Visual search for verbal material in patients with obsessive–compulsive disorder

Psychiatry Research 264 (2018) 244–253

Contents lists available at ScienceDirect

Psychiatry Research journal homepage: www.elsevier.com/locate/psychres

Visual search for verbal material in patients with obsessive–compulsive disorder

T

⁎

Fabiano Bottab,1, , Nicolas Viberta,1, Ghina Harika-Germaneaub,1, Mickaël Frascaa, François Rigalleaua, Eric Fakrac, Christine Rosa, Jean-François Roueta, Florian Ferrerid, Nematollah Jaafarib,e a

Centre de Recherches sur la Cognition et l'Apprentissage, CNRS, Université de Poitiers, Université de Tours, Poitiers, France Unité de Recherche Clinique Pierre Deniker du Centre Hospitalier Henri Laborit, Poitiers, France c Service Hospitalo-Universitaire de Psychiatrie, Secteur Saint-Etienne, Hôpital Nord, Saint-Etienne, France d Département de Psychiatrie et de Psychologie Médicale, Hôpital Saint-Antoine, Université Pierre et Marie Curie Paris 6, Assistance Publique – Hôpitaux de Paris, Paris, France e Laboratoire de Neurosciences Expérimentales et Cliniques, INSERM U 1084, Université de Poitiers, INSERM CIC-P 1402; CHU de Poitiers, Poitiers, France b

A R T I C LE I N FO

A B S T R A C T

Keywords: Visual search Obsessive–compulsive disorder Eye-tracking Selective attention Working memory Word list

This study aimed at investigating attentional mechanisms in obsessive–compulsive disorder (OCD) by analysing how visual search processes are modulated by normal and obsession-related distracting information in OCD patients and whether these modulations differ from those observed in healthy people. OCD patients were asked to search for a target word within distractor words that could be orthographically similar to the target, semantically related to the target, semantically related to the most typical obsessions/compulsions observed in OCD patients, or unrelated to the target. Patients’ performance and eye movements were compared with those of individually matched healthy controls. In controls, the distractors that were visually similar to the target mostly captured attention. Conversely, patients’ attention was captured equally by all kinds of distractor words, whatever their similarity with the target, except obsession-related distractors that attracted patients’ attention less than the other distractors. OCD had a major impact on the mostly subliminal mechanisms that guide attention within the search display, but had much less impact on the distractor rejection processes that take place when a distractor is fixated. Hence, visual search in OCD is characterized by abnormal subliminal, but not supraliminal, processing of obsession-related information and by an impaired ability to inhibit task-irrelevant inputs.

1. Introduction Patients with Obsessive Compulsive Disorder (OCD) suffer from time consuming obsessive thinking, which provokes anxiety and distress, and interferes with everyday activities (Abramowitz et al., 2009). Usually, the obsessions result in associated compulsions, such as repetitive behaviours, rituals, or constant checking, by which patients reduce the anxiety generated by their obsessions (Menzies et al., 2008). There is broad agreement that OCD involves abnormalities of the fronto-striatal circuits (Pujol et al., 2004). Accordingly, OCD patients are impaired on the neurocognitive functions sub-served by these brain areas (Chamberlain et al., 2005), showing alterations in executive functions such as response inhibition and working memory (Jaafari et al., 2013; Snyder et al., 2015), in decision-making processes (Sachdev and Malhi, 2005) and in attentional control mechanisms ⁎

1

(Morein-Zamir et al., 2013). According to the literature review by Kuelz et al., (2004), there is little evidence for a specific impairment of basic attentional abilities in OCD patients. However, several studies suggest that OCD patients have trouble ignoring irrelevant stimuli and show impaired selective attention abilities. In recent reviews, Kashyap et al., (2013) and Snyder et al., (2015) argued that the attentional deficits observed in OCD patients were mainly a by-product of their impaired executive functions, pointing to reduced skills in the organization of incoming information and optimization of cognitive resources (Kashyap et al., 2013). According to Snyder et al., (2015), the depression and motor slowing which are often observed in OCD patients cannot explain the broad executive functions’ impairment typically observed in these patients. In accordance with this idea, Chamberlain et al., (2005) suggested that failures in cognitive and behavioural inhibitory processes underlie

Correspondence author. Department of Experimental Psychology, University of Granada, Spain. E-mail address: [email protected] (F. Botta). These three authors contributed equally to this work

https://doi.org/10.1016/j.psychres.2018.03.054 Received 15 May 2017; Received in revised form 26 February 2018; Accepted 22 March 2018 Available online 23 March 2018 0165-1781/ © 2018 Elsevier B.V. All rights reserved.

Psychiatry Research 264 (2018) 244–253

F. Botta et al.

fixations made on each type of distractor words was taken as an index of the words’ ability to attract attention, while fixation durations were taken as an index of the processing time needed to consciously reject the distractor words once they were fixated. The experiment was designed to compare the impact of target-distractor similarities on attentional guidance and word processing times in OCD patients versus healthy controls. In particular, the lists containing obsession-related words were compared with the lists containing neutral distractors, which represented the baseline condition, to assess whether obsession-related information is processed differently from other kinds of verbal input by patients. Consistent with Unoki et al., (1999), obsession-related words were expected to influence the unconscious attentional guidance process, but not the conscious rejection of non-target words. Many previous studies reported working memory impairment in OCD patients (Jaafari et al., 2013; Moritz et al., 2002; Nakao et al., 2009; van der Wee et al., 2003). According to the literature (see Jaafari et al., 2013), the patients’ working memory deficit would result from abnormalities in memory control and/or executive functioning. Since visual search tasks require the active maintenance of a representation of the target (the “target template” ) in working memory (see for instance Soto et al., 2005), the verbal and visuo-spatial components of the participants’ working memory were measured to establish possible relationships with participants’ performance in the search task. The verbal and visuo-spatial components of working memory were evaluated using tests known to involve both the storage and manipulation of information, namely the reading span test (Desmette et al., 1995) and the backward location span test (Fournier-Vicente et al., 2008). Consistent with previous literature, OCD patients were expected to show lower working memory scores than control participants. Hence, given the prominent involvement of working memory in visual search processes, OCD patients were expected to need more time and more fixations than controls to find the target in the visual search task. More precisely, there should be a reliable inverse correlation between working memory scores and visual search efficiency among OCD patients. The patients with higher working memory were expected to show less reduction of efficiency in the search task than those with lower working memory scores. Another reason to expect a reduction of the efficiency of visual search for words in OCD patients follows from the fact that, as stated above, OCD patients have trouble ignoring irrelevant stimuli and show impaired selective attention abilities, probably because of their impaired executive functions (Kashyap et al., 2013; Snyder et al., 2015). As a result, OCD patients should have more difficulties rejecting irrelevant distractor words while searching for the target word than control participants.

many of the neurocognitive symptoms found in OCD patients. More precisely, they proposed that two main types of inhibitory processes are impaired in OCD patients, namely the cognitive inhibition processes that control internal thinking and prevent intrusive thoughts and mental rituals, and the behavioural inhibition processes that control motor actions such as checking behaviours. Some theories suggest that abnormal processing of the information related to obsessions constitutes an essential element of OCD. More specifically, the hypothesis is that attentional processing would be biased towards obsession-related information, thus contributing to develop and sustain intrusive obsessive thinking (Rachman, 1997). For example, Lavy et al., (1994), using a Stroop color-naming task, observed that OCD patients were slower in naming the colour of OCDrelated words with negative valence than that of neutral words. Using a similar task, Unoki et al., (1999) observed that OCD patients were more sensitive to compulsion-related words than healthy controls, but only when the words were presented subliminally and could not be consciously identified. However, other studies failed to find any attentional bias for OCD-related information presented either supraliminally (Moritz et al., 2008) or subliminally (Kampman et al., 2002). An alternative way to study attentional bias in OCD is to study the patients’ eye movements during the visual search paradigm, which requires people to find a target object presented among other visual stimuli that are called “distractors”. When people are free to move their eyes, current eye movement models indicate that the location of the attentional focus usually coincides with the location of gaze fixations (Deubel and Schneider, 1996; Zelinsky, 2008). Hence, eye movement recordings are often used to study human attentional behaviour. More precisely, the assumption of a strong relationship between attention and gaze is supported by strong psychophysical evidence, which shows that the focus of attention moves towards the location of the next fixation just before the corresponding saccade occurs (Deubel and Schneider, 1996). In other words, the locations of successive saccades within the search display would reflect how the searcher's attention is guided within the display. According to Zelinsky's model (Zelinsky, 2008), visual search relies on the elaboration of a retinotopic target-map, which gives a point-bypoint measure of the similarity between the items present in the visual field and the target, and which is updated after each eye movement. The search process consists in an alternation of fixations and saccades until the target is found. Each saccade is the result of an attentional guidance mechanism, which identifies the most likely target candidate on the current target map. According to current visual search models (Soto et al., 2005; van Zoest and Donk, 2010; Wolfe et al., 2011), the observer is generally unaware of the particular features that attract her or his attention towards this particular item. Once the saccade is made and the most likely target candidate is fixated, the observer must decide whether the candidate is really the target, or instead a distractor item that must be rejected. In the latter case, programing of the next saccade begins right away. In contrast with what happens during saccade programing, visual search models assume that the observer has some awareness of the decision he makes about the currently fixated item (Dampuré et al., 2014). To our knowledge, the few studies that employed the visual search paradigm to investigate attentional mechanisms in OCD patients were only based on response times and accuracy (Kaplan et al., 2006; Morein-Zamir et al., 2013), but did not analyse patients’ eye movements. The main goal of the present study was thus to investigate online attentional processes in OCD by recording OCD patients’ eye movements during a visual search paradigm with verbal material. Participants had to search for target words within lists of distractor words (Léger et al., 2012). The distractor words were manipulated to be orthographically similar to the target, semantically related to the target, obsession-related words (i.e., words that were semantically related to the patients’ most frequent obsessions) or neutral, target- and obsession-unrelated words. According to Zelinsky (2008), the number of

2. Material and methods 2.1. Participants Thirty-six patients with a primary diagnosis of OCD, and 36 healthy control volunteers with no history of or current psychiatric or neurological illness participated in the experiment (Table 1). Patients were recruited at a specialized psychiatric hospital (Centre Hospitalier HenriLaborit, Poitiers). The controls were recruited from local community and were individually matched with patients for sex, age (to within 5 years) and years of education. All participants gave their informed consent to participate in the study, and the experimental protocol was approved by the local ethics committee.2 Patients were examined by a psychiatrist using the Mini 2 The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.

245

Psychiatry Research 264 (2018) 244–253

F. Botta et al.

experimental session included five successive blocks of five sets of sentences. Each block included one series of two sentences, one of three sentences, one of four sentences, one of five sentences and one of six sentences. According to Friedman & Miyake (2005) the total span score of each participant was calculated as the total number of correctly recalled words with a maximum possible value of 100.

Table 1 Mean values and standard deviations (SD) of the main participants’ characteristics and OCD patients’ clinical variables. Participants’ characteristics and clinical variables

Patients with OCD (n = 36)

Healthy controls (n = 36)

Gender (F/M) Age

22/14 37.3 (13.3; range 16–73) 17.1 (9.9; range 6–51) 20.2 (11.2; range 2–48)

22/14 37.2 (13.2; range 15–68) Not applicable Not applicable

11.4 (2.9; range 7–18) 26.3 (3.9; range 21–36) 6.7 (3.3; range 4–17) 27 out of 36 (75%) Current comorbidity

Not available Not applicable Not applicable None No history of axis I psychiatric disorder

Age at onset of pathology Duration of the pathology (in years) HARS score Y-BOCS score BABS score Pharmacological treatment Comorbidity

2.2.2. Backward location span task The backward location span task (Fournier-Vicente et al., 2008) consisted of a modified, computerized version of the Corsi blocks tapping task (Milner, 1971). Participants were presented with a five by five grid, in which increasingly long sequences of two to nine randomly located cells turned black one after the other for 1500 ms. Immediately after each sequence, participants had to reproduce it in the opposite order in an empty grid by clicking on the corresponding cells. The test started with a practice session of three two-cell and three three-cell sequences. The experimental session included 32 sequences of increasing difficulty, four for each set size, namely four two-cell sequences, four three-cell sequences and so on up until four nine- cell sequences. In accordance with recommendations for scoring span test performance (Friedman & Miyake, 2005), the participant's visuospatial span score was defined as the total number of correctly recalled cells for a maximum possible score of 176.

Social phobia (7 patients) Generalized anxiety disorder (3 patients) Pathological gambling (1 patient) Bipolarity (1 patient) Past comorbidity Major depressive episode (26 patients)

2.2.3. Visual search task The material consisted of 96 lists of French common nouns that were derived from the lists used by Léger et al., (2012). Each list included a target word, six distractor words and six “target-unrelated” words that were neither orthographically nor semantically related to the target displayed in a single column. All words in the experimental lists had 1–4 syllables, 4–10 letters (M = 7.4 letters, SD = 1.4) and a lexical frequency superior to 1 per million according to the “Lexique 2” database (New et al., 2004). The 96 lists included 24 sets of three “experimental” lists (i.e., 72 lists) that were used to manipulate the type of distractor words seen by the different participants, and 24 “filler” lists that were the same for all participants. Each set of “experimental” lists and each filler list was built around a different, OCD-unrelated target word. Each participant searched once for each of the 24 “experimental” target words in one of the three lists of the corresponding set, and once for each of the 24 “filler” target words in the 24 “filler” lists (see procedure below). Twelve of the 24 experimental sets were similar to those used by Léger et al., (2012) and included the following three types of lists (Table 2):

International Neuropsychiatric Interview (MINI version 5.0.0.). Provided that OCD was their dominant disorder, patients with comorbid psychiatric diagnoses were not excluded, except those with current mood episodes. Patients’ anxiety was assessed through a French version of the Hamilton Anxiety Rating Scale (HARS). The patients’ OCD symptoms (see Table 1) were assessed through a French version of the Yale-Brown Obsessive–Compulsive Scale (Y-BOCS, Goodman et al., 1989b, a), which measures the severity of OCD symptoms on a 0 to 40 scale. Thirty of the patients displayed checking compulsions. The patients’ insight on their pathology was assessed by a French version of the Brown Assessment of Beliefs Scale (BABS, Eisen et al., 1998). The BABS score ranges from 0 (excellent insight) to 24 (no insight). Twentyseven of the patients were receiving antidepressants with serotonin reuptake-inhibiting properties alone, or combined with neuroleptics (n = 5), anxiolytics (n = 1), or both (n = 1). 2.2. Procedure and apparatus Participants were tested individually and performed the reading span test, the backward location span task and the visual search task in that order. The reading span test and the search task were performed on the 17” monitor of a TOBII 1750 eyetracker, which provided gaze positions at 50 Hz with a precision of 0.5° of visual angle. The backward location span task was performed on a PC-compatible laptop using homemade executable software. The same versions of the reading span test and backward location span task as in Jaafari et al., (2013, 2015) were used.

1. Orthographic lists included the target word, six target-unrelated words and six “orthographic” distractors that shared at least their first and last two letters with the target word. 2. Semantic lists included the target word, the same six target-unrelated words as orthographic lists and six “semantic” distractors that were semantic associates of the target word. 3. Neutral lists included the target word, the same six target-unrelated words as the two other types of lists and six other “neutral” distractor words that were also unrelated to the target word.

2.2.1. Reading span test A French version (Desmette et al., 1995) of the original test (Daneman and Carpenter, 1980) was used. The material consists of 100 unconnected test sentences and 10 practice sentences. Each sentence contains from 12 to 17 words. All the last words of the sentences are nouns matched for mean length and frequency. Participants were required to read aloud series of two to six sentences and to remember the final noun of each sentence of a series. The sentences were presented one by one on the screen. After having read all sentences in a series, the participant had to recall the last word of each sentence following the order of presentation. The practice session included a series of two sentences and a series of three sentences. The

The 12 other experimental sets used by Léger et al., (2012) were modified to create obsession-related material and included the following three types of lists (Table 2): 1. Orthographic lists included the target word, six target-unrelated words and six “orthographic” distractors that shared at least their first and last two letters with the target word. 2. New, obsession-related lists were built to replace the semantic lists. They included the target word, the same six target-unrelated words as in the orthographic lists and six “obsession-related” distractor words. To cover as much as possible the whole range of obsessions 246

Psychiatry Research 264 (2018) 244–253

F. Botta et al.

Table 2 Examples of experimental word lists used for the experiment. Orthographic list

First set of lists

Target Distractor words

Filler words

Semantic list

French words

Translation

French words

Translation

French words

Translation

corbeau cerceau caveau ciseau cadeau carreau chameau absent moyenne poinçon faïence soda roulette

raven hoop small cellar scissors present window pane camel absentee average punch, stamp earthenware soda roulette, caster

corbeau plumage perchoir envol branchage nichée forêt absent moyenne poinçon faïence soda roulette

raven plumage Perch taking flight branches nest, brood forest absentee average punch, stamp earthenware soda roulette, caster

corbeau froment répression novice meeting paupière tribut absent moyenne poinçon faïence soda roulette

raven wheat repression beginner meeting eyelid tribute absentee average punch, stamp earthenware soda roulette, caster

Orthographic list

Second set of lists

Target Distractor words

Filler words

Neutral list

Obsession-related list

Neutral list

French words

Translation

French words

Translation

French words

Translation

sculpteur sauteur sauveteur sécateur secteur senateur senteur crampon ondée eglantine préjudice raidillon canard

sculptor jumper lifeguard shears sector senator scent stud downpour eglantine prejudice steep path duck

sculpteur tourment hygiène symétrie addition brossage pervers crampon ondée eglantine préjudice raidillon canard

sculptor torment hygiene symmetry addition brushing pervert stud downpour eglantine prejudice steep path duck

sculpteur tournant hôtesse sucrerie adhésion braderie potages crampon ondée eglantine préjudice raidillon canard

sculptor bend hostess sweet adhesion clearance sale soups stud downpour eglantine prejudice steep path duck

experimental trials were used (see also Léger et al., 2012, who used the same procedure). All participants were tested one by one as in Léger et al., (2012). Each trial began with the presentation of a target word at the top-left corner of the screen (Fig. 1). Then, the word disappeared and participants had to fixate a cross on the left side of the screen. After 2 seconds, the list in which they had to search for the target appeared. Once they had found the target, the participants had to click on it as fast and as accurately as possible. The response time was the time elapsed between the appearance of the list and the mouse click on the target word. For each participant, as mentioned above, the experimental session consisted of 24 experimental and 24 filler trials presented after 10 practice trials. As the filler trials, the practice trials consisted in searching for a target word within 12 target-unrelated distractor words and were the same for all participants. Practice trials did not include any word contained in either the experimental or filler trials, and in addition all words used within the experimental and filler trials were seen only once by each participant. As already stated above, the 24 experimental trials consisted for each participant in searching once for each of the 24 experimental target words in one of the three lists of the corresponding set. The experimental session was divided in two parts consisting each of 12 experimental trials and 12 filler trials presented in random order. In the first part, the participant searched for the 12 target words associated with orthographic, semantic and neutral lists. In the second part, the participant searched for the 12 target words associated with orthographic, obsession-related and neutral lists. Within both parts of the session, the assignment of experimental target words between the three types of lists was counterbalanced so that the type of list was crossed with three sets of target words (four per set) and twelve groups of three participants. Altogether, each participant searched for eight target words within orthographic lists, eight target words within neutral lists, four target words within semantic lists and four target words within obsession-

and compulsions that can be experienced by OCD patients, each one of these words was related to one of the most typical types of obsessions/compulsions described by OCD patients (Rasmussen and Eisen, 1990). The six types of obsession/compulsions that were used included fear of contamination and illness, washing and fear of dirt, pathologic doubt and checking, counting, order and symmetry, and sexual or aggressive obsessions. The words were selected by two of the clinicians involved in this study (GHG and NJ), who routinely see OCD patients, to be easy to understand and as related as possible to these different types of obsessions/compulsions. The analysis of the obsessions and compulsions displayed by the 36 OCD patients tested in this study indicated that 27 of them (75%) suffered from 3 or more of the six types of obsessions/compulsions targeted by the six obsession-related words included in each list, and that all of them suffered from at least one of these obsessions or compulsions. 3. Neutral lists included the target word, the same six target-unrelated words as in the two other types of lists and six other “neutral” distractor words that were unrelated to the target word. In the results section below, only the data from experimental lists were analysed. The rationale for including filler trials, which were not analysed, within the experimental procedure was that previous studies revealed that displaying the words in columns led most participants to search for target words from top to bottom (Léger et al., 2012; Ojanpää et al., 2002). Hence, to maximize the impact of distractor words, all the target words of experimental lists were located randomly within the bottom half of the lists (positions 7 to 13), and the non-target words were distributed randomly across the remaining positions. In the filler trials, the target word was placed within the top half of the lists (positions 1 to 6). For each participant, the 24 filler trials were then intermingled with the 24 experimental trials. The filler trials were inserted to prevent participants from only searching for target words at the bottom of the lists, as they would quickly do if only the 24 247

Psychiatry Research 264 (2018) 244–253

F. Botta et al.

Fig. 1. Sequence of events in an experimental trial. In this example, the target is presented within an orthographic list.

score. The absence of significant correlation with the HARS score might be due to the overall low level of anxiety displayed by OCD patients. As already shown by others (Jaafari et al., 2013; Mataix-Cols et al., 2002), neither the patients’ working memory span scores nor their performance on the visual search task depended on their medication status. All patients were therefore pooled together for analyses.

related lists. Each patient and his or her individually matched control performed the same experimental trials. 2.3. Design and data analysis Dedicated TOBII software (ClearView 2.7.1) was used to analyse eye movements, to time slide presentations, and to time and locate mouse clicks. Eye fixations were defined, assigned to particular words and analysed as in Léger et al., (2012). The dependent variables were the number and average duration of the fixations made on the six distractor words and the six filler words of each list before the participant clicked on the target word. All trials in which the participant did not click on the target word were considered as errors and excluded from further analysis. All statistical analyses were done with linear mixed models that take into account both the independent variables that are manipulated (fixed factors) and the random effects caused by inter-participant and interitem variability (Kliegl et al., 2010). Separate models were constructed using the lmer function of the “lme4” R statistical package, version 1.1–7 (Bates et al., 2014; R CoreTeam, 2013), for each one of the three dependent variables used to describe the visual search behaviour, i.e. the number of fixations on non-target words, the average duration of these fixations, and the response times. All models included participants and lists as random intercepts to control for by-participant and by-list variability. The fixed factors were Pathological Status (patients or controls), List Type (neutral, orthographic, semantic or obsession-related) and Word Type (filler or distractor). Finally, regression models were built to assess whether the reading span and backward location span scores could predict the patients’ and controls’ visual search behaviour, as well as the level of each patient's reduction of efficiency in the visual search task relative to control participants.

3.1. Number of fixations on non-target words The model accounting for the number of fixations on non-target words (Fig. 2a) revealed a main effect of Pathological Status (χ2(1) = 29.1, p < .001), with patients making more fixations than controls, and a main effect of Word Type (χ2(1) = 13.2, p = .002), with a higher number of fixations on distractor words than on filler words. The main effect of List Type was also significant (χ2(3) = 16.1, p = .001) but was qualified by an interaction with Pathological Status (χ2(3) = 11.1, p = .01).3 Patients made less fixations on obsession-related lists than on neutral (p < .001), orthographic (p = .003) and semantic lists (p = .01), whereas there was no significant difference between the number of fixations they made on neutral, orthographic and semantic lists (all ps > 0.075). Healthy controls, instead, made more fixations on orthographic lists than on neutral (p = .02), semantic (p = .04) and obsession-related lists (p = .01), whereas there was no significant difference between the number of fixations they made on neutral, semantic and obsession-related lists (all ps > 0.48). In addition, patients made more fixations than controls on neutral (p < .001) and 3 The absence of a significant three-way interaction implies that the impact of the nature of distractors was visible on both the filler and distractor words rather than just on distractor words. Due to the vertical arrangement of the words within the lists, it is actually plausible that more than one word could be identified in a single fixation on the list (Ojanpää et al., 2002), thus blurring the differentiation between fillers and distractors. Since the main goal was to understand participants’ behaviour on distractor words, specific t-test comparisons were also performed by taking into account only the distractor words and excluding the filler words, which were the same in all lists. The general pattern of results was confirmed. Patients made less fixations on obsession-related distractors than on neutral and orthographic ones (p < .001 in both cases), but the difference between the number of fixations made on obsession-related and semantic distractors was not significant (p = .08). Healthy controls, instead, made more fixations on orthographic distractors than on neutral (p = .01), semantic (p = .05) and obsession-related ones (p = .008).

3. Results Both OCD patients and controls made very few errors (0.5%). Among patients, none of the dependent variables that characterized the visual search behaviour was significantly correlated with the Y-BOCS score, the insight score, the duration of the pathology or the HARS 248

Psychiatry Research 264 (2018) 244–253

F. Botta et al.

semantic lists (p = .003), but not on orthographic (p = .15) and obsession-related lists (p = .54). 3.2. Mean duration of fixations on non-target words The model accounting for the mean duration of fixations on nontarget words (Fig. 2b) showed a main effect of Word Type (χ2(1) = 24.1, p < .001), with longer fixations on distractor words than on filler words. There was also a main effect of List Type (χ2(3) = 19.7, p < .001), with longer fixations on orthographic lists than on all other kinds of list (all ps < 0.01). The main effect of Word Type was actually qualified by a List Type x Word Type interaction (χ2(3) = 18.6, p < .001). Participants’ fixations were longer on distractor words than on filler words in orthographic (p < .001) and semantic lists (p = .03), but not in neutral and obsession-related lists (both ps > 0.56). The main effect of Pathological Status was not significant (χ2(1) = 2.81, p = .09). 3.3. Response times The linear mixed model run on the log-transformed response times (Fig. 2c) showed a main effect of Pathological Status (χ2(1) = 165.1, p < .001) indicating that patients needed more time to find the target than controls. There was a main effect of List Type (χ2(3) = 18.9, p < .001). Participants were faster to find the target within obsessionrelated lists than within all other kinds of lists (all ps < 0.05). To check whether the pattern of data observed for the number of fixations was reflected in participants’ response times, planned comparisons were performed despite the absence of significant interaction between List Type and Pathological Status. The pattern of data was similar to that observed with the number of fixations. Patients were faster to find the target within obsession-related lists than within neutral (p = .007) and orthographic (p = .002) lists, whereas the difference between obsession-related and semantic lists did not reach significance (p = .07). Conversely, controls showed faster response times for the targets presented in neutral, semantic and obsession-related lists than for those presented in orthographic lists (p = .01, p = .01 and p = .003, respectively). 3.4. Impact of working memory spans Working memory span was higher in healthy controls than in patients for both the verbal and visuo-spatial working memory tasks (respectively t(35) = 4.4, p < .001 and t(35) = 4.4, p < .001). The regression models displayed on Fig. 3 (see panels 3a and 3b) revealed a significant interaction between the participants’ pathological status and the impact of both working memory spans on participants’ response times, such that the regression slopes were significantly different between OCD patients and controls (spatial working memory F (1,68) = 8.14, p = .005; verbal working memory F(1,68) = 12.75, p < .001). Among OCD patients, there were significant negative correlations between both the verbal and spatial working memory spans and response times (verbal β = −0.52, p = .001, Fig. 3a; visuo-spatial β = −0.63, p < .001; Fig. 3b), whereas among controls response times were not significantly correlated with either one of the working memory spans. Similarly, among OCD patients, when either one of the working memory spans increased, the number of fixations on non-target words (verbal working memory β = −0.37, p = .026, visuo-spatial working memory β = −0.43, p = .009), and the mean duration of the fixations (verbal β = −0.52, p = .001, visuo-spatial β = −0.35, p = .039) decreased significantly. Among controls in contrast, there was no significant correlation between the working memory spans and either the number or mean duration of fixations on non-target words. To check whether the patients’ reduction of efficiency in the search task was related to their working memory, regression models were built to assess whether the patients’ reading span and backward location span scores could predict the increase in the number of fixations on

Fig. 2. (a) Mean number of participants’ fixations as a function of pathological status and list type. (b) Mean duration of participants’ fixations as a function of pathological status, list type and word type. (c) Mean participants’ response times as a function of pathological status and list type. Error bars represent standard errors of the means. O = Orthographic; S = Semantic; N = Neutral; OR = Obsession-related. * for p < .05, ** for p < .01 and *** for p < .001.

249

Psychiatry Research 264 (2018) 244–253

F. Botta et al.

Fig. 3. (a) Correlation between verbal working memory scores and response times in healthy controls and OCD patients. (b) Correlation between spatial working memory scores and response times in healthy controls and OCD patients. (c) Correlation between the patients’ increase in response times relative to their individually matched controls and the patients’ verbal working memory score.

response times β = −0.49, p = .002) with the patients’, but not the controls’ visuo-spatial working memory scores.

non-target words and response times observed in each patient relative to her or his individually-matched control. The differences in the number of fixations (β = −0.36, p = .033) and response times (β = −0.59, p < .001) between patients and their matched controls were both negatively related to the patients’, but not the controls’ verbal working memory scores (Fig. 3c). Similarly, these differences were both negatively related (numbers of fixations β = −0.31, p = .07,

4. Discussion As expected, OCD patients needed more fixations on non-target words and more time than healthy participants to perform visual search 250

Psychiatry Research 264 (2018) 244–253

F. Botta et al.

fixations is the same on both types of distractors. Accordingly, the number of fixations on non-target words decreased in the obsessionrelated lists where obsession-related words were present, whereas the average duration of these fixations was not modified in comparison with the other kinds of lists. Greene and Rayner (2001) precise that the familiarity effect results from low-level learning processes associated to repeated exposure to the stimuli, which are independent of the meaning of the stimuli. In other words, familiar distractor words are recognized very quickly through their perceptual features (i.e., their visual appearance as strings of letters) without even being semantically processed. In OCD patients, the long-term, repeated exposure to obsessionrelated information would make the visual appearance of obsessionrelated words extremely familiar. Consequently, OCD patients would thus be able to recognize and reject the obsession-related words just based on their visual appearance, without directly fixating them. However, another interpretation is possible. The facilitating effect of obsession-related words could be explained also in terms of avoidance. In this case, the reduced number of fixations on obsession-related words would result from OCD patients unconsciously avoiding information related to their obsessions. Some of the previous studies that used the Stroop task support this idea. According to Kyrios and Iob (1998), for instance, the avoidance of OCD-related words would result from “automatic” decisions taken without real awareness at early stages of information processing, which would prevent the stimuli to reach conscious processing levels. However, using contamination-related pictures, Najmi et al., (2010) showed that OCD patients were slower at approaching these pictures, but did not reject them faster. Altogether, because of the differences between the precise experimental tasks that were used in these studies and the difference between these tasks and our own paradigm (see also below), the way OCD patients process obsession or compulsion-related stimuli is still controversial. Establishing a general model of the processing of OCD-related stimuli by OCD patients would require a better control of the participants’ individual levels of familiarity with both the OCD-related material and the neutral material used as control. In contrast with the numbers of fixations, the average duration of fixations was not significantly different between OCD patients and controls. Since the duration of fixations represents the decision process by which the fixated words are identified as target or distractors (Dampuré et al., 2014; Krajbich et al., 2010; Zelinsky, 2008), this suggests that the patients’ conscious decision processes involved in the rejection of distractor words were not impaired. This is consistent with the data reported by Unoki et al., (1999) who observed that obsessionrelated information affected subliminal, but not supraliminal processing in OCD patients. As expected, the present research indicates that OCD patients do process obsession-related information in a specific way (Lavy et al., 1994; McNally et al., 1994; Unoki et al., 1999). However, the differences between the Stroop color-naming task often used to study attentional bias in OCD and the visual search paradigm make direct comparisons between the different studies difficult. One of the most important differences relates to the exact attentional mechanisms that are assessed respectively by the Stroop task and the visual search task. In the Stroop paradigm, the OCD-related information is the to-be-ignored dimension of a unique stimulus. Participants must inhibit an obsession-related information while accessing a target information (the colour of the writing) with both pieces of information contained in the same item, i.e., the word they must fixate. Thus, in the Stroop task, the target and distracting information are both within the attentional focus. In the visual search paradigm in contrast, the threatening information never coincides with the location of target stimulus. Moreover, since the target is placed within an array of distractor words, attention guidance processes become essential for the task. As far as we know, Morein-Zamir et al., (2013) were the only ones to use a visual search task including obsession-related material to analyse OCD patients’ attentional bias. They failed to observe reliable

for words. Interestingly, the various types of distractor words did not have the same impact on patients’ and controls’ visual search behaviour, suggesting that attentional guidance mechanisms differ between OCD patients and healthy controls. As previously observed (Dampuré et al., 2014; Léger et al., 2012), healthy participants made more fixations on orthographic distractors than on other types of words. Their attention was preferentially attracted by the words that were visually similar to the target. Orthographic distractors were more likely than neutral distractors to be selected as potential targets and fixated because of their visual resemblance with the target (Zelinsky, 2008). In contrast, OCD patients less frequently fixated obsession-related distractors than the neutral, orthographic and semantic ones, but fixated the neutral distractors as often as orthographic and semantic distractors. The similarity of patients’ oculomotor behaviour within neutral, semantic and orthographic lists shows that patients’ attention was attracted in the same way by items that did not share any feature with the target (i.e., the neutral words) and by items that were visually similar to the target. These results suggest that while in healthy participants attentional guidance was based as usual on target-distractor visual similarity (Zelinsky, 2008), this was not the case for OCD patients. The patients’ oculomotor target maps may be less diversified than those of controls and may consist of samples of equally salient objects (hence equally liable to be fixated) without any strong differentiation between relevant and irrelevant items. This is consistent with previous claims that OCD patients’ attention tends to be captured by local, irrelevant features of stimuli (Soref et al., 2008). These findings are in accordance with Chamberlain et al., (2005) model, which states that OCD patients’ cognitive deficits result mainly from failures in inhibitory control processes. More precisely, the present results suggest that OCD patients have trouble optimizing the distribution of their attentional resources because they do not properly inhibit task-irrelevant stimuli. In other words, OCD patients would be unable to attend selectively the items that are relevant for the task they must perform. The resulting capture of patients’ attention by any kind of distractor may explain why OCD patients were altogether slower and less efficient than controls in the search task, and might be directly related to the patients’ overload of working memory. Consistently, in OCD patients only, both the verbal and spatial working memory spans correlated with visual search efficiency. Patients with lower working memory spans needed more time to find the target because they made more and longer fixations on non-target words. Moreover, each patient's working memory span predicted the increase in response times and number of fixations observed in the visual search task relative to his or her matched control. This indicates that the anomalies of patients’ visual search processes were closely related to the alteration of their working memory, as already reported by Jaafari et al., (2013). The apparent incapacity of OCD patients to focus on relevant information (Yovel et al., 2005) might overload the working memory with task-irrelevant inputs thus generating the working memory deficits. In that case, consistent with Harkin and Kessler's model (Harkin and Kessler, 2011), OCD patients’ working memory impairment would be secondary to executive dysfunction. Another intriguing result is that while patients made fewer fixations and had faster response times on obsession-related lists than on all other kinds of lists, healthy participants showed instead similar results for neutral, semantic and obsession-related lists. The presence of obsession-related material was somehow “beneficial” for patients, which suggests that they processed obsession-related material differently from controls. A first possible interpretation is that patients make fewer fixations on obsession-related lists just because the obsession-related words are very familiar to them as part of their obsessions. According to Greene and Rayner (2001), familiar distractors are less fixated than unfamiliar ones during visual search, even though the duration of 251

Psychiatry Research 264 (2018) 244–253

F. Botta et al.

References

differences between OCD patients and controls, but they only measured the time that participants needed to discriminate whether the target (an obsession related item set among positive distractors or a positive item set among obsession-related distractors) was present or not, and did not record participants’ eye movements during the task. In the present study, the use of eye-tracking measures during visual search made it possible to evaluate much more precisely the various attentional processes involved in this task as detailed in the introduction section, which revealed specific alterations of the attentional guidance mechanisms in OCD patients. In future studies, it might be interesting to combine eye-tracking with EEG measures and systematically manipulate the distance between the obsession-related information and the target stimulus, to analyse more precisely the possible dissociation between attentional guidance and the inhibitory mechanisms involved in the rejection of distracting material by OCD patients. A better coherence with previous studies is obtained when taking in consideration the role of inhibition in the search task. In the present paradigm, during the search for the target word, the top-down mechanisms involved in the active maintenance of the target representation in working memory interact with the bottom-up capture of attention by the distractor words. The generally larger amount of fixations to distractor words observed in OCD patients, except on obsession-related words, suggests that in OCD patients the distractor words produce more interference than in healthy participants. Considering that distractor words represent the to-be inhibited information, these data point to a dysfunction in OCD patients of the processes that allow to actively maintain task-relevant information while inhibiting task-irrelevant information, in accordance with Chamberlain et al., (2005) model. As the first of its kind, this study has several limitations that must be considered for future research. First, the absence of specific measures of participants’ attentional performance and processing speed makes it difficult to know to which extent differences in basic cognitive abilities might have contributed to the present findings. Second, the use of vertical wordlists made it difficult to precisely attribute each fixation to a specific word (Ojanpää et al., 2002). Future eye-tracking studies should increase the distance between items and randomize their locations to reduce the possibility to foveate more than one item at once and minimize the impact of item layout on ocular guidance (Dampuré et al., 2014; Léger et al., 2012). Moreover, future experiments would benefit from the use of idiographic stimuli adapted to each patient's obsessions and compulsions (Morein-Zamir et al., 2013; Muller and Roberts, 2005) and/or from the division of the patients’ sample in different sub-samples depending on the specific OCD subtype they suffer from (e.g., washers vs. checkers, see Mataix-Cols et al., 2004). Finally, since intrusive, distracting thoughts are observed also in other psychiatric pathologies (i.e. depression, anxiety, eating disorders etc.), other kinds of psychiatric patients could be tested using the same paradigm in future studies to assess whether the present data are OCD specific or not. Despite these limitations, the present findings constitute a promising line of research to improve our understanding of the disorders of attentional control in OCD patients.

Abramowitz, J.S., Taylor, S., McKay, D., 2009. Obsessive–compulsive disorder. Lancet Lond. Engl. 374, 491–499. http://dx.doi.org/10.1016/S0140-6736(09)60240-3. Bates, D., Mächler, M., Bolker, B., Walker, S., 2014. Fitting linear mixed-effects models using lme4. ArXiv:14065823, https://scholar.google.es/scholar?q=fitting+linear +mixed-effects+models&hl=en&as_sdt=0&as_vis=1&oi=scholart&sa=X&ved= 0ahUKEwi24IzV46XaAhWJ0xQKHWPxCJ0QgQMIJTAA. Chamberlain, S.R., Blackwell, A.D., Fineberg, N.A., Robbins, T.W., Sahakian, B.J., 2005. The neuropsychology of obsessive compulsive disorder: the importance of failures in cognitive and behavioural inhibition as candidate endophenotypic markers. Neurosci. Biobehav. Rev. 29, 399–419. http://dx.doi.org/10.1016/j.neubiorev.2004. 11.006. Dampuré, J., Ros, C., Rouet, J.-F., Vibert, N., 2014. Task-dependent sensitisation of perceptual and semantic processing during visual search for words. J. Cogn. Psychol. 26, 530–549. Daneman, M., Carpenter, P.A., 1980. Individual differences in working memory and reading. J. Verbal Learn. Verbal Behav. 19, 450–466. http://dx.doi.org/10.1016/ S0022-5371(80)90312-6. Desmette, D., Hupet, M., Schelstraete, M.-A., Van der Linden, M., 1995. Adaptation en langue française du «Reading Span Test» de Daneman et Carpenter (1980). Année Psychol. 95, 459–482. Deubel, H., Schneider, W.X., 1996. Saccade target selection and object recognition: evidence for a common attentional mechanism. Vis. Res. 36, 1827–1837. Eisen, J.L., Phillips, K.A., Baer, L., Beer, D.A., Atala, K.D., Rasmussen, S.A., 1998. The brown assessment of beliefs scale: reliability and validity. Am. J. Psychiatry 155, 102–108. http://dx.doi.org/10.1176/ajp.155.1.102. Fournier-Vicente, S., Larigauderie, P., Gaonac'h, D., 2008. More dissociations and interactions within central executive functioning: a comprehensive latent-variable analysis. Acta Psychol. (Amst.) 129, 32–48. Friedman, N.P., Miyake, A., 2005. Comparison of four scoring methods for the reading span test. Behav. Res. Methods 37, 581–590. Goodman, W.K., Price, L.H., Rasmussen, S.A., Mazure, C., Delgado, P., Heninger, G.R., Charney, D.S., 1989a. The Yale–Brown obsessive compulsive scale. II. Validity. Arch. Gen. Psychiatry 46, 1012–1016. Goodman, W.K., Price, L.H., Rasmussen, S.A., Mazure, C., Fleischmann, R.L., Hill, C.L., Heninger, G.R., Charney, D.S., 1989b. The Yale–Brown obsessive compulsive scale. I. Development, use, and reliability. Arch. Gen. Psychiatry 46, 1006–1011. Greene, H.H., Rayner, K., 2001. Eye movements and familiarity effects in visual search. Vis. Res. 41, 3763–3773. Harkin, B., Kessler, K., 2011. The role of working memory in compulsive checking and OCD: a systematic classification of 58 experimental findings. Clin. Psychol. Rev. 31, 1004–1021. Jaafari, N., Chopin, N., Levy, C., Rotgé, J.-Y., Lafay, N., Hammi, W., Rigalleau, F., Millet, B., Krebs, M.-O., Vibert, N., Group, InsightStudy, 2015. Excessive checking behavior during an image comparison task in schizophrenia. Eur. Psychiatry J. Assoc. Eur. Psychiatr. 30, 233–241. http://dx.doi.org/10.1016/j.eurpsy.2014.11.012. Jaafari, N., Frasca, M., Rigalleau, F., Rachid, F., Gil, R., Olié, J.-P., Guehl, D., Burbaud, P., Aouizerate, B., Rotgé, J.-Y., Vibert, N., for Insight Study Group, 2013. Forgetting what you have checked: a link between working memory impairment and checking behaviors in obsessive–compulsive disorder. Eur. Psychiatry J. Assoc. Eur. Psychiatr. 28, 87–93. http://dx.doi.org/10.1016/j.eurpsy.2011.07.001. Kampman, M., Keijsers, G.P.J., Verbraak, M.J.P.M., Näring, G., Hoogduin, C.A.L., 2002. The emotional stroop: a comparison of panic disorder patients, obsessive–compulsive patients, and normal controls, in two experiments. J. Anxiety Disord. 16, 425–441. Kaplan, O., Dar, R., Rosenthal, L., Hermesh, H., Fux, M., Lubow, R.E., 2006. Obsessivecompulsive disorder patients display enhanced latent inhibition on a visual search task. Behav. Res. Ther. 44, 1137–1145. Kashyap, H., Kumar, J.K., Kandavel, T., Reddy, Y.C.J., 2013. Neuropsychological functioning in obsessive–compulsive disorder: are executive functions the key deficit? Compr. Psychiatry 54, 533–540. http://dx.doi.org/10.1016/j.comppsych.2012.12. 003. Kliegl, R., Wei, P., Dambacher, M., Yan, M., Zhou, X., 2010. Experimental effects and individual differences in linear mixed models: estimating the relationship between spatial, object, and attraction effects in visual attention. Front. Psychol. 1, 238. http://dx.doi.org/10.3389/fpsyg.2010.00238. Krajbich, I., Armel, C., Rangel, A., 2010. Visual fixations and the computation and comparison of value in simple choice. Nat. Neurosci. 13, 1292–1298. http://dx.doi. org/10.1038/nn.2635. Kuelz, A.K., Hohagen, F., Voderholzer, U., 2004. Neuropsychological performance in obsessive–compulsive disorder: a critical review. Biol. Psychol. 65, 185–236. http:// dx.doi.org/10.1016/j.biopsycho.2003.07.007. Kyrios, M., Iob, M.A., 1998. Automatic and strategic processing in obsessive–compulsive disorder: attentional bias, cognitive avoidance or more complex phenomena? J. Anxiety Disord 12, 271–292. Lavy, E., van Oppen, P., van den Hout, M., 1994. Selective processing of emotional information in obsessive compulsive disorder. Behav. Res. Ther. 32, 243–246. Léger, L., Rouet, J.-F., Ros, C., Vibert, N., 2012. Orthographic versus semantic matching in visual search for words within lists. Can. J. Exp. Psychol. Can. Psychol. Exp. 66, 32. Mataix-Cols, D., Alonso, P., Pifarré, J., Menchón, J.M., Vallejo, J., 2002. Neuropsychological performance in medicated vs. unmedicated patients with obsessive–compulsive disorder. Psychiatry Res. 109, 255–264. Mataix-Cols, D., Wooderson, S., Lawrence, N., Brammer, M.J., Speckens, A., Phillips, M.L., 2004. Distinct neural correlates of washing, checking, and hoarding symptomdimensions in obsessive–compulsive disorder. Arch. Gen. Psychiatry 61,

Financial support This research received no specific grant from any funding agency, commercial or not-for-profit sectors. Conflict of interest None Acknowledgments We thank Ms. Rebecca Lane for checking the English language of the manuscript before submission. 252

Psychiatry Research 264 (2018) 244–253

F. Botta et al.

Gen. Psychiatry 61, 720–730. http://dx.doi.org/10.1001/archpsyc.61.7.720. CoreTeam, R, 2013. R: A Language and Environment For Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. Rachman, S., 1997. A cognitive theory of obsessions. Behav. Res. Ther. 35, 793–802. Rasmussen, S.A., Eisen, J.L., 1990. Epidemiology of obsessive compulsive disorder. J. Clin. Psychiatry 51, 10–13 Suppldiscussion 14. Sachdev, P.S., Malhi, G.S., 2005. Obsessive–compulsive behaviour: a disorder of decisionmaking. Aust. N. Z. J. Psychiatry 39, 757–763. http://dx.doi.org/10.1080/j.14401614.2005.01680.x. Snyder, H.R., Kaiser, R.H., Warren, S.L., Heller, W., 2015. Obsessive–compulsive disorder is associated with broad impairments in executive function: A meta-analysis. Clin. Psychol. Sci. J. Assoc. Psychol. Sci. 3, 301–330. http://dx.doi.org/10.1177/ 2167702614534210. Soref, A., Dar, R., Argov, G., Meiran, N., 2008. Obsessive–compulsive tendencies are associated with a focused information processing strategy. Behav. Res. Ther. 46, 1295–1299. http://dx.doi.org/10.1016/j.brat.2008.09.001. Soto, D., Heinke, D., Humphreys, G.W., Blanco, M.J., 2005. Early, involuntary top-down guidance of attention from working memory. J. Exp. Psychol. Hum. Percept. Perform. 31, 248–261. http://dx.doi.org/10.1037/0096-1523.31.2.248. Unoki, K., Kasuga, T., Matsushima, E., Ohta, K., 1999. Attentional processing of emotional information in obsessive–compulsive disorder. Psychiatry Clin. Neurosci. 53, 635–642. http://dx.doi.org/10.1046/j.1440-1819.1999.00618.x. van der Wee, N.J.A., Ramsey, N.F., Jansma, J.M., Denys, D.A., van Megen, H.J.G.M., Westenberg, H.M.G., Kahn, R.S., 2003. Spatial working memory deficits in obsessive compulsive disorder are associated with excessive engagement of the medial frontal cortex. NeuroImage 20, 2271–2280. http://dx.doi.org/10.1016/j.neuroimage.2003. 05.001. van Zoest, W., Donk, M., 2010. Awareness of the saccade goal in oculomotor selection: your eyes go before you know. Conscious. Cogn. 19, 861–871. http://dx.doi.org/10. 1016/j.concog.2010.04.001. Wolfe, J.M., Võ, M.L.-H., Evans, K.K., Greene, M.R., 2011. Visual search in scenes involves selective and nonselective pathways. Trends Cogn. Sci. 15, 77–84. http://dx.doi.org/ 10.1016/j.tics.2010.12.001. Yovel, I., Revelle, W., Mineka, S., 2005. Who sees trees before forest? The obsessive–compulsive style of visual attention. Psychol. Sci. 16, 123–129. http://dx.doi. org/10.1111/j.0956-7976.2005.00792.x. Zelinsky, G.J., 2008. A theory of eye movements during target acquisition. Psychol. Rev. 115, 787–835. http://dx.doi.org/10.1037/a0013118.

564–576. McNally, R.J., Amir, N., Louro, C.E., Lukach, B.M., Riemann, B.C., Calamari, J.E., 1994. Cognitive processing of idiographic emotional information in panic disorder. Behav. Res. Ther. 32, 119–122. Menzies, L., Chamberlain, S.R., Laird, A.R., Thelen, S.M., Sahakian, B.J., Bullmore, E.T., 2008. Integrating evidence from neuroimaging and neuropsychological studies of obsessive–compulsive disorder: the orbitofronto-striatal model revisited. Neurosci. Biobehav. Rev. 32, 525–549. http://dx.doi.org/10.1016/j.neubiorev.2007.09.005. Milner, B., 1971. Interhemispheric differences in the localization of psychological processes in man. Br. Med. Bull. 27, 272–277. Morein-Zamir, S., Papmeyer, M., Durieux, A., Fineberg, N.A., Sahakian, B.J., Robbins, T.W., 2013. Investigation of attentional bias in obsessive compulsive disorder with and without depression in visual search. PloS One 8, e80118. http://dx.doi.org/10. 1371/journal.pone.0080118. Moritz, S., Birkner, C., Kloss, M., Jahn, H., Hand, I., Haasen, C., Krausz, M., 2002. Executive functioning in obsessive–compulsive disorder, unipolar depression, and schizophrenia. Arch. Clin. Neuropsychol. 17, 477–483. http://dx.doi.org/10.1016/ S0887-6177(01)00130-5. Moritz, S., Fischer, B.-K., Hottenrott, B., Kellner, M., Fricke, S., Randjbar, S., Jelinek, L., 2008. Words may not be enough! No increased emotional Stroop effect in obsessive–compulsive disorder. Behav. Res. Ther. 46, 1101–1104. http://dx.doi.org/10. 1016/j.brat.2008.05.005. Muller, J., Roberts, J.E., 2005. Memory and attention in obsessive–compulsive disorder: a review. J. Anxiety Disord. 19, 1–28. http://dx.doi.org/10.1016/j.janxdis.2003.12. 001. Najmi, S., Kuckertz, J.M., Amir, N., 2010. Automatic avoidance tendencies in individuals with contamination-related obsessive–compulsive symptoms. Behav. Res. Ther. 48, 1058–1062. http://dx.doi.org/10.1016/j.brat.2010.06.007. Nakao, T., Nakagawa, A., Nakatani, E., Nabeyama, M., Sanematsu, H., Yoshiura, T., Togao, O., Tomita, M., Masuda, Y., Yoshioka, K., Kuroki, T., Kanba, S., 2009. Working memory dysfunction in obsessive–compulsive disorder: a neuropsychological and functional MRI study. J. Psychiatr. Res. 43, 784–791. http://dx.doi.org/10.1016/j. jpsychires.2008.10.013. New, B., Pallier, C., Brysbaert, M., Ferrand, L., 2004. Lexique 2: a new French lexical database. Behav. Res. Methods Instrum. Comput. 36, 516–524. Ojanpää, H., Näsänen, R., Kojo, I., 2002. Eye movements in the visual search of word lists. Vis. Res 42, 1499–1512. Pujol, J., Soriano-Mas, C., Alonso, P., Cardoner, N., Menchón, J.M., Deus, J., Vallejo, J., 2004. Mapping structural brain alterations in obsessive–compulsive disorder. Arch.

253