may be influencing the psychological functioning of Puerto Ricans with OCD and their neuropsychological performance when tests involve emotional content. These findings can provide a further understanding of OCD in Latinos and emphasize the importance of considering cultural differences, such as the impact of emotional content, in the treatment and assessment of this population to obtain effective outcomes.
Funding: The project described was supported by Award Number S21MD001830 (K. Martínez) from the National Center on Minority Health and Health Disparities. The content is solely the authors' responsibility and does not necessarily represent the official views of the funding institutions.
Acknowledgments: We would also like to thank Dr. Carlos Morales for helpful assistance with the statistical analyses.
Conflict of Interest: None of the authors has any financial conflicts of interest to disclose.